World Cup February 11, 2011

World Cup 2011: an alternate preview

A statistical analysis of team strengths going into the 2011 World Cup
131

India: statistically the favourites
India: statistically the favourites © AFP

The tenth edition of the World Cup is not far away. In my last article I looked at the previous 9 World Cups from an alternate analysis point of view. In this article I will look at the ensuing World Cup, similarly from an analytical point of view.

For this article I have used some of the results of the proprietary work on recent form and match simulation I have done for a company which has a web presence exclusively for Cricket. As such I will not be presenting any detailed tables but refer to the conclusions extracted from those. I request readers to kindly bear with me. There are contractual restrictions to be observed.

The cornerstone of this analysis is the Team Strength index (TSI) of the participating teams. This index has been extracted using a complex process which involves the following.

- An estimated Final XI.
- The career figures of selected players.
- The recent form of players, bowling and batting.

The base TSI is determined using the following data.

- Career Runs per innings, after excluding single-digit not outs.
- Career Strike rate.
- Career Bowling strike rate- Balls per wicket.
- Career bowling accuracy - Runs per over.
- Recent Form Runs per innings, after excluding single-digit not outs.
- Recent Form Strike rate.
- Recent Form Bowling strike rate- Balls per wicket.
- Recent Form bowling accuracy - Runs per over.

The teams have been selected by me based on the assumption that ALL the players would be available and would be injury-free. Key players such as Tendulkar, Sehwag, Gambhir, Ponting, Collingwood, Bresnan, Kallis, Vettori and a few others have injury concerns. If these players do not play, the numbers would change as also the predictions. Mike Hussey and Eoin Morgan are already out. The teams concerned have already gone down in strength.

The recent form encompasses the last 10 innings played or last 10 spells bowled, provided these have been played on or after 1 January 2009. Most players have done this during 2010 itself. However someone like Tendulkar has played 2 innings in 2011 and 2 in 2010 and I had to go back one more year. 2009 performances are weighted slightly lower. Runs per Innings makes more sense than Batting average especially since recent form has to be considered. 3 not outs out of 10 distorts the recent form a lot. The weight for recent form numbers as against career numbers is decided based on the number of career matches played. The career figures have the highest weight of 75% if the player has played more than 100 matches. The recent form numbers become more significant if the player has played fewer matches.

After the base TSI is determined it is further adjusted based on the following three factors.

- The recent form of the teams - results, margins, venues et al.
- The venue of the matches (to decide on home/familiarity measures).
- The team performances in the two recent World Cups (2003 and 2007). This will enable us to assign due weight to the manner in which big-matches and big-stage have been handled. Going back beyond 2003 is not right since most of those players would have retired.

Recent form of teams: The recent form of teams considers the last 20 matches played by the team, provided these have been played on or after 1 January 2009. Most teams have done this during 2010 itself. The results are analyzed from results (wins/ties/losses) point of view and the match venue (home/neutral/away) point of view. In addition the margins of wins is incorporated. Finally, unlike the batsmen/bowler recent form calculations, the team numbers are determined based on the recent matches only. Needless to say, the West Indian successes of the 1970s should have no bearing on the chances of today's West Indies. The form related changes work out to around 3% on either side of 100.

Match venue factor: India, Bangladesh and Sri Lanka have been assigned a 2.5% benefit value for playing at home. Pakistan has a 1.25% benefit for playing in familiar sub-continent conditions.

Recent World Cup form: Upto 2% is allotted for this factor. A simple one based on the performances of teams in the recent two World Cups. Australia gets 2% (2 wins), Sri Lanka gets 0.75% (One final and one semi-final), India gets 0.50% (one final), New Zealand gets 0.35% (one semi-final and one super-six) and so on.

These numbers might seem arbitrary. However these have been arrived at after lot of trial runs. Moreover the benefit cannot be made greater than these since these numbers are used for simulation which is very sensitive to these numbers.

The recent form tables are listed below. It should be noted that the actual match team strength of India in its inaugural match against Bangladesh will be slightly lower since they would be playing "away" and Bangladesh would be playing at "home". Similarly actual match team strength of Sri Lanka in its match against New Zealand will be very slightly lower since both of them would be playing at "neutral" venue.

Team         BtIdx BwIdx TmIdx LocAdj RF_Adj Wc_Adj BtIdx BwIdx TmIdx

India 31.25 28.16 59.41 1.0250 1.0220 1.0050 32.89 29.65 62.54 South Africa 29.20 30.81 60.01 1.0000 1.0290 1.0025 30.12 31.78 61.91 Sri Lanka 24.30 33.79 58.09 1.0250 1.0300 1.0075 25.85 35.94 61.79 Australia 27.90 30.01 57.90 1.0000 1.0120 1.0200 28.80 30.97 59.77 Pakistan 23.98 27.16 51.14 1.0125 0.9990 1.0000 24.26 27.48 51.73 England 23.55 27.50 51.05 1.0000 1.0050 1.0010 23.69 27.67 51.35 Bangladesh 20.88 26.04 46.92 1.0250 0.9970 1.0010 21.36 26.64 48.00 West Indies 22.61 24.43 47.04 1.0000 0.9820 1.0010 22.23 24.01 46.24 New Zealand 20.70 24.81 45.50 1.0000 0.9670 1.0035 20.08 24.07 44.16 Ireland 17.60 22.86 40.47 1.0000 1.0160 1.0010 17.90 23.25 41.16 Zimbabwe 14.85 23.80 38.65 1.0000 0.9880 1.0010 14.69 23.54 38.23 Netherlands 12.02 17.50 29.52 1.0000 1.0020 1.0000 12.04 17.53 29.58 Canada 13.73 15.91 29.64 1.0000 0.9930 1.0000 13.64 15.80 29.43 Kenya 13.75 15.61 29.35 1.0000 0.9740 1.0025 13.42 15.24 28.66

It should be noted that the table would look different if the World Cup was going to be played outside, say South Africa. The pace bowlers would get back their potency. But the difference would probably be no more than 5%.


Not so surprising that India leads the team strength table, albeit by a hairsbreadth, closely followed by Sri Lanka, South Africa and Australia. India has the best batting lineup amongst all, and good bowling strength. Sri Lanka has the best bowling credentials, not matched by the batting. South Africa is placed high in both areas. Australia is also similarly placed. Their top quality pace bowling makes up for their average spin bowling. England, weakened by the loss of a key player and their indifferent form and Pakistan, with the loss of two key bowlers and their inability to play at home, are in the middle. Bangladesh, New Zealand and West Indies are in the third group. Ireland is very good, but is probably out of its league. Zimbabwe has an excellent spin attack but their batting is pathetic. Netherlands has one truly world class player. Canada, with an almost wholly expatriate team, are here to fill up the numbers.

The recent form of South Africa, Australia and Sri Lanka has been excellent. India and Pakistan have kept their heads above water. England's form was the best until recently. Unfortunately the Australian disaster reversed this. The other way around for Australia. The recent form of West Indies and New Zealand has been awful.

Now to the groups.

Team         BtIdx BwIdx TmIdx LocAdj RF_Adj Wc_Adj BtIdx BwIdx TmIdx

Group A

Sri Lanka 24.30 33.79 58.09 1.0250 1.0300 1.0075 25.85 35.94 61.79 Australia 27.90 30.01 57.90 1.0000 1.0120 1.0200 28.80 30.97 59.77 Pakistan 23.98 27.16 51.14 1.0125 0.9990 1.0000 24.26 27.48 51.73 New Zealand 20.70 24.81 45.50 1.0000 0.9670 1.0035 20.08 24.07 44.16 Zimbabwe 14.85 23.80 38.65 1.0000 0.9880 1.0010 14.69 23.54 38.23 Canada 13.73 15.91 29.64 1.0000 0.9930 1.0000 13.64 15.80 29.43 Kenya 13.75 15.61 29.35 1.0000 0.9740 1.0025 13.42 15.24 28.66

Group B

India 31.25 28.16 59.41 1.0250 1.0220 1.0050 32.89 29.65 62.54 South Africa 29.20 30.81 60.01 1.0000 1.0290 1.0025 30.12 31.78 61.91 England 23.55 27.50 51.05 1.0000 1.0050 1.0010 23.69 27.67 51.35 Bangladesh 20.88 26.04 46.92 1.0250 0.9970 1.0010 21.36 26.64 48.00 West Indies 22.61 24.43 47.04 1.0000 0.9820 1.0010 22.23 24.01 46.24 Ireland 17.60 22.86 40.47 1.0000 1.0160 1.0010 17.90 23.25 41.16 Netherlands 12.02 17.50 29.52 1.0000 1.0020 1.0000 12.04 17.53 29.58

Group A has an average Team strength value of 44.82, considerably lower than the other group. However this is the clearer group in that the top four teams are going to encounter very little opposition from Zimbabwe, Canada and Kenya are likely to share 3 wins amongst themselves. It would be a momentous upset if they defeat any of the top four teams. The team strength numbers substantiate this conclusion. These three teams are more than 15% away from the lowest placed of the top four teams. The order is quite difficult to predict. Let us say Sri Lanka, Australia, Pakistan and New Zealand qualify, in some order or other.


Group B is quite difficult to predict and is fraught with possibilities. The average team strength is 48.68, around 7-8% higher than the other group. However the dark horse is Bangladesh. They have a team strength which is understandably and justifiably higher than West Indies. They are the only team which would be playing all their matches at home. India and Sri Lanka play one match away. So there is a very good chance (I would put it as high as 50%) of Bangladesh winning three matches. That should put them in with a great chance of qualifying. So there is a good chance that one of the top four teams would miss out. That seems likely to be West Indies. Also note that Ireland are also quite strong. So let us say India, South Africa, England and Bangladesh, the order of the top three uncertain.

What happens afterwards is almost a lottery. The team which has three great days would win the cup, that is all. The two teams which have very little chance of having three great days are Bangladesh and New Zealand and should be ruled out. Out of the other six, England and Pakistan are likely to have two great days, but probably not three. So this leaves us with the four teams, India, South Africa, Sri Lanka and Australia. It could be any one of these four.

I can hear some readers saying that this is what everyone and their neighbour's cat is saying. However my statement is based on the fact that I have done complete simulation of the World Cup a few thousands of times. But as I have already mentioned, that is proprietary information for my client. Hence I am not able to divulge that information, until it is published. I can only say that these four teams are closely bunched together and, as a group, have around 80-85% chance of winning the World Cup. Pakistan and England follow next. The readers can draw their own inferences. The final simulation results show a very high degree of correlation with the Team strength values.

To view/down-load the list of selected players for each team, please click/right-click here.

Finally a request to the readers.

This is a World Cup played between 14 teams and 210 players. About half of these players would have fond expectations of winning and reaping the rich rewards. They would like to win the World Cup for themselves and their countries. It is quite unlikely that they do it for some other player in their team, however great the player might be. So let us stay out of this "win for Jayawardene" or "win for Tendulkar" or "win for Kallis" etc. Let your comments be centred around the teams not individuals. In this analysis, as in real life, the 11 players bring with them their performance-related numbers and contribute to the Team Strength. Then it is the team which performs together.

Anantha Narayanan has written for ESPNcricinfo and CastrolCricket and worked with a number of companies on their cricket performance ratings-related systems

Comments have now been closed for this article

  • Animesh on March 7, 2011, 14:58 GMT

    Sincere apologies; would like to check it out but i seem to be unable to locate it. Could you direct me as to when that was written? [[ http://blogs.espncricinfo.com/itfigures/archives/2010/01/_sachin_tendulkar_on_top.php Ananth: ]]

  • Animesh on March 5, 2011, 16:55 GMT

    Hats off. The unpredictability of sports i don't need to talk about. But an analysis-cum-preview like this is way too engrossing as a fan to set aside and move on.. just spent over an hour on the article and comments. Thanks man. Keep 'em coming.

    Incidentally, as a die-hard Sachin fan.. it is disappointing to see him consistently drop down the lists on batting parameters whenever you do your analyses. I could go on I guess, but perhaps in some other article. [[ It only shows that you hve not read through the articles. In the combined Test-ODI article Tendulkar was no.1 and that was the last pure batsman analysis. Ananth: ]]

    Once again - brilliant work. All simulations, experts and statistics aside - hoping India clobber Australia silly in the final :)

  • Obelix the Fat on February 28, 2011, 1:23 GMT

    I propose a group stage that qualifies you for a best of 3 final.

  • Boll on February 20, 2011, 9:49 GMT

    Little did I realise that this site would also introduce me to Tamil cinema. Just been looking at a bit of Kamal Hasan(?) that people were mentioning earlier. Fantastic stuff! Is he watching the game though?

  • love goel on February 19, 2011, 18:58 GMT

    If we REALLY want the strongest team to win, and not have any one match cause major ripple effects, the format actually has to be simpler...

    Well the definition of the strongest/best is not same for everybody. While you believe that group format followed by best of 3 is the best format, I truly believe that SF/F is the best do way to go in any sports. While you want the strongest teams to win,sports must provide for upsets and heartbreaks. It must allow for sudden change in momemtum, for the dark horses to steal the limelight

    For me SF/F are the way. Without that it will all be too predictable. [[ I see merits in both arguments. We have to allow for upset results since otherwise the charm is lost. At the same time we have to probably ensure that a bad day at the office of one player should not sinlk them. Baggio in 1994 did not have a bad day, but a bad few seconds, and Italy lost the Final. So the biggest of all World level tournaments, the Football WC, has similar flaws. Probably the all-play-all League, followed by semi finals and a 3-match Final is the best. If a team loses in the SF, tough luck. After all life should not be. helping the best team win. If India loses in the knock-out rounds of 2011 WC, despite every help provided to help them win, give credit to the team that derails them, against odds. Ananth: ]]

  • Alex on February 19, 2011, 6:24 GMT

    Ananth: besides multiple pen-names, the necessity of claiming a different nationality is baffling. Some readers claim to be Aussies but I will bet that they have never been to that beautiful country. Anyway, 3 more hours and WC '11 is ON. Your analysis is fine but I will add the following bits.

    1. The toss will prove crucial.

    2. Much like in '87 & '96, the skipper will play a big role. Border and Ranatunga were absolutely great as captains and as leaders. Media praises Sanga but Mahela was a far better leader. Afridi seems to have the mojo but I think India has the edge with MSD.

    3. This is the first world cup since T20's got popular and will hold some surprises thanks to T20 ... very tough to predict just how. All things being equal, I think the team with better T20 players will win the match. So, given the subcontinental conditions, I think Pak is the dark horse even without Aamer and Asif.

  • Kartik (the old one) on February 19, 2011, 6:06 GMT

    To give some idea of how far the West Indies have fallen..

    In the mid-80s, WI vs. SL matches were expected to be one-sided. Performances by Walsh, Richards, etc. were eye-popping. The first 10 matches between the two were won 10-0 by WI.

    Bangladesh did not even have full ODI status in the mid 80s.

    Today, not only is SL stronger than WI, but even Bangladesh is stronger than WI.

    Younger cricket fans may have trouble believing that WI were once the best.

  • Kartik on February 19, 2011, 5:58 GMT

    Ananth said : 'The team which has three great days would win the cup, that is all. '

    Yes. The one thing that really bugs me about this world cup (and the last one and the one before that) is : The format.

    If we REALLY want the strongest team to win, and not have any one match cause major ripple effects, the format actually has to be simpler.

    10 teams (only 1 minnow, no more).

    All teams play each other once (as in 1992).

    The top two teams then play a final (no semis, and certainly no ridiculous quarterfinals).

    The finals are a best-of-three. If the old WSC in Australia had a best-of-three final, why can't the actual WORLD Cup?

    This will ensure that the two best teams are the two finalists, and that the winner is pretty much confirmed as the very best team.

    No strong team gets knocked out by just on bad match (as in 2007). Groups and Quarterfins are both bad ideas.

    This truly is the best format for determining the Champion. Beauty in simplicity.

  • shanto551 on February 19, 2011, 1:37 GMT

    Can I watch the cricket live through cricinfo.com? [[ I am afraid you should contact cricinfo.com or go through their website. I am not sure whether free streaming of matches in India is being done. Ananth: ]]

  • Obelix the Fat on February 18, 2011, 18:09 GMT

    *These romans must be crazy* Simulate all you want Ananth here is what I predict :) Semis Australia vs Pakistan and South Africa vs England I have no idea what happens after ! [[ Your educated guess is as good as my scientific calculations. Once the QF is reached, it is almost anybody's Cup. Why Obelix ??? You could as well have been Astrix the Wise !!! All the best. Ananth: ]]

  • Animesh on March 7, 2011, 14:58 GMT

    Sincere apologies; would like to check it out but i seem to be unable to locate it. Could you direct me as to when that was written? [[ http://blogs.espncricinfo.com/itfigures/archives/2010/01/_sachin_tendulkar_on_top.php Ananth: ]]

  • Animesh on March 5, 2011, 16:55 GMT

    Hats off. The unpredictability of sports i don't need to talk about. But an analysis-cum-preview like this is way too engrossing as a fan to set aside and move on.. just spent over an hour on the article and comments. Thanks man. Keep 'em coming.

    Incidentally, as a die-hard Sachin fan.. it is disappointing to see him consistently drop down the lists on batting parameters whenever you do your analyses. I could go on I guess, but perhaps in some other article. [[ It only shows that you hve not read through the articles. In the combined Test-ODI article Tendulkar was no.1 and that was the last pure batsman analysis. Ananth: ]]

    Once again - brilliant work. All simulations, experts and statistics aside - hoping India clobber Australia silly in the final :)

  • Obelix the Fat on February 28, 2011, 1:23 GMT

    I propose a group stage that qualifies you for a best of 3 final.

  • Boll on February 20, 2011, 9:49 GMT

    Little did I realise that this site would also introduce me to Tamil cinema. Just been looking at a bit of Kamal Hasan(?) that people were mentioning earlier. Fantastic stuff! Is he watching the game though?

  • love goel on February 19, 2011, 18:58 GMT

    If we REALLY want the strongest team to win, and not have any one match cause major ripple effects, the format actually has to be simpler...

    Well the definition of the strongest/best is not same for everybody. While you believe that group format followed by best of 3 is the best format, I truly believe that SF/F is the best do way to go in any sports. While you want the strongest teams to win,sports must provide for upsets and heartbreaks. It must allow for sudden change in momemtum, for the dark horses to steal the limelight

    For me SF/F are the way. Without that it will all be too predictable. [[ I see merits in both arguments. We have to allow for upset results since otherwise the charm is lost. At the same time we have to probably ensure that a bad day at the office of one player should not sinlk them. Baggio in 1994 did not have a bad day, but a bad few seconds, and Italy lost the Final. So the biggest of all World level tournaments, the Football WC, has similar flaws. Probably the all-play-all League, followed by semi finals and a 3-match Final is the best. If a team loses in the SF, tough luck. After all life should not be. helping the best team win. If India loses in the knock-out rounds of 2011 WC, despite every help provided to help them win, give credit to the team that derails them, against odds. Ananth: ]]

  • Alex on February 19, 2011, 6:24 GMT

    Ananth: besides multiple pen-names, the necessity of claiming a different nationality is baffling. Some readers claim to be Aussies but I will bet that they have never been to that beautiful country. Anyway, 3 more hours and WC '11 is ON. Your analysis is fine but I will add the following bits.

    1. The toss will prove crucial.

    2. Much like in '87 & '96, the skipper will play a big role. Border and Ranatunga were absolutely great as captains and as leaders. Media praises Sanga but Mahela was a far better leader. Afridi seems to have the mojo but I think India has the edge with MSD.

    3. This is the first world cup since T20's got popular and will hold some surprises thanks to T20 ... very tough to predict just how. All things being equal, I think the team with better T20 players will win the match. So, given the subcontinental conditions, I think Pak is the dark horse even without Aamer and Asif.

  • Kartik (the old one) on February 19, 2011, 6:06 GMT

    To give some idea of how far the West Indies have fallen..

    In the mid-80s, WI vs. SL matches were expected to be one-sided. Performances by Walsh, Richards, etc. were eye-popping. The first 10 matches between the two were won 10-0 by WI.

    Bangladesh did not even have full ODI status in the mid 80s.

    Today, not only is SL stronger than WI, but even Bangladesh is stronger than WI.

    Younger cricket fans may have trouble believing that WI were once the best.

  • Kartik on February 19, 2011, 5:58 GMT

    Ananth said : 'The team which has three great days would win the cup, that is all. '

    Yes. The one thing that really bugs me about this world cup (and the last one and the one before that) is : The format.

    If we REALLY want the strongest team to win, and not have any one match cause major ripple effects, the format actually has to be simpler.

    10 teams (only 1 minnow, no more).

    All teams play each other once (as in 1992).

    The top two teams then play a final (no semis, and certainly no ridiculous quarterfinals).

    The finals are a best-of-three. If the old WSC in Australia had a best-of-three final, why can't the actual WORLD Cup?

    This will ensure that the two best teams are the two finalists, and that the winner is pretty much confirmed as the very best team.

    No strong team gets knocked out by just on bad match (as in 2007). Groups and Quarterfins are both bad ideas.

    This truly is the best format for determining the Champion. Beauty in simplicity.

  • shanto551 on February 19, 2011, 1:37 GMT

    Can I watch the cricket live through cricinfo.com? [[ I am afraid you should contact cricinfo.com or go through their website. I am not sure whether free streaming of matches in India is being done. Ananth: ]]

  • Obelix the Fat on February 18, 2011, 18:09 GMT

    *These romans must be crazy* Simulate all you want Ananth here is what I predict :) Semis Australia vs Pakistan and South Africa vs England I have no idea what happens after ! [[ Your educated guess is as good as my scientific calculations. Once the QF is reached, it is almost anybody's Cup. Why Obelix ??? You could as well have been Astrix the Wise !!! All the best. Ananth: ]]

  • Gerry_the_Merry on February 18, 2011, 10:36 GMT

    Hi Ananth

    This is just a message to you adn not a comment. i am somewhat drunk with all the batting analysis. My favorite topic in cricket is fast bowling. i understand that sometime back there was a detailed test bowling analysis in this blog by one of you but cannot trace it. would you be able to send me the link. many thanks. Gerry [[ In December I had done an analysis of Test bowling based on the aggregate of wickets in 1-10 tests and higher. You probably have seen that. I had also done one on Test third innings bowling and Test fourth innings bowling. I am not sure whether you have seen those. And the BCG charts for Test bowlers was third quarter 2010. One on Muralitharan. Another on Test bowling sequences. One on Test-Odi-Best bowler. I have just come to Feb 2010. I feel you should go back to the Cricinfo archives. Unfortunately it is quite confusing to go through the archives, You have to go month by month. But that is not my domain !!!

    Ananth: ]]

  • Spanish_Armada on February 17, 2011, 21:25 GMT

    Hola, a thought provoking article indeed.

    Simulation and modeling is only as good as the input variables. In cricket there are many variables to account for unlike lets say soccer or even tennis. That is why simulation in cricket has been pretty useless so for-I mean other similar projects in cricket have been a failure in predicting or estimating the future, after all, thats precisely why simulation is done so one can plan, and simulation in itself is not a goal . You have tried to validate your model with past results which is fairly common and standard practice but in cricket complicating factor is change of rules like power play, aging of players etc. As someone has mentioned, but may be in a very negative tone, that if you publish results then you have responsibility to disclose more. Otherwise keep private analysis private. I hope you get my point. [[ I have tried to ignore this despite knowing about the duplicity. When Alex warned of possible multiple postings I did not want to respond, despite my knowledge. I thought this would go away. But it seems to continue. In the past 10 days, you have posted under the following names. Spanish_Armada Yours Truly Rohit Singh Jat Baldev Singh. Your theme is the same. A seemingly mild attack on the article. Why this subterfuge. What pleasure do you derive. If you want to say something, say it openly under your own name. I have never hidden genuine comments, critical those may be. How do I know that these are all from the same person. These six messages have come from the same machine, tracked through the IP Address. The mail ids are all, of course, invalid, expected. I even know the location of the IP address. I will not reveal the IP address nor the location. That is not me. Suffice to say that we are talking about the south of USA. The answer to your comment: I have answered comments made in your other avatars and other genuine readers already. So there is nothing more to be said. Thanks Ananth: ]]

  • Rangarajan R on February 17, 2011, 10:32 GMT

    Does this tool also simulate scenarios? I am sure there would be some randomization involved in the simulation but still? How many times in competitive tournaments have the particpant teams won 3 consecutive matches (I am sure many teams have a yo - yo trend) . . . . If you look at the past 4 worldcup, the x - factor has reduced and the most consistent team (with the best balance for any condition) won. (Favourites always won).

    SL of 96 had players who could win on all conditions (it so happened that it was in Sub continent), SO were the subsequent Aussie teams. If you look in to that count, then in today's rosters, which team has players who have delivered on all conditions? I do feel that would be the winner (May be SAF) [[ I will do a total of 3 WC Simulation efforts for CastrolCricket. The first is a pre-tournament one and which has already been published. The second one will be at the mid-point of the League phase incorporating the real-life results of the 21 matches, including player form. No different to first. However the third one is quite different. That will be done at the conclusion of the league phase. So it will only be a simulation of 3 knock-out matches. I am looking at radically modifying criteria and introducing, head-to-head, tournament form, previous 3-key-match streaks et al into the simulation. I have not worked out the details since that is more than 5 weeks away. Ananth: ]]

  • Rangarajan R on February 17, 2011, 10:20 GMT

    Ananth, This is an excellent series of articles and as Tony rightly said, none of us have a compulsion to accept your articles. So the abuses are unwarranted. If i didnt like, i may not read yours. Period. I guess the readers are used to commenting on Harsha / Ian Chappel, et al's blogs where there is no response from them and so they are used to abusing. whereas in your blogs, they get a response. [[ I remember an excellent Sambit Bal article on Tendulkar which was awash with vicious and crude responses. But these were all published and never read by the author himself. Sambit told me that he could not really afford to read all the responses because he had no time and he does not want to be influenced. His opininions have to be expressed free of any bias. Maybe there is a filter to catch four-letter words and then anything goes. I work differently. Your comments are my lifeblood. I get inspired by these. I correct my mistakes (there have been many, I accept) from the comments. Ideas for my articles come from you all. I must say that over the past three and half years, readers have come to understand the way this blog works and more than 90% of the comments are fine. Ananth: ]] But apart from the usual stats covered, do they also see the performance in multi country tournaments? Indians have always shown two clear traits: Start with a bang and fade off (or) come back from behind and raise visions of victory, only to falter in the end. In most cases, India have always gone to the finals by winning the last match (which would be a virtual eliminator)where they are observed t lose all their steam. They have done well in smaller triangular series (2 + 1 finals) but have invaribly faltered (but for 2008) in longer tourneys. (cotd...)

  • Alex on February 17, 2011, 8:20 GMT

    Ananth - As an intro to your team strength article, early this week on CNN IBN, Rajdeep Sardesai asked the panel of Imran, Viv, and Border to nominate the greatest WC team ever. Border picked '79 WI ... he thought '03 and '07 Oz can compete with '79 WI but will lose ultimately. Viv, bless him, didn't even know which year world cups are held. He eventually nominated '87 & '03 Oz teams as the teams that would have competed well with WI. Imran thought the greatest ODI team ever was actually the '83 WI team that got beaten by India.

    If we think about it, the only weakness of '83 WI was #6: Bacchus. They lost because (1) Haynes played a stupid shot, (2) Viv went for one shot too many, (3) Greenidge got Sandhu's delivery of life, and (4) Lloyd pulled a hamstring. Had Richardson been born 3 years earlier, he would have replaced Bacchus and this team would have had no weaknesses. [[ Alex My next article is on Test Team strength not on ODIs/WCs. If one concludes that Croft was as good as Marshall (in ODIs) and Murray was as good as Dujon, it comes down to Kallicharan/King vs Gomes/Bacchus. I think the first pair wins hand down. So I have to go with Border and the 1979 WIndies.In my numbers, the 1979 team is ahead slightly. Ananth: ]]

  • Tony Purcell on February 16, 2011, 23:48 GMT

    We're all keenly awaiting your next article - but it will undoubtedly be worth the wait. As an Australian supporter I'm not too worried by this end of the tournament. If 1999 proved anything it was that it's not how you start it's how you finish. I don't think any team can bank on getting past the quarter finals, and there is a lot of time for form to change as the foreign teams become better acclimatised. Whoever wins will have thoroughly deserved it - and while my personal favourite is Australia the main thing is that we see some great cricket played - and watched and discussed - in the right spirit.

  • Alex on February 16, 2011, 18:17 GMT

    Ananth: this article has generated a number of comments (Rajan Parrikar is creating a similar community driven blog on WC '11). I sometimes wonder if it is a case of a person posting comments under different pen-names. It is difficult to filter that out and everybody can say just about what they want but I wish the comments add some cricket-related info and not just personal feelings. A community effort like such a blog together with cold stats can unearth several great performances which are otherwise ignored.

    E.g., Kapil's 175 & Viv's 189* are considered the pre-eminent ODI innings ... however Sanath's 189* might be considered their equal esp. as it came in a final. Likewise, no one talks about Powell's match-winning 124 vs Ind in '99 Coca-Cola final. Words were shed in readers' comments on SRT's WC performances in response to your previous article but no one pointed out that, due to injury, SRT entered WC '99 without playing an ODI for full 5 months. [[ Yes, that was one great innings. I watched on television. West Indies had no hance of winning.Entering at 67 for 4, chasing 250+ with almost no chance. Powell played a wonderful innings. Incidentally it is ranked in the top-25 in my all-time list. Ananth: ]]

  • Boll on February 16, 2011, 13:22 GMT

    Ananth, I take it this is your last article until the end of the World Cup. Disappointing if that`s the case, but just wanted to thank you for all the work you put in. I`m sure I`m not just speaking for myself when I say that this is simply the best cricket blog on the web.

    Hope you have a great World Cup. All the best from (a bloody chilly) Japan.

    ps. Go the Aussies! [[ Are you by any chance intercepting my mails !!! I have informed Cricinfo a few days back that after my article next week, I will take a break until the first week of April. Possible alternative is that you have put 2 and 2 together and come out with a very educated and correct guess. But my next article is very important. I have completely re-done the Test Team Strength analysis, starting from scratch and incorporating a completely radical period adjustment method. In fact I start the WC simulation on my desktop and during that 20-minute period, look at this article on the laptop. It will be a very thought-provoking article, I can assure you. It is not a bad idea that Australia has started badly in the practice matches. They would clearly go to being fourth favourites and I get the feeling it would help them. India has started like a bomb. This might very well be the problem since suddenly people will anoint them as clear favourites with some daylight from, say Saf and Slk. Really appreciate your kind words. If I have brought a little bit of sunshine into the harsh Japanese winter, readers like you are my inspiration and are like the zephyr wafting through the window.. Ananth: ]]

  • Alex on February 16, 2011, 10:36 GMT

    Ananth & Sarosh: Sarosh is right. SRT overcame great adversity.

    1. In '06 Eng-Ind series, as reported in press, some Eng players referred to SRT as a walking wicket.

    2. Over the entire year 2006, plenty of anti-SRT articles were written demanding an ousting of SRT: e.g., TOI had an "Endulkar" article. SRT's supporters in press were reduced to asking Lara (who was near his best) what SRT needs to do to perform better.

    3. Even after winning the MoS vs WI in Feb '07, the impression that SRT was finished was so strong that after the WC '07 debacle, BCCI called him on carpet to explain his comment on Chappell & dropped him from the ODI series vs BD. Then, the amazing revival started.

    Aside: Although SRT failed in ODI's in '05, he did OK in tests that year --- 441 runs in 10 innings with 4 50+ scores is good. He really was hampered with injuries & recoveries over Apr '04 --- Nov '06. He still managed some good knocks but whatever was on the display was not the real thing

  • Alex on February 16, 2011, 4:36 GMT

    Ananth - TV viewers in India had an excellent opportunity last night to flip channels and time-share 2 dud events which deserved to make a great viewing: NDTV's Indian of the Year and CNN IBN's interview of Imran, Viv & Border by the one and only Rajdeep Sardesai.

    Enough can be surmised about the NDTV awards by noting that Rajanikanth got the Indian of the Decade award and SRT got the Indian of the Year award. On the Sardesai show, Viv was asked to evaluate and make predictions ... unlike Don/Imran/Lloyd/Waugh and much like SRT, Viv has never had anything intellectually insightful to say. Border rated '79 WI to be the best WC team while Imran thought WI '83 was the best ever. They all thought Pak the dark horse in this WC.

  • Tony Purcell on February 16, 2011, 2:25 GMT

    Fellow readers:

    On reading Ananth's articles it is tempting to think that we could do better or that the analysis doesn't consider elements that we think it should. However, he has spent many hours on each project & has applied his considerable talents to their analysis. This isn't to say that questions are unwarranted or unwelcome. Many readers produce fascinating insights (& Ananth happily incorporates suggestions that might improve his results).

    Assuming that we, on the basis of a few minutes idle consideration, possess understanding & judgement superior to that Ananth has gained through many hours slogging at the coalface of cricketing knowledge, is neither well-mannered nor well-founded.

    I firmly agree that we shouldn't be unquestioning, and I would think Ananth does too, but neither do we need to be rude, dismissive or unappreciative. I enjoy reading the blogs and all your comments - let's temper our curiosity with courtesy - as the vast majority of you already do. [[ Tony Most of your points are well-made and reflects the current situation. Many thanks for that. The only reason I stay on this energy-sapping, time-consuming, but ultimately satisfying blogging effort is because of the interactions I have. What I cannot understand is someone who sends a comment at 10 PM Indian time, which I see an hour later, but keep it for answering the next day since I am really tired (AFter all I was born in between the two great Chappells !!!) and the person sends a rude and vicious comment six hours later. What drives these people. Ananth: ]]

  • Tony Purcell on February 16, 2011, 2:04 GMT

    Ananth,

    It is understandable that some readers want to see the gory details but they should reflect on this: This software & study were commissioned by castrolcricket - who paid real money & entered into a contract.

    We can thus conclude:

    a) You're bound by the terms of the contract not to distribute for free information your client has paid for. There is a firm legal limit on how much detail you can share (beyond the fact that cricinfo just won't offer gigs of test runs for download).

    b) Your code & methodology are solid enough that a commercial entity considered it worthwhile to purchase use of them. Just like the inestimable Messrs Lynch & Martin-Jenkins, they wouldn't do so if you weren't able to demonstrate its validity & power. Further, if any schmo could do just as well with a laptop and a spare weekend they'd use in-house talent rather than pay you. [[ The Times effort was a gift from me to them. Castrol is a commercial arrangement. To their credit they have never been touchy. The final numbers, I leave it to them to publish. Ananth: ]]

    In summary: Strong evidence of the accuracy of the technique is already available.

    To be continued

  • Mohammed Amdani on February 16, 2011, 0:17 GMT

    Who cares? As a Pakistani I pray to Allah that Pakistan wins the world cup.

  • Gaurav on February 15, 2011, 22:07 GMT

    Tendulkar - the walking wicket..Just made my day !! LOL ! Ananth, would we be able to see your match predictions with CastrolCricket anywhere on the web or their site? Also,by any chance, are you playing the Cricinfo Fantasy Game? [[ A funny way to call one of the all-time greats. The Castrol Predictors have been announced in press. Sometime these should be coming up on the websites also. I do not know the exact schedule. The match predictors, I presume will appear from 18th onwards. I send my simulation results one day before the match.. No to the Fanfasy game. Ananth: ]]

  • DOUGLABOY@WINDIES on February 15, 2011, 14:32 GMT

    Dear Anath, I just want to say that this is an outstanding article and that I really respect all the work you have put in. My comment was not trying to have any innuendos or insinuations, it was just out of my own curiosity. Also, any literatue on stats that you would recommend? [[ Douglas Pl don't think you are the one I referred to in my reply. No way at all. Your comment was an excellent one and I responded in detail to that. Ananth: ]]

  • jeet on February 15, 2011, 8:12 GMT

    May India reach WC final but it could be a difficult task for them due to overconfident and more expectation from us.... We wish always be a winner India at top ...hope soon our dream come into reality..and they give us a chance to joy like 83...!!!

    Wish Team India All The Best..!!

  • Ramesh Kumar on February 15, 2011, 7:51 GMT

    Ananth,

    Very good work and the follow-up discussions are thouroughly enjoyable but for a few unwarranted provocations.

    1. While you do simulation, do you differentiate 80/20 & 55/45 possibilities for determining the result? [[ A single match simulation is never done on the computer other than for testing. Serious work involves running a minimum of 100 runs, whether it is a head-to-head which i will be doing for CastrolCricket (that is my client) on a daily basis or the Tournament simulation (4900 matches). Each set will come out with a different result. For the Tournament I posted each set's results into an Excel file containing the Team strengths and got the CC. I finally selected one with nearly 0.9 as CC. I have yet to decide how I would select for the head-to-head simulation. One thing is certain, it will be based on relative team strengths and not on my personal preferences. So the simulation, per se, is a totally independent process So 80/20 or 55/45 do not come in. The only things which matter are Player data, ground data, innings/match status and the strategies (unfortunately on the computer, not very dynamic) Ananth: ]]

    2. What is the sensivity of upsets in the league phase? Given QF being A1 vs B4, it is important to finish at top and any upset in league phase can change the equation in QF. Given the strength of Group B(Ireland looking good), one upset can make a team to be B3/B4 and hence to meet A1/A2 in QF. Is this point relevant in the simulation? [[ No my tournament engine does not look at the table status at each match that time. The league table and the NRRs are developed only at the end and then the QF pairings determined. Each match is simulated on an independent basis. Otherwise it would become too complicated to delvelop, but more importantly, to test. Ananth: ]]

    Sorry if I have not caught the explanation or the methodology.

    Ramesh

  • Tony Purcell on February 15, 2011, 7:40 GMT

    Developing numerical simulations is an art form. One doesn't want to incorporate too many variables, lest you end up with an unwieldy mess whose response to parameter changes is confusing and counter-intuitive. Conversely you don't want too few parameters because then your results are very coarse & lack resolving power.

    How does one decide what the "right" number of variables is?

    Generally, by constructing models and testing them, running the simulation many times, & checking that the results accord with the observed behaviour of the modelled system. This is what Ananth has done. Barring contractual obligations (& his desire to preserve his intellectual property rights), Ananth could make the code & the test runs publicly available so that we can verify his work.

    But:

    a) I don't *want* to slog through thousands of printouts b) If I did I'd write my own simulation c) If I don't trust Ananth's judgement or veracity in presenting his results, why on Earth would I be reading his blog? [[ Tony You have made my day. When I ran the individual player testing program during 1989, with an IBM PC XT (16mhz), it took 9 hours to complete the job. I did not have Excel with me to do the summarizing of tabulations. That was another program and 3 hours. I can tell you, when I got Bradman's average over 1000 innings as 74.5 (his base was 0.8 * 99.96) and Chandrasekhar's as 2.7, it was as if I had won the Nobel Prize. Last week I ran my final simulations, the last of the 72 sets took 20 minutes (The program simulated 4900 matches). I decided on set no 68 which had the highest Correlation Coefficient at 0.893, I got the same feeling of satisfaction that I had got a statistically relevant solution (the CC) to the vexed problem of avoiding biased selection. In fact you have questioned me most. However you have done that from a base of belief and with a view to learning more. I have explained the simulation methodology for hours on end to Steven Lynch and Christopher Martin-Jenkins. Only after they were satisfied did they consent to doing a complete coverage of the simulation in Times. Now you see why I see red at innuendos and insinuations. Ananth: ]]

  • Alex on February 15, 2011, 7:32 GMT

    Alex - Madan Lal was a surprise in '83 WC but Binny was not. Binny thrived in conditions that aided seam & swing -- e.g., Wood & Border had no clue how to play him in ODI's down south. Had he been born in Eng, he would probably have enjoyed an excellent career. Kapil & Binny was definitely a potent ODI attack in England. It was a different story on Indian pitches though and WI absolutely destroyed India in India --- seeds were sown earlier ... I often think Holding's delivery (which is worth watching from the beginning of the run-up) to Mohinder in the WC '83 final itself started Mohinder's horror phase.

    The Pak team that toured India in '83 after the WC was quite ordinary, esp. in the bowling department.

  • bks123 on February 15, 2011, 5:43 GMT

    excellent job..my inner feeling is SL vs AUS final. And AUS 4th in a row. However, I have a request to you. If you look at the WC champions so far from 1975 to 2007, we see the teams that won has their best captain of all time in charge (except may be Aus, to me probably mark taylor was Aus's best captain who could not win it). Sometimes they even led teams multiple times in WC like is doing. And they led from the front...Is there a way to find out which captain has contributed most for their country's win in the world cup title and a way to predict one for 2011. Thanks [[ I don't think I am going to anything more on the WC other than one comprehensive WC performance analysis at the end of this WC. Ananth: ]]

  • Yours Truly on February 15, 2011, 5:33 GMT

    That is the point, we don't really know how much and what kind of work you have done until you tell us the details. I don't mean you did not do it but certain procedures should be followed and simulation is not an exact science as you may like to believe. I was not making a literal comparison as far as you selling a car is concerned. It is just an analogy just like one you used on Kamal Hassan and Sachin Tendulkar. What's wrong with that? Sometimes your replies to blog readers is really over the top. I know you get irritated by some readers from time to time but you can't use the same stick for cow and a crow. If you are so confident in your analysis why don't you tell us who will win the WC in 2011? I am not writing this to demean your analysis but to show that explaining past performance by a model (your reply to douglaboy) is inconsequential and marginally deceptive. That is called retrofitting the data a trick that many stock analyst use to show why their model will fare better. [[ 1. Only the naive and superficial guy will ask the foolish question "If you are so confident in your analysis why don't you tell us who will win the WC in 2011?". Do you think I can run the simulation once and proclaim to the world that team abc would win the WC. Only a fool will do that. I am not a fool nor am I a soothsayer nor an astrologer nor a journalist. Instead I have run the tournament 100 times and have said that these four teams are the leading contenders, a fact which seems to have escaped you. 2. In my detailed replies to Ludd, Tony Purcell and Fuzzy, amongst others, I have explained in quite a bit of detail what has been done. I am not sure you have gone through that. Anyhow that is it. That is all I can give. 3. In my reply to Douglas, I wanted to prove that the 1983 Final was the only really upset victory in all the World Cup Finals. So I ran the simulation 100 times and got a 5-1 ratio in favour of West Indies. That is all. There was no retro-fitting nonsense you are talking about. In fact it is the other way around. If I had simulated the match today, 27 years later, and said that India-West Indies were 48-52 or something like that what you said might be true. You have not even realized that I am admitting that the simulation program would not have predicted the Indian success. Anything goes, right. As far as my responses to readers are concerned these are reflection of the tone and content of the readers comments. Both Ludd and Tony Purcell were in fact questioning many of my assumptions like a few others have done. Except that they understood and appreciated the work and then asked pointed questions. These were all answered by me in depth. I suggest let us close this now. Ananth: ]]

  • Yours Truly on February 15, 2011, 2:02 GMT

    You are selling a car without letting us know what is under the bonnet. it is queer for you to publish these articles without full transparency in to the methods and processes. You agree? [[ Am I selling. Are you buying anything from me. I am not a stock-broker or automobile saleman or Insurance salesman. Do we have a contract. I suggest kindly get off such nonsensical comparisons. I have put in thousands of hours of work and come out with multiple simulkation models, analyzed, accepted and published by top publications around the world, including London Times, during the past 20 years. I have provided here a small part of that model. You can accept the results and/or question these. I have answered almost all the questions diligently. What you cannot expect is that, in order to convince you, I will put every single bit of 20-year expertise in an "Open Source" arena, kindly get off that idea. I have been given a blog space by Cricinfo. You have come here yourself. It is your choice. No one forced you. Do not come here and say that I am selling you something under false pretenses. I have nothing to sell, no products, no services, absolutely nothing. Ananth: ]]

  • DOUGLABOY@WINDIES on February 14, 2011, 20:55 GMT

    Dear Ananth thank you for a thought provoking article. However,I am wondering whether the criteria you have chosen truly has any bearing on the results. For example, before the 2007 World cup, I remember an Australian team (albeit not at 100% strength) lose to New Zealand 3-0. Furthermore, that Sri Lanka team of 1996 had in no way a high batting average. Furthermore, I doubt that a home field advantage would help Bangladesh overcome a team like England, considering they were whitewashed in previous ODI series. Have you seen the stats from before previous world cups and found a correlation to success in the group stages? As a budding statistician, I have learnt while reading Sabermetric books and reading books like 'Moneyball' on baseball that often seemingly importants statistics, like RBIs do not have as profound an impact on the game as seemingly lesser statistics, like on base percentage. I guess when we decide to look more carefully at cricket statistics we may be surprised ...... [[ Out of the 9 WCs, 5 were won by the favourite teams. Let us look at the other 4. 1996 was won by Sri Lanka with 3 confirmed match-winners. When I simulate that Final, Sri Lanka has the edge over Australia. 1992 was won by a team of brilliant players over a team of competent players. Similar to 1996. This is again either side of 50 situation. 1987 was the year when England had a 55-45 edge over Australia. Who knows what might have happened if Gatting had played a normal shot. 1983 was the only real upset in the history of WCs, as far as the Final is concerned. When I simulated the match now I get 80+ for West Indies and 20- for India. So if the match is played 6 times, Win would win 5 times and India once. That is what the pundits said before the match and Alex said a few hours back. It was a combination of various things. The 125-km delivery of Sandhu pitching and moving perfectly, the attitude of Richards, the catch by Kapil, the 110-115 kmph bowlers of India putting the ball perfectly on spot et al. The point is that the win was the one in six situation. No one knows what is going to happen. A few guys are saying these atatements off the top of their heads. Not that they are wrong. I have put in more than couple of hundred hours of work and have said that the top four teams have just above 80% chance of winning. That is all. In my program I cannot plan for unexpected things to happen. You also have to go by the fact that in 9 WCs, there has been only one wholly unexpected winner. Ananth: ]]

  • RS on February 14, 2011, 16:52 GMT

    Can you please write the same kind of article after the world cup on the basis of world cup performances and compare that on with your current analysis. Thanks [[ If Ireland or Netherlands win, I will be the happiest person around despite the fact that I would have been proved wrong with my analysis. Anyhow 100 other guys are making their statements and predictions. Everyone cannot be right or wrong. I could do all analysis. Tomorrow Tendulkar or Amla might be injured and that changes everything. Ananth: ]]

  • Kaundinya on February 14, 2011, 15:05 GMT

    Hi Ananth, I have read both the analysis of yours' about individual performances and about the team strengths. A lotta good work really. All I want to know is how much weightage is given to 2009 performances and should they carry any weightage at all. When I look at it with about 20 to 25 matches played every year (on an avg) perhaps it is more for soem teams, should the 2009 performance indicate recent in any sense of the word. I can understand world cup performances carrying weightage indicating how well they perform for the occasion(even this should carry only a little weightage), but 2009 is far from indicating the current form. I mean if it carries a weightage of let us say 25 % and 2010 carries 65% and all the other factors(carreer averages and world cup performances to 10 %) will the 200 of sachin be counted as equivalent to 84 in 2010 right( which I don't think is very reasonable). Career averages make sense too for more than 100 innings(they do indicate consistency). [[ Two things. I go to 2009 ONLY if I do not have 10 performances in 2010-11. That is there for very few players. The other thing is that the 2009 performances carry only 50% weight. Ananth: ]]

  • Sarosh on February 14, 2011, 14:18 GMT

    Ananth: Though I understand the difficulty of conceiving of SRT as a 20 something average batsman , the fact is that at certain times in his career he was actually operating at a much reduced level for extended periods of time. For eg. If we consider Test cricket. In 2005 he averaged 44 over the course of 10 inn. In 2006 he averaged 24 over 12 inn. due to form/injuries. In contrast when in good form in 1998 he averaged 81 over 9 inn. And in 2010 he averaged 78 over 23 inn. These are not statistical anomalies. In 2006 his performance was actually so poor that it was strongly believed he may retire. He was known as a walking wicket and it was almost embarrassing to watch a great batsman reduced to such a level. In 2006 it was quite conceivable that he would actually not score more than his year average of 24 in his next match. In 2010 it was reasonable to expect he could get a hundred or at least his year average of 78 over the course of the next match. So, though we may not associate SRT or some other batsman with a particular figure, in actual fact for certain periods of time it is often the reality. [[ You have a point. However we must give the great batsmen the opportunity because of the great track record over the years. I will ask you only one question. Whether he was a walking wicket (where did you get that !!!) or not, at no time was there any thought of dropping Tendulkar. Ananth: ]]

  • Yash Rungta on February 14, 2011, 13:55 GMT

    Ananth, just wondering... as you told, if India qualifies(which it obviously should).. the 2nd quarterfinal which is to be played in Ahmedabad will the one??? Or will some other quarter-final be played in India?? Because it says A2 vs B3.. but what if India are B2 or B1 or B4?? [[ I get the feeling that some of these matches would be shifted in a hurry. Anyhow even now the schedule says A1 vs B4 etc. It is possible that, come what may, the QF in which India is involved might be played in Ahmedabad. Similarly if Sri Lanka avoids India, they would be able to play in Colombo. Re the other matches, anyhow they would all be playing away. Ananth: ]]

  • Karthik on February 14, 2011, 13:22 GMT

    Hi Ananth- I always read your article, but writing a comment for the 1st time. This is interesting analysis with the available data (how can one rate X-factors, rain, dews, etc)! I've 3 questions: 1)How do you consider players who have either not played in the last 2 years -like Piyush Chawla, who incidently was quite good in the warm-up- or have never played Intl cricket -like Imran Tahir, who also seemed decent in his warm-up and who has a very good chance of being selected in the 11 considering how spin-friendly some of the warm-up match wickets were? I wasnt sure if i missed that bit in the analysis/ comments. [[ At least Piyush Chawla has a career of 22 matches and 28 wkts. So that will be the basis. For Imran Tahir I would start with a bowling average of 30, for want of no other information. 30 is what I always start for debutants. Ananth: ]]

    2)Do you consider toss in your analysis? This might become crucial with the dew-factor and captains winning toss will always have better use of conditions (i hope MSD also practices tossing the coin apart from cricketing-skills:-). [[ The Toss becomes crucial in the human-controlled matches since there is atime for strategies and counter-strategies. In the computer-controlled simulation, this does not come in. There is a generic determination of strategies, that is all. Ananth: ]] 3)Don't you think it might be worth-while looking at how teams performed chasing/ defending targets recently, if we consider toss? [[ The problem with this that, if you consider a number of matches, these factors even out. However factors like chasing a big total and the consequent pressure on the batsmen have been built in. Ananth: ]]

  • Satish on February 14, 2011, 12:05 GMT

    NZ: SF,QF,SF,Super Six,SF.. Consistently good performances over the past editions.So expect a first round exit..

    My analysis was based purely on the law of averages and I know it is very flawed.It is a well known fact that under real match conditions, with a packed house baying for the players' blood , all these laws and predictions go for a toss.. [[ If India meets a team like Pakistan in the Final, remember that Pakistan, like India, has two explosive matchwinners at 7 and 8. In fact I feel that no other team has this asset like these two. Mike Hussey gone. Morgan gone. Mathews is the only other possibility. Ananth: ]]

  • Satish on February 14, 2011, 12:01 GMT

    Great Analysis Ananth.Statistically on the dot and it is totally unbiased.However,here is an analysis going by the law of averages.

    India: We do well in alternate WCs-87,96,2003.Everytime we perform miserably in a WC, we more than make up for it in the next edition.96- SF, 2003- Finals. So going by the same yardstick, 2011- Winners.

    Australia: They have won the last 3 WCs. So they are due a bad performance.An exit in the QFs sounds plausible.

    SA: Always consistent.Except for 2003, have made it to the knockout in every other edition.If you look at their performance from 1992, it reads as follows: SF,QF,SF,Super Six,SF.. So going by the law of averages, a QF exit.

    England:Finals, QF, First Round,Super Six,Super Six from 1992-2007. A SF spot seems very possible.

    SL: First Round,Winners,First Round,SF,Finals from 1992-2007.Due a bad performance.So most probably in exit in the QF.

    Pak:Winners,QF,F,Super Six,First RD..The law of averages never hold good for them!So no predictions..

  • Alex on February 14, 2011, 11:31 GMT

    Ananth - you should stress that the numbers your formula generates are valid for estimating performance over a batsman's next 7-8 matches and not in a single match ... what happens in a single match has a lot more randomness which hopefully averages out over 7-8 matches.

    1. Therefore, way too much was read into batting failure of SRT & others in '07 WC. 3 games is too little a sample set to draw those types of conclusions. India flogged the minnows in the 2nd game but played poorly in Games 1 & 3 and were gone just like that. [[ That is a very valid point made by Alex. Incidentally that happens in real life also. When Sehwag walks into bat on Feb 19, no one can predict how many he would make. He might want to back up his words on Bangladesh, go for a few big hits and get out for 10 or go on to make a 100. What is certain that if Sehwag gets to bat 9 times in the WC, he is likely to score 350 runs + - 10%, almost certainly at a strike rate exceeding 100. If he has a phenomenal tournament he might go 25+% more and if he has a poor tournament he might go -25%. If Tendulkar gets to bat 9 times he is likely to score 450+ - 10% at a lower scoring rate. That is real life and what has been built into the simulation engines. Ananth: ]]

    2. For this reason, India's '83 win was a real fluke/outlier. The bowlers (Binny and Madan) massively over-performed and were among the best across the tournament. The batting was mediocre but the batsmen chipped in (Kapil top-scored for India!!). After the big upset in the final, they went back to the old ways and were thrashed 6-0 by WI _in India_ within 5 months of the '83 win. Likewise, any team that sticks to this blue print can win '11. [[ Yes, what you say is probably true. Indian bowlers were on a zone, not reached before or after. However remember that even though India were 19-9-10 for the year, they had a run of 6 consecutive wins starting at Tunbridge Wells and ending with 2 wins against Pakistan after the WC. Then lightning struck. Madan Lal, Binny and Amarnath were not bowling in England but in India. Ananth: ]]

  • QRK on February 14, 2011, 11:29 GMT

    WHO'S ur fav team... plz let us know.... hahaha [[ I will leave you to do the guessing for some more time !!! Ananth: ]]

  • Mizanur on February 14, 2011, 11:20 GMT

    What happened ICC? will they going to be mad? Now they are already half mad. The Indians, The English and The Australian will be most guilty for destroyed cricket. Only one problem about increasing supporters or fan of cricket in one country. That is ICC. How foolish they decided that world cup play 16 to 14 and next world cup 14 to 10. How Rabbish is it? I think that If they want a cricket world of its fan they should give test status to holland, Kenya and Ireland with good ground facilties and All played world cup without prequalify. and 4 teams would be play after qualify. If thay have no oppurtunity why they play cricket. At least 18 to 20 country must be played in world cup. Like example The Asian, The Europian, The Africans, The Amirans Top 8 teams played in frequalify final round and four of thems will be play in world cup. [[ Your anguish is understandable. However this is not the forum for expressing that. Please go to the ICC website and mail them directly. Ananth: ]]

  • Sarosh on February 14, 2011, 10:11 GMT

    Ananth, Srikanth: In my view, RF (recent form) should be given very high weightage. Higher than career average at all times. My understanding of “average” is very different. When considering a long career - A player with an average of 45 would often have innings in bunches where he averaged perhaps 80. At other times this may have come down to 20. These sequences may extend over months.The “average” over his career may work out to be 45. However, this career average is quite misleading in many instances. Form is an inescapable reality for most top flight batsmen. Career average is useable if for example you had to chose a probable score SRT would make if he played 450 ODI innings at random. A good guess would then be 45. However(when not at random and in reality) at various points of time this average would diverge considerably, depending largely on his form. Going only by Recent form may actually yield more accurate results. [[ If Tendulkar has a 450+ match career average of 45 and over the last 10 matches his form is 20, it would be silly to consider him as a 20-level player. Similarly if he averages 80, he should not be put at 80+ level. Hence I have added both after assigning weight to the values. The weight ranges from 25% for players who have played over 100 matches to 100% for players who have played 10 matches or fewer. If you see the final numbers, soon I should be able to show these to you, either directly or as a link, you will know how well this works. Ananth: ]]

  • Gans on February 14, 2011, 9:41 GMT

    Ananth, I am still confused about the QF line-ups... The schedule is defined as A1 vs B4 at Mirpur... If India is B4, SL is A1 then will the match not be played at Mirpur?

    btw, What is A1 - A4 and B1 - B4? Is it based on the group league points? (or like they had it before, it is the ranking at the start where by if the A1 is assigned as Aus regardless of Aus points in the league stage and if Aus is knocked out, then A1 becomes Zim/Ken/Can (quite unlikely though!) :) ) [[ If I am not wrong, any time there is a match-up between two Asian co-hosts (3 teams) in the knock-out rounds, other than the Final, the match will be played at the higher ranked country. India always and Sri Lanka against Bangladesh. Ananth: ]]

  • Ashraf Uddin Ahmed on February 14, 2011, 8:40 GMT

    Thanks for your analysis. I will be very happy if this becomes true. Every nations want to be winners. No need to tell that we want to win also. At past we were being ignored but now the name of my country is appearing in the news and analysis and all are realising that we can win also. Really I am proud of this. Thanks again for you article.

  • Sid on February 14, 2011, 7:53 GMT

    Ananth, Any reason why you have taken only the last two world cups? Especially given that we do have players who have been around for much longer - Ponting and Tendulkar being obvious examples and their past world cup records (especially Ponting with his World cups wins) will be a factor. [[ In fact you have given me the answer yourself. Out of the 210 players who are playing in this WC it would be difficult find even 10 survivors (Ponting, Tendulkar, Afridi, Razzak, Akhtar, Kallis and there we stop) from 1999. So I have taken only the past two WCs I can accept that it is only Pakistan, with three survivors from 1999, could say they would carry forward the experience. There is a case for giving this a slight weight. Would not make much difference, though. Good point. Ananth: ]]

  • Alex on February 14, 2011, 7:06 GMT

    Rizwan and Ananth - whoever has the money & the power, gets his/her way. It has always worked like that. No use blaming the BCCI for them doing what is good for Team India. With all these factors in place, India might actually win this WC ... in that case, regardless of whatever the scorecards read, the real MVP would be MSD.

    Ananth - I don't go overboard on SRT but calling him "Kamal Hassan of cricket" is too much. Sachin is arguably the world's best of his generation at his specialized task; Kamal Hassan is a Yuvraj at most. If it has to be a simile from India, calling Sachin "MS Subbulakshmi or Alladiya Khan or Ramana Maharshi of cricket" is more to the point. [[ I have no problems in deciding on Tendulkar as the best of his generation. Similarly Kamal Hassan is the best of his generation and possibly earlier also. Only the people down south of Hydearabad will know the range of films he has done. He cannot be judged by his forays into Hindi films. They were average. He is not a star, but an actor par excellence. He is the one actor the other actors down south watch.I will stand by my statement. And this is showing no disrespect to Tendulkar, and let us say, from my point of view. And let me also say, Kamal Hassan, in his sphere of excellence, was as good as MSS. Let us close this thread. Ananth: ]]

  • Vinish Garg on February 14, 2011, 5:36 GMT

    Ananth

    I read the analysis 2 days back and realised that I would not be able to comment for many days, owing to some work commitments. However, some insightful and intelligent comments by readers and your trademark polite way to respond made me squeeze in time today.

    Excellent analysis and I second your thoughts on the analysis parameters. I guess it will be history (South Africa may not win it), pressure (India may stumble again at semi-final), and unpredictability (Pak may make the semi-final). I do not fancy Sri Lanka either. So, my vote for final is England (Yes, they would rise to the occasion this time till Final at least) and Australia.

    For you – It is no fun that fans/readers post comment when they have time. The challenge is to do it when we are screwed, and you help us value ourselves being cricket fans. Thanks for that. And for your temperament – what else do you do apart from Cricket and Yoga? I doubt if you can be so cool to respond to nasty or irrelevant comment [[ No, I do not do Yoga. However when I get one (or two or five) nasty comments I can only think of the hundreds of other wonderful readers who live from Japan to SF and Britain to New Zealand and who understand and appreciate what I do. They criticise constructively. They would take me to task, but not in a vicious manner. Half of them would not agree with me but not one would cross the line. There is no malice. These readers make me want to continue this thankless but fulfilling part of my life. Ananth: ]]

  • ak on February 14, 2011, 3:19 GMT

    Ananth - beautiful analysis, and even wonderful interactions with the readers!! Thank you!

  • india is the best of all on February 14, 2011, 3:11 GMT

    All the comments is right on the situation but the question remaining who can defeat pakistan?

    rgds

    long live india

  • rizwan on February 14, 2011, 2:04 GMT

    Ananth

    I agree with you that millions do not bring happiness .However , that will provide me the opportunity to take a break for one whole year and visit the places I can only dream of like Ooty ,The Himalayas , the Amazon , Petra-Jordan and Argentina (to pay homage to my hero Che ) I hope also it would provide me the opportunity to meet IMO the best actor ever in any genre/language the great Kamal Hassan. [[ I understand. I endorse everything, especially the last sentence. When somone once said "Kamal Hassan is the Tendulkar of films", I corrected him as "Tendulkar is the Kamal Hassan of Cricket". My son is doing his Ph.D in US on "Human Geography", a far cry for an Engineering graduate. He is a strong left-winger and his idol is Che. What a man, what a life. Unfortunately the Gods loved him and took him away before he was 40. It is an ironic thing to talk about God when we talk of a Marxist revolutionary. 43 years after his death he remains the symbol for anti-fascist forces throughout the world. Ananth: ]]

  • Tony Purcell on February 13, 2011, 23:33 GMT

    Ananth,

    Very interesting analysis, thank you very much for sharing it with us. You mentioned above that it has been tested, are you able to share any details of that process? [[ There were different types of tests. Individual player tests wherein a player would play thousands of innings continuously against a mix of typical bowlers comprising of 3 good and 1 average. At the end of 1000 innings or so, the simulation results were matches against their averages. The variance was there but within 10% on either side. The batsmen ranged from Bradman to Chandrasekhar. This was done way back. Team results were always matched against the team strengths. Scores against bowling strength faced and bowling performances against average strength of batsmen. Earlier this was done on an aggregate basis. However this time I did the Correlation test as a sure way of ensuring that Team results matched the overall Team strength values, incorporating career and recent form since this was the first time I had incorporated recent form.. Ananth: ]] The technique seems fairly generic, it could it seems be used to simulate the results of test series or ODI series. Has it been applied in that context? [[ You seem to have missed an earlier article I had published on the Test simulation work I had done in colloboration with Christopher Martin-Jenkins and the results were published in Times of London during 2002. This was a Test series played across the world between an all-time English team and all-time RoW teams.The link is below. http://blogs.espncricinfo.com/itfigures/archives/2010/04/ Ananth: ]]

    Understandably, your analysis cannot incorporate critical small-scale events,so the result of any one particular game remains subject to great uncertainty. [[ That is the charm. However if the game is not simulated in the computer at rapid pace with no manual intervention but played at a leisurely pace between two external captains, there is great scope for strategies, bowler deployment, batting order, do-or-die instructions for a short period etc. I have conducted many tournaments like this between schools (a 3-day event) and what we simulated for Sportstar during 1990. Ananth: ]]

    Given two teams in a head-to-head contest, how many games would be required for this technique to make a reliable prediction of the result? I guess that depends on the margin of the difference between the teams, so let's say the two teams in question are Australia and England playing at a neutral venue. [[ In the head-to-head work I am doing for my client I am doing 100 runs of the match. I have found that that produces the final results reasonably close to the team strengths. Unfortunately I cannot reveal the numbers. However I can only say that where the two teams are quite close in Team strength indices, as the four leading teams of the WC are, the numbers do not vary too much on either side of 50. But if Australia play Canada, it is almost always a complete domination. Incidentally this match simulation was also done during 1999 WC and published on the day of the match, half an hour after the teams were known. Ananth: ]]

    One further question and this might be a bit tricky. Has the introduction of the UDRS, by removing egregious umpire error, made your system more accurate, less accurate, or produced no change? [[ No my system has no umpires or umpiring decisions. That is real life. Ananth: [[ Finally a word of sincere thanks. Some readers have resorted to innuendos and crude statements without understanding that these simulation programs have been written, tested, modified, re-tested, published for over 20 years. I thank you for asking me pertinent questions. That is the way to go. You respect my work and I will respond in kind. Ananth: ]]

    ]]

  • Gulshan Singh on February 13, 2011, 19:30 GMT

    Looks like my questions and suggestions are too uncomfortable for you so you are not ging to publish it, because you know you are not doing the right thing on that front. So despite all your pretenses you are just another creep n the making. [[ I saw your comment at 11 PM and wanted to answer it first thing in the morning. However you have jumped the gun and resorted to crude vulgarisms. So the only way is to answer in kind, **** ****. Ananth: ]]

  • Fuzzy Feeling on February 13, 2011, 18:39 GMT

    You are at-least right about the title of the article "picking the favorite" :) [[ The article does not have a single existence of the word "favourite" and the title of the article is "World Cup 2011: an alternate preview". And Cricinfo's caption for the photograph is their choice, not mine. Ananth: ]]

  • sudharsan on February 13, 2011, 17:11 GMT

    Its all on the particular day performance matters. Nothing to go with stats

  • Anand on February 13, 2011, 16:42 GMT

    Ananth:

    Just realised that my previous comment may seem out of place for this article. Just wanted to add another one. Going by players' RECENT form in terms of match changing performances, can we come up with a list of players to watch out for. Your analysis already provides the teams to expect to be doing well and potential dark horses. But if we go by players who can change things (I am not including the likes of Sachin, McCullum, Ponting, Kallis, Steyn, Gayle, Afridi, Sehwag in this list. They can do wonders on any day irrespective of current form), then than can help identify dark horses better. In the Indian team, I would think Yusuf Pathan fits that definition. In Pak Aamer could have been one but for this unfortunate ban. Do you think thats a wothwhile exercise?

  • Anand on February 13, 2011, 16:34 GMT

    Ananth:

    Just a thought.. Can you come up with a baker's dozen for defining moments in world cups (I can think of Kapil's catch to dismiss Richards, Wasim's two in two, Gatting reverse sweep to begin with) but is it worthwhile doing an analysis of defining moments (not necessarily in the final) which went on the change the fate of the world cup. I understand extrapolation of events happening or not happening is very hard, but some cases (e.g., Aamir Sohail vs Prasad in 1996 was surely a defining moment). I personally believe that defining moments will have more to do with wickets than with runs e.g., I personally wont put Jadeja's slaughter of Waqar as a defining moment. This is because batting moments have to be over a period where as wickets are like one ball (and hence one moment). Again, not sure about the feasibility of this but was curious ... [[ This is a readers' contribution list.I am amazed at the memory of readers and readers can send their suggestiions. There was already an excellent set of comments in the previous article. Ananth: ]]

  • F. Shaikh on February 13, 2011, 9:03 GMT

    Ananth. The detail that you provided on one of the above comments about Q/F being held in the country that is currently ranked above the other, isnt it biased or somethin....for all i know, cricket is a Gentlemen's game, but these acts are actually distroying the honour and sacracy of a biased free Worldcup. Good analysis though...May the best team wins....

  • habeeb on February 13, 2011, 8:42 GMT

    if shewag will come form india nothing to worry he will manage.....

  • Kesavan on February 13, 2011, 8:32 GMT

    hi, An excellent analysis..My query is, though England had a bad series down under, playing badly against the aussie spins, somehow can we attribute it to the long test series which took out the energy from the English.. if taken into consideration guess English index is pulled down badly..my own stats calc (am not gr8 in it) says Ind, SA, Eng, SL in semis..thereon lets wait for surprise.. [[ Australia is miles ahead of England in the ODI rankings. They walloped England 6-1 without Ponting and M Hussey. Lee has come back in great form. Australia has lost M Hussey and England, Morgan, both their best finishers. However you have only talked of the semi finals. After the league round anything can happen. Ananth: ]]

  • Shajadul on February 13, 2011, 6:17 GMT

    Bangladesh is rising and we all hope that they can win at least 4 matches out of 6 in the group. Bangladesh was minno for long time but now it's the platform for them to perform. I honestly hope that they do well and ofcourse get a bit of luck. this is what you always need in cricket.

  • Jahid Russel on February 13, 2011, 6:13 GMT

    I did not go thru any statistical analysis. but my assesments are exactly same as yours. I am bangladeshi. so my support is for Bangladesh (I know it would be amazing if Sakib and his team can go to even semis). Otherwise, south Africa, (they deserved at least one WC in those occurred in 90s) nice article

  • muzammil on February 13, 2011, 5:58 GMT

    hi..good analysis have you tested this on already existing facts...i.e doing the same analysis for previous world cups and match the analysis resutls with actuals???

    i bet you would be wrong for 1987,1992,1996 and even dare i say about the semi-finalists of almost all by 50% (2 wrong teams). else i will convert to the religion of monte carlo simulation cheers [[ There is no need to do this for the previous WCs. Enough testing has been done otherwise. Ananth: ]]

  • shrikanthk on February 13, 2011, 4:27 GMT

    Sorry...The Jonathan Trott example was probably not apt in the previous comment, since both his career average and recent string of scores point in the same direction. Just goes to show how dicey this business is. Will Trott sustain his brilliant run or is his average due for a correction?

    It's anybody's guess. [[ A better example would be Yousuf Pathan. Given below is the extract from my analysis re Pathan. Ind,Pathan Y.K , 31, 3, 694, 24.79, 10, 429, 42.90, 173.1 He has played 31 innings, the first 21 in a very average manner. The last 10 have been explosive. So that is the indicator which sets him as a batsman going into the WC in great form. On the other hand Trott has played only 18 innings. His last 10 innings have been below his career numbers. Ananth: ]]

  • shrikanthk on February 13, 2011, 4:22 GMT

    rizwan: The "Random walk" is a natural concomitant of "efficient market hypothesis". If markets aren't "efficient", then "random walk" theory won't hold.

    Technical analysis refers to the use of recent past historical movements to predict future movements. This is fundamentally inconsistent with the "efficient market" hypothesis as per which all information about a stock is already factored into the stock price and hence studying the past is of no use.

    What Ananth has done is a classic case of "Technical analysis". This is based on the assumption that the recent string of scores is somehow more important than the average itself! There is something to be said in favour of Ananth's approach. It works only when the "Average" is not a reliable metric, as in the case of young cricketers like Jonathan Trott.

  • love goel on February 13, 2011, 4:14 GMT

    An off topic comment

    Re.Black Swan , alas by the time I bought Taleb's book from a used book seller , it was too late .....the crash of 2008 had occurred.If only I had read Taleb sooner ....

    Well I am sure there are many books which predicted even more boom in 2008.If you had bought one of them....

    and I am sure of one thing. Though Taleb made a lot of money from the book, he also didnt made money following his observation in the book. Because if he did, we would be hearing about that and not the book! Hindsight is 20/20 and that is why even the best experts of any sports do not make lot of money betting on it.

  • love goel on February 13, 2011, 3:57 GMT

    I think people are underestimating WI. They have some good players, a little bit of luck and they can a squeeze in a couple of win against good sides

    However for the world cup winner, I think the team that is most stubborn will win the cup. There will not be massive victories but close hard fought ones and the team which just refuses to die will take the cake. For this reason alone, I put Australia, India and Pakistan as the favourites. England is weak, south africa brittle , and Sri lanka while good lacks backup and will crack.

    I will be very disappointed if India meets Australia or pakistan in QF. These 3 should move till the SF.

    Also I will never put my money on SA. They haven't won a single knock-out match in 3 attempts in WC; I dont see them winning 3 knock outs in 1 attempt

  • rizwan on February 13, 2011, 3:41 GMT

    Ananth

    Re.Black Swan , alas by the time I bought Taleb's book from a used book seller , it was too late .....the crash of 2008 ( corectly predicted by Taleb earlier) had occurred.If only I had read Taleb sooner and paid heed to his advice I would be have retired with my millions in Kandy ! [[ Ah! Kandy, one of the most beautiful places we have ever gone to. The lovely Mahaveli river is like the jewel in the crown. We also saw the quaint Asgiriya, unfortunately, a museum piece now. And it is true that millions do not bring happiness. Ananth: ]]

  • ww on February 13, 2011, 3:28 GMT

    Final:Aus vs Ind. Result:India def Aus Mom:_Tendulkar for scoring 134.

  • Mehedi Hasan on February 13, 2011, 3:01 GMT

    I think you guys picked couple of nice point.

    I am Bangladeshi. We, Bangladeshi people will be extremely happy if we pass Semi-final.

  • rizwan on February 13, 2011, 0:33 GMT

    Ananth , your comment abbout India having clout and therefore ensuring home advantage is undemocratic , despicable & unprecedented in the new millennium.Of course in the early 80 s , the only two permanent members of ICC viz.England and Australia were guilty of numerous shenanigans but India's stance can't be condoned because of abuses by the previous powers in the ICC.Is the ICC spineless to allow such preferential treatment ? Where is fairplay and meritocracy ? This is not fair and I am surprised that the rest of the permanent members did not throw a hissy fit . 'Noblesse Oblige ' and India's attitude is a shameless.

    Kudos to you for pointing it out and I wonder if any success as a DIRECT result of that particular favouritism will taint India's performance ! [[ I was shocked to read your first sentence. It seemed as if you were accusing me of "undemocratic , despicable & unprecedented" et al. Then I realized that the sequence of words gave that impression. My complaint is that, come what may. a India-Sri Lanka QF will be played in India, irrespective of the Group standings. That seems to me be wrong. And, if I am not wrong, if this tie happens in the SF it will be the same case. I also blame the Slk Board for not fighting this issue. But, unfortunately, they would be silenced with promise of aid. Ananth: ]]

  • rizwan on February 13, 2011, 0:20 GMT

    shrikanthk

    Were you referring to random walk thory rather than efficient market hypothesis or technical analysis?

    As for your assertion,I believe the analogy of random walk theory does apply in Ananth's analysis ONLY on the RARE instances when a a moment of brilliance (like Kapil's catch off Richards) can affect the outcome of a match.However , technical analysis works MOST of the time and by including all KNOWN factors & Ananth has also ensured that efficient market hyothesis holds true.In this regard Ananth has done a wonderful job of incorporating all KNOWN information in his model and that is the best anyone can do.

    Of course if Ananth had a crystal ball , his job would be a lot easier

  • JB on February 12, 2011, 21:47 GMT

    Congratulations on an excellent piece of analytical work. I appreciate that you have connected high level mathematics to a sound cricketing brain. I agree with you that the dark horse here is Bngdsh. Considering the fact that all, if not most of their group games are held in their own back yard and also from what we all seen of their capability in the 2007 WC, I can sense their presence in the last 8. I am sure India & SA will be very careful with them this time around; it could only be that their qualification will be at the expense of either the WI or Eng. As for winning the world cup, I agree again with you about having three great days. India and SL are my strong favourites. I am thoroughly confused about this venue saga at the K/O stages and beyond, could you please help me on this? From the Cricket Info website, I note the following A1 vs. B4@Dhaka, A2 vs. B3@India, A3 vs. B3@Dhaka and A4 vs. B1@Colombo.

    Supposing the final group tables turned out like Group A- SL(1), Aus(2), PK

  • Zenn Banda from Canada on February 12, 2011, 20:56 GMT

    Hi, This is one of the best articles I have read based on facts and not fiction or whims and fancies of individuals.Yes I agree India , SL, SA and Australia could be the probable semifinalists; However, India will be knocked out because of pressure(from 1.2 billion arrogant fanatics!!) and also they are proven chokers.SA are a very good side but they are also proven chokers.So that leaves SL and Aus in the finals.One of these 2 teams will win the cup as happenned in 1996, 1999, 2003 and 2007. However, Pakistan is the dark horse. They could be brillient if they want to be. Keep up the good work; This is a brillient simulation based on facts.My preference based purely on sentiments (not facts) are SL,Canada, Aus,Pak,SA,NZ, WI in that order. India and Eng are my least favorites because of the arrogance of the fans (fanatics). All the best from Canada; keep up the good work. Zenn

  • Mohsin Irshad on February 12, 2011, 18:33 GMT

    Some wrong assumptions:

    1. As you said Pakistan will be handicapped by the absence of two bowlers, but in this case Asif is not a part of starting eleven. Only Amir was a loss.So figures will be different.

    2. Did you consider the 'big match factor' . South Africa and word Chokers comes to my mind and on the other hand Pakistanis are good during important matches.

    3. "India has good bowling"? You are kidding. Thats a major weakness of Indian team. Still not sure who are there 3 best seamers

  • Fuzzy Feeling on February 12, 2011, 18:12 GMT

    Wow, I always suspected that you play around with your numbers to make them semi objective and semi subjective and produce a soup salad of an article which is neither here nor there. Now you have accepted it.. Does that make you a person who shoots first and then makes a circle around it to make it a target :) [[ Since you have made your point in an inoffensive manner I will publish your comment and answer the same. First it is obvious that you have not read the article nor the comments completely. You have had a cursory run through the article and come out with off-the-top comments. 1. I have mentioned clearly that a lot of work I have done is proprietary and confiidential for another client and I have only used the extracts. You have missed that point. 2. I had answered Ludd's comment, made in a nicer manner, with a clear explanation. You seem to have missed that. 3. The Team Strength is based on 100% objective analysis. There is nothing subjective about it. The career numbers, the recent form numbers are all facts, not fiction. The weights are given based on a number of trials. I have not picked up some number from the top of my head and thrown it on the readers. 4. You may not understand the significance of this. The complete tournament was simulated over 7000 times, yes you read it correctly, 7000. I have a 5000 line C-program which does the simulation. Each ball is the result of 25 different randomizing decisions. This is wrapped by a Tournament engine program which runs the tournament including determining NRRs to the nth decimal. 5.This was not one 5-minute article, generated by the ex-cricketers/journalists, in between dinner and drink. There was well over 100 hours of effort supporting this work. 6. Again you may not understand the significance of this point. The final set was selectee not randomly, not because it matched my expectations, not because that was what the readers wanted to hear. But it was pure statistics. After running each set I did a Correlation analysis of the Team strengths vs the Set results in terms of Tournament wins. The set that was finally selected had a very high, the statisticians would conform this, Correlation coefficient of 0.893. The CC exceeded 0.90 a few times but this was with an Australian team with Hussey. So I had to reject those sets. 6. Unfortunately I have not been able to share the simulation results for reasons explained twice in the article. The results will be published in the next few days and I will provide the links. Then you will know the magnitude of the work that was been done. So contrary to what you have mentioned I have not shot and drawn a circle around it. I have tried to shoot as close to a correlation of 1.00 and stopped when I reached the highest. Maybe you should take off your coloured glasses. Unfortunately you are the one shooting in the night in a random manner hoping you will hit the target. This is it. This is the extent of explanations I can give. Ananth: ]]

  • R.Venkatachalam on February 12, 2011, 17:24 GMT

    Great work boss.. I feel West Indies will qualify even in case bangaladesh enters QF. WI have been good at subcontinent conditions - And some individual brilliance from Gayle, Latest KP with sarwan and c'paul hanging around.. Looks a balanced team with bat. Benn, Gayle, miller ll be very effective. I feels Its going to be England that ll miss out in that group. [[ Yes I agree. My numbers show slightly differently. However West Indies can defeat any team on the day. 3 great days will, however, normally elude them because of their inconsistency. However they have WC wins even in my simulation. Ananth: ]]

  • NALINWIJ on February 12, 2011, 14:22 GMT

    Excellent analysis. I just wonder how to factor in home ground advantage and the knowledge against a specific opponent like India's familiarity with Sri Lankan bowling [especially Mendis] after so many matches recently. There is an unusual world cup pattern where Pakistan has the wood over Sri Lanka who has done well over India who has the wood over Pakistan.

  • Sharif Faisal on February 12, 2011, 13:40 GMT

    Thanks. I meant If SL, Aus, Saf, and Ind make it to the semifinals by being the top 2 in their respective groups, what are the match ups in the semis and where they play? Thanks. [[ The Semi finals are scheduled at Mohali and Colombo. I presume that the higher rank benefits will continue. In which case India will play at Mohali and Sri lanka, if not against India, will play at Colombo. Ananth: ]]

  • Mudassar on February 12, 2011, 12:43 GMT

    great analysis man. I agree with semi final line up. I also think that India & Sri Lanka will feature a semi final at Colombo (1st semi final). Unfortunately that is going to wreck the dreams of one of the hosting side :(

  • Umar Irshad on February 12, 2011, 12:30 GMT

    hi ananth..

    great article from your side as always.

    as your artcle possess estimation of stats from last two years. wht about the players who just started their careers like ahmed shehzad and junaid khan. what if they perform something spectacular on their day along with a blinder from afridi or razzaq. do put some light on this matter too. [[ There is a default values for players who have played fewer than 10 matches. Ananth: ]]

  • Manzoor Raza on February 12, 2011, 11:48 GMT

    Nice article, though I think that anybody who is following cricket and results as a layman can also arrive at the same team numberings without getting into technical details and analysis for at least top 8 teams. The only difference may be that Australia and Westindies may be rated higher than their existing positions. Anyway, its a wonderful and well executed mindboggling exercise. My team - South Africa. Still, My wish team will be Pakistan. It would be great if I see an encounter in Final between these two teams. Pakistan Paindabad.

  • malik on February 12, 2011, 11:33 GMT

    One question here... is there any possibility of an india-pakistan Q/F ?? just like the one in 96. [[ Yes, India could be B2 or B3 and Pakistan A3 or A2 or any such combination. Ananth: ]]

  • m udasir on February 12, 2011, 10:20 GMT

    u cant pridict winner by analysing .it is a game of cricket where u can`t pridict any thing although my choice wl be pak because of its un pidictibility

  • shrikanthk on February 12, 2011, 8:47 GMT

    Ananth: History is replete with enough examples which illustrate the hopelessness of predicting immediate future using recent past performances. Some instances - - Geoff Allott's performance in '99 WC - Matty Hayden's performance against India on 2001 tour - Wasim Akram's show in '92 WC Final (having somewhat struggled in the earlier games) - The remarkable failure of a seemingly powerful Australian team in '69-70 against SA after defeating India in India. - Shane Warne's remarkable comeback on 2001 Ashes tour after being decimated by Indians earlier the same year.

    One can go on and on. My takeaway: Just go by the career averages as the best forecasts for experienced, proven players with some allowance made for the conditions and circumstances. Eg: One would expect a Tendulkar or a Warne to fare better than their career averages on a big stage like the World Cup.

    For less experienced players, make very tentative predictions based on recent form. But with very low confidence levels

  • zulfqarhyder on February 12, 2011, 8:10 GMT

    a good work, stats may be proved wrong this time as team which do not speculates will be the winner as most of matches are fixed by the players and the result is altered. I think India and Austriala will be the victim of Q/F, pakistan, srilanka, SA and Newzealand will play semis, and India will host the Final of its neigbouring country with the chokers and both can choke at any time.

  • sufyan on February 12, 2011, 6:56 GMT

    excellent work. dont forget Pakistan, as pak is the only team who need only one push to change the whole momentum. i feel Pak will top in A group and will face India in final

  • Manjala Wijenayake-Angamm on February 12, 2011, 6:31 GMT

    Sri Lanka is the favorite to win the world cup. Our boys will show the colours and dominate the game. My mother is a very big politication in Galle, there is no matches in Galle. Very Boring. If not my political mother and us would have gone to see matches. Good Luck Sri Lanka. Show your colours. [[ Surprising that there is no match in Galle, a wonderful ground. It would have been a great tonic for the Tsunami-affected people. Maybe your mother should protest at this !!!~ Ananth: ]]

  • Sharif Faisal on February 12, 2011, 4:48 GMT

    If the order holds, QF match-ups are SL-BD, Aus-Eng, Pak-Saf, Ind-NZ. If favourites win their respective QFs, SL, Aus, Saf, and Ind make it to the semis. What are the match ups then, and the venues?? Not very clear from ICC fixture. Thanks. [[ SL-BD in Slk. Ind-Nzl in Ind and the other two in the respective assigned places. Ananth: ]]

  • shrikanthk on February 12, 2011, 4:32 GMT

    Ananth: It is a good thing that you are giving greater weight to recent form for a younger player than for a grizzled veteran. It makes perfect sense.

    This is how I understand it. The career average represents our fundamental valuation of the player. So, in an ideal world, any future performance is completely random...i.e it is a random walk around the current career average. Eg : Tendulkar's average in the World Cup = Career Average +/- dx where dx is totally random. Hence, the best forecast of Sachin's WC average is his career average itself! [[ I agree to lower weight of RF but not zero weight. That ignores reality. Dhoni has a RF RpI of 21.3 in the last 10 innings against his career RpI of 37.2. I agree class is permanent and form is temporary. However until Dhoni comes out of the slump, this has to be factored in. However in deference to his having played 265 matches this will have a lower weight. Ananth: ]]

    This theory may not hold for a young player like Yousuf Pathan. That's because his career average DOES NOT represent his fundamental valuation. Partly because he hasn't played in enough games for us to trust his average. Also partly because he still has scope to improve himself / or be found out as a player. Same thing holds for someone like Jonathan Trott. His average of 50 in ODIs and 60 in Tests is NOT completely reliable at this point.

  • Abhi on February 12, 2011, 4:11 GMT

    Srikanth, Your comment is simply a variant of the good ol' "Form is temporary , Class is permanent" theory. I think Ananth has got it right. If a CLassy player is out of form , for whatever reasons,- it must be taken into account. [[ Yes, I agree to lower weight of RF bot not zero weight. That is an ostrich-like stance. Ananth: ]]

  • Abhi on February 12, 2011, 4:07 GMT

    Ananth, Just couldn't resist! Which stocks do you own?! [[ Possible only when I can mail you. Both your mailids are bouncing. In general, the blue chips. Safe, but no outrageous increase in value Ananth: ]]

  • sabkababaone on February 12, 2011, 3:45 GMT

    as per my analysis india is a good side of the overall but keep in line that pakistan is now having dangerous bating side untill last the bowling always is extremly nice of this opposition the credit of asian ground is be add to this one side imagine so dangerous of all if they come on the angry way they such great player,s like Afridi,Razaq,Umar Akamal,Kamran Akmal,Asad Shafiq,Misbah,Younis,speed gum SHOAIB,

  • shrikanthk on February 12, 2011, 3:36 GMT

    Ofcourse, this is a very rough analogy.

    Stock prices are forward-looking (reflecting a consensus about future performance) unlike career averages which represent total historical information.

    But nevertheless, I thought the comparison was worth pondering about.

    Technical analysis cannot quite explain several careers. For instance, no expert of past trends could have predicted the resurgence of Warne and Lara post 2001 in Tests! Whereas a purist who bases his judgment on complete career performance and not just the past few matches wouldn't have been surprised by their strong comebacks! A "Technical analyst" from 1932-33 would've predicted that Bradman has finally settled down to a "normal" average after the Bodyline series and predicted a career average of 60 for him. But quite inexplicably, Bradman's average rocketed back to 100 rendering his recent Bodyline past irrelevant!

    Which is why I remain a purist and take "recent performance history" with a pinch of salt. [[ In general I agree with you. I am a completely random stock buyer and the buy-and-forget type of guy. Maybe the reason why the stocks have been kind to me over the years. I never analyze there. Just after the Bodyline series Bradman's "stock" would have come down to 60. However wfter two more series of 100+ it would have gone back to 100 since the 1934 and 1935 series wouyld have become recent form and Bodyline would have ben history. Here let us take the cases of Amla and Yousuf Pathan, The recent form of both has been phenomenal. They are expected to anchor Saf innings and provide a whirlwind finish to the Indian innings in the World Cups. Amla is more likely to do than Pathan. Hence the RF that they come into the tournament with, is relevant. How much. I have mentioned that for players who have played over 100 matches the RF has a lower weightage as compared to others who have played fewer matches. There are checks and balances. Finally these are theoretical exercises. What the players do on the field is what it matters. I can say that every single analysis/report/article on the WC, be it by Harsha, Gavaskar, Ian C, Sanjay M, Ananth, Rajesh, Omar S, Ayaz M, Roebuck et al can be confined to the dustbin. These are not like film reviews by established critics or stock buying recommendations by reputed analysts. There the audience decides whether to see the film/buy a stock or not based on the reviews/suggestions, okay at least 50%. Here it does not matter what is written. The crowd would go and make a noise for 9 hours. Ananth: ]]

  • shrikanthk on February 12, 2011, 3:27 GMT

    This analysis made me wonder about the "efficient market theory" that I learnt about in my MBA classes.

    Estimating future performances by using the recent past figures as a yardstick can be roughly likened to "Technical analysis", in stock analysis jargon.

    However the purists who believe in efficient markets would object to your analysis. For them, "Technical analysis" is all hogwash. All information available about a player is already reflected in his figures! Hence, studying the recent past can be of little use in predicting the future. A poor run leading up to a tournament may either signal an impending return to form or a permanent decline towards retirement. A purist ceases to speculate and instead bases his judgment solely on the average (i.e the stock price) which reflects the sum total of his knowledge about the player (stock).

    I am a purist myself who takes stock prices (i.e career averages) as a given and never speculate on future prices based on past trends!

  • CricStat on February 12, 2011, 2:43 GMT

    Hi Ananth,

    Thanks for compiling a very good and fact backed article. As its obvious from the article, there are few more results that are still to be unearthed from your side due to contract. For me, the semifinals would be among Ind Vs RSA and SL Vs AUS, and world cup final being played at wankhede which is newly laid stadium, advantage is with teams with more balanced attack. SL is frontrunner followed by Ind(provided zak is fit).

  • darklion on February 11, 2011, 23:52 GMT

    Your analysis is flawed in that some of the best players the game has ever seen are given very low ratings (Tendulkar, Murali, Ponting, Vettori, Kallis to name a few.) While I understand the need for consistency in how you assign ratings to players, this cannot be right (just by inspection this is false--these simply cannot be the weakest players on their respective teams!) Unless I am interpreting the numbers on the player sheets wrongly? [[ This is priceless and I could die laughing. The numbers are the Player-Ids in my Database. The presence of the number means that the concerned player is in the eleven. The numbers are also chronological. Hence Tendulkar has got the lowerst. Ananth: ]]

  • ibbi on February 11, 2011, 22:34 GMT

    Im sorry but based on your analysis, favorites are the most probable to win.. which I find to be totally against the reality as I know... India's 1983, Pakistan's 1992 and 2009 are examples of it... Truth is that events in life are random, there is a Black Swan right around the corner, waiting to ambush you... and the best part, you don't even know it exists... So, hold your horses, sit back and watch... You will be puzzled and perplexed as the events begin to unfold [[ H great. First time Black Swans have come into the comments. Wonderful. Recommend Nassim Taleb's epochal book to all readers. Frankly once I do the analyis I do not bother. I hope that my favourite team (India, no!!!) wins. Ananth: ]]

  • bks123 on February 11, 2011, 21:54 GMT

    If QF are hosted by teams having higher ICC ranking then it's really sad. Team having higher points in WC stage should host the match. If its BAN vs NZ then obviously BAN is going to host the QF irrespective of the points they have but if its BAN vs SL then the team having higher points must host the QF.

  • Milind on February 11, 2011, 20:46 GMT

    In 1992, co-hosts AUS & NZ played the round robin game in NZ with senior host AUS guaranteed to play semifinal at home. India gets the priviledge to play the knockout games at home, after playing the round robin game against the co-host away. Joint bids always use an understanding for the knock stage.

  • Deepak on February 11, 2011, 20:41 GMT

    i think u should do analysis on match winners of every team like 2 from every team example yusuf and yuvi-india, afridi and razak-pak, brovo and polard-wi, mathue and parera-lanka they all r going to Decide atleast 3-4 matches which is very big thing this article was very good nice to see india as favorite.

  • mub on February 11, 2011, 20:35 GMT

    I somehow think that it's not as if you need 3 good days. A bad day for the opposition can do the trick most often.

    If you take the '99 final, would you say that Australia had a good day? Or Pakistan had a bad day?

    That's why I'd feel any team can win after the quarters! [[ The teams that win are the ones which make their good days and not wait for the other team to have a bad day. The Australian bowlers should get the credit for dismissing Pak for 132. Similarly 4 years later the Australian batsmen should get credit. And why take credit away from Gilchrist for a match-smasher. It is a credit to Sri Lanka that they still came back reasonably well unlike their two other Asian compatriots. Ananth: ]]

  • Monish Menon on February 11, 2011, 19:52 GMT

    Another great article... and this time i made sure i read ur article from home and not work :P..... [[ [[ Thanks. Ananth: ]]

    Ananth: ]]

    anyways my semifinalists would be India, Aussies, Sri Lanka and Bangladesh....

    Yea i personally feel Bangladesh are stronger than the Windies and one good day in the quarters (or an ordinary day against New Zealand) can fetch them a spot in the semis!

  • Iftikhar Ahmed on February 11, 2011, 18:29 GMT

    Fantastic and interesting analysis. I am expecting India-Pakistan, Australia-Sri Lanka Semifinals.

  • deepak on February 11, 2011, 17:52 GMT

    I THK ANY TEAM'S STRENGTH SHD BE SUBTRACTED FROM IT'S WEAKEST PT WHICH IN INDIA'S CASE IS THE BOWLING ...ONE NEEDS TO RATIONALISE ON WHETHER THE BATTING IS GOOD ENF TO COMPENSATE FOR IT'S POOR BOWLING.THATS WHY DESPITE BOASTING OF A SPLENDID BATTING LINE UP,FRANKLY MEDIOCRE RESOURCES SHD MAKE US LAND 4/5 IN THIS CUP TOO

  • Gaurav on February 11, 2011, 17:11 GMT

    Interesting analysis, though I see New Zealand as the dark horse. Ananth, what are some of your predictions for upsets during the league stage? Mine are as follows : Pakistan beat Sri Lanka - Afridi/Razzaq special England beat South Africa - KP special West Indies beat India - Gayle special [[ Not one of these results is an upset. Upset is if a team from the second division defeats a first division team and I can only see Banglaesh against any of the four teams and possibly Zimbabwe against New Zealand as candidates. Of course Ireland vs Banglaesh is another possibility. Ananth: ]]

  • Vikram on February 11, 2011, 17:01 GMT

    I wish there were mathematical ways to calculate the choke factor. [[ All teams, other than Australia and Sri Lanka, to a lesser extent have choked at some time or other in the last 12 years. Ananth: ]]

  • Anand on February 11, 2011, 16:58 GMT

    Ananth: Fantastic article again. Kudos for all the hard work you put in to get these reults and then make an analysis. As a cricket lover and a stats lover I get the best fo both when I read your articles.

    Can I safely say that based on your predictions or calculations, teams nos. 5 and 6 can be called "dark horses" (which is England and Pak, as you also mentioned in your response to someone)?

    Is it possible (again, not sure how computationally complex this is) to then rank the upsets possible in this worlc cup, e.g., intuitively, I would not rate the result between any of the top 4 teams (India, SL, Aus and SAF) as an "upset". Personally, I would be upset if INdia lose to any of the other three but I am talking about an "objective" upset here (a low rank beating a higher rank team and bigger the difference bigger the upset).

    Can a simulation be run to see how "exciting" the world cup can get? I am just curious :-) [[ Two problems. One is "exciting" is a subjective term. Two is the Simulation for the WC is in private domain. Ananth: ]]

  • cric_options on February 11, 2011, 16:56 GMT

    Ananth...continuing on the earlier discussion, like you suggested, a few days back I came up with a risk index of the top teams and their chances in the knock outs and this is how it looks: Have assigned 1 to low risk of loss, 2 to medium chances of loss and 3 to high chances of loss, then added and multiplied the columns. Quar Semis Finals Added Multiplied ind 1 3 2 6 6 pak 2 2 2 6 8 sa 3 2 2 7 12 aus 1 2 1 4 2 sl 1 2 2 5 4 eng 2 3 2 7 12 nz 3 3 3 9 27 bang 3 3 2 8 18 wi 3 3 2 8 18

    Based on this, we see Aus, SL, India, Pak have the better chance in knock outs. But I totally acknowledge and agree to the form superiority of SA and thus will put the last spot open to either SA or Pak. Combining your analysis and mine, would rank SL, India, Aus, SA/Pak. For Aus to matter, their fast bowling in my opinion has to click, so its Lee/Tait/Johnson/Bollingers cup to win. Thanks again. [[ Very Interesting and simple idea. I would have put Pak at 2/2/2 if a.they had the two bowlers and if they had been playing at home. Why I would have had them at 1+1+2. They are starting the tournament with a great handicap. I like your putting Aus-Final at 1. No doubt you have given weighjt to their 12 year history. Again if SA get through the first knock-out they will be a force to reckon with. Ananth: ]]

  • Haris on February 11, 2011, 16:53 GMT

    Interesting Article Ananth: I have a question though regarding quarterfinal rules. I understand you explained about how higher ranked asian team will host the game but what about the rule, is it still based on teams previous ranking i.e. australia ranked 1st in ICC table will play the 8th ranked team or it will be based on the points they score in the league stage? Thanks! [[ No there is no change to A1 vs B4, A2 vs B3, A3 vs B2 and A4 vs B1. It is only in determining the venue of matches will the ODI rankings come into play, that too between Aisn teams who are co-hosts. Ananth: ]]

  • cric_options on February 11, 2011, 16:46 GMT

    Once again great analysis Ananth. Was wondering if in your analysis you put in the rolling data, such that after each match played in the world cup, that match is included in the calculations and the last of the 10 discarded. Also to this criteria, would be interesting to add a coefficient of rivalry between two teams which play each other, based on historical data. Once the predictions include these two criteria, it would become important how the matches were scheduled in the WC, as in who played whom earlier and thus the confidence and momentum could be created based on the results. Most often this is what we have seen, Pak in 92', Aus in 99', Ind in 83'. This will balance out the statistical edge the consistent performers have with the dark horses. Regarding hosting a Ind-SL match, guess the best option would have been to play that in Dhaka, a very much possible scenario when there are 3 hosts. But its hard to expect BCCI to think of bias when there is so much money to be made [[ As the WC progresses I expect to feed in the actual results, WC player performances (at a much higher weight) and do the simulations. Once these are published in the appropriate site I have no problems in providing the link or publish extracts here. Ananth: ]]

  • Winner on February 11, 2011, 16:43 GMT

    Sun Feb 13 09:00 GMT | 14:30 local 04:00 EST | 03:00 CST | 01:00 PST India v Australia

    This match will tell us who will be the world cup favourite [[ I don't think either team will go with their A-team for this warm-up match. Ananth: ]]

  • Santhana Krishnan on February 11, 2011, 15:30 GMT

    Ananth, your point is well made. But, I think it's unlikely. Both India and Sri Lanka would be extremely disappointed, if they don't finish in the top two of the respective groups. [[ Your point is also well made but does not take away from the subtle pressure which has been brought in. Ananth: ]]

  • Abhi on February 11, 2011, 15:30 GMT

    Ananth, Sometimes I feel you work more for this analysis "hobby" of yours than I do full time.

    As someone so correctly remarked- what about the "X factor".

    I feel the "X factor" is "Pressure". I don't think India can handle it at home. As we saw in 1996 once Tendulkar got out the whole thing collapsed.

    It is all very well if you get off to a good start with the crowd roaring behind you. That is "Home support"....A poor start, with a deathly silent crowd on the other hand is more unnerving than supportive.

    That is why my pick is Australia (as usual). They have mastered the art of absorbing pressure with the trick of "diffusing" it among the whole team equally- not separately among individual players. With the Indian team you somehow sense that the pressure is still "quantised" among individual players...Big difference when the chips are down and the S... hits the fan.

    India on paper. Australia actually take the Cup. [[ If I say that very few wins separate the first and fourth, then the X-factors can always play an important fact. Ananth: ]]

  • Syed on February 11, 2011, 14:57 GMT

    Hi Ananth,

    A very good article. I would like to say that Bangladesh, even though surpassed WI's in rankings, would not be able to compete with them despite the home ground advantage. Its a very formidable side, however WI's have some pretty good experienced campaigners who can perform better in sub-continent. I would go for WI's any day.

    Thanks. [[ On the day. Imagine Bangladesh scores 250. Look at how good they are at defending such scores at home with the quality spin attack they possess. For that matter this scenario exists in all their matches. Ananth: ]]

  • General_Public on February 11, 2011, 14:40 GMT

    Interesting analysis, just goes on to substantiate what everyone is saying - India are favorites really. Though frankly I am slightly hesitant about India's bowling pulling us through the three knockout games. And if I am not mistaken the host knockout matches being held in their venues was there in 1992 World Cup.

  • Santhana Krishnan on February 11, 2011, 14:18 GMT

    Though the idea of the quarters being played on the country with higher rank seems to have been introduced to favour India, logically there is nothing wrong in it. We don't want India v New Zealand quarter to take place in Sri Lanka [[ No the real problem is the India - Slk match. Imagine India gets 6 points and is fourth (guys don't jump on me). Slk has got 12 points and is first in A. Still the match would be played in India. Ananth: ]]

  • Rachit Gupta on February 11, 2011, 13:26 GMT

    Hi Anantha,

    In response to your comment to Gans... Why dont the ICC just hand over the cup to India in that case and release the public from this sham! Having said that I believe that chances of India winning the cup in 2011 this way are more :-) [[ And this is the first time such a rule has come in. Ananth: ]]

  • Pawan Mathur on February 11, 2011, 13:24 GMT

    I just want to ask you twoquestion. As per your calculations, are there are major possibility of a World Cup Upset and secondly who is the dark horse of this world cup [[ 1. Possibly Bangladesh qualifying. 2. The dark horse, I feel, is Pakiatan and to a lesser extent, England. Ananth: ]]

  • Shiraz Mahmood on February 11, 2011, 13:19 GMT

    You forgot one thing. The X factor. [[ I have not forgotten. That is for you guys to come out with. Ananth: ]]

  • Alex on February 11, 2011, 13:19 GMT

    Ananth - SL's has high bowling indices are surprising: Oz has a better bowling unit. Haven't seen all SL matches the last 2 years but Indians hammered their bowling home & away convincingly.

    The hype around Team India reminds me of '87 which was supposed to produce an Ind-Pak final but the two dark horses took them both out in semis. In this edition, Oz might still have the best shot if their #3 & #4 click big time. [[ India's scoring off Sri Lankan bowlers is not relevant. Please look at Sri Lankan bowlers (Murali, Mendis, Malinga, Mathews, Kulasekara). That will explain their high number. You could say Mendis would be hit but his career figures are still exceptional. The numbers are 4.3/4.5/3.9/4.6/4.4 (Adj RpO) and 42/50/55/31/46 (Adj BpW). Absolutely top-drawer figures. I personally think Australia should have waited on Hussey and replaced only Hauritz. Ananth: ]]

  • Ludd on February 11, 2011, 13:03 GMT

    Hi, Thanks for another great article. For what its worth I think your rankings table looks pretty accurate; I find it very hard to decide what order to put the top 4 in, but you too have them very close. My question is - does this mean that the statistics support me, or that you agree and I agree? You say that you decided the numbers after a lot of trial runs, so to what extent are you picking the figures to match your opinions? Is that clear? If the cold hard stats say that Kenya will win, you would say the numbers were wrong and rework them, but what if they said England would win, or Australia? Are the numbers 'right' when they give the right result? No criticism meant, just curioisty from a statistical incompetent. (and - whilst I'm being picky - in the second para below the group tables you say SAf play only one match away. For a moment I thought I'd missed something about this WC, but I think you mean Sri Lanka) [[ My hands are still tied. However you have raised a valid point in a nice manner and I will still try and answer it. I have simulated the WCs in units of 100, many many times. Each time I have run a correlation analysis of the team win table with the team strength table and finally selected one which had the highest correlation coefficient. Within the 100, Bangladesh could still win, probably once or maybe twice, out of 100. While the top four teams would win quite a few times. At no stage would I ever predict a single result. That would be stupid and allow my subjective bias to come through. Not here in the manner described. Thanks for pointing out the mistake. Will be corrected. Ananth: ]]

  • Dr.Limple on February 11, 2011, 12:51 GMT

    Its sad to see that you spent so much time on something that is utterly useless. Whichever team has a bit of momentum going into quarter final and semis, is going to win, and surely you would agree that sudden shifts of momentum are very hard to predict "statistically". So all of your "analysis" would carry little meaning once the world cup gets underway, because only then the clear favorites will emerge. [[ Only shows that you have not read the article completely. What you see is not even 5% of what I have done. Unfortunately I cannot help it. If you do not like predictions, would you accept these if they come from writers (ex-players and others) with very little analytical background and who talk about potential winners from the top of their heads. Ananth: ]]

  • Gohar on February 11, 2011, 12:09 GMT

    Hi,

    Brilliant analysis, well i would say advantage in the knock out stage would go to the Team batting second because of the Dew factor and playing under lights. So for a match to be won; captain needs to win the toss as well. Also, home advantage for India will only counts when they are playing against Pakistan, in other cases Pressure will be on Team India because of high expectations, so my suggestion to all Subcontinental team is that start your campaign as a Low as possible. My Semi finalist are India Australia South africa and Pakistan, I am feeling srilanka will face india in quarters!! [[ And India will have the advantage. Pl see the response to the porevious comment. Ananth: ]]

  • Gans on February 11, 2011, 12:01 GMT

    I am curious about the rules for Q/F.. is it going to be based on the ranking after the group matches or as per the seedings at the start of tournament?

    That would be a useful one to simulate who is likely to play at that stage and who is likely to be there in the S/Fs [[ This time ICC (probably prodded by BCCI) has made aure that India gets all assistance in coming through. Only recently I learnt the following. If any of the Asian teams do meet in the QF, the higher ranked team (in this case, India) will host the match. So India will host any QF match against the two Group A teams, Pakistan and Sri Lanka if such a match comes through. Of course if Sri Lanka and Bangladesh meet, the match will be held in Slk. Whatever happens, Pakistan will travel. It certainly helps to have clout !!! Ananth: ]]

  • QRK on February 11, 2011, 10:22 GMT

    Well Ananth,

    India and South Africa having good chances are about 90-95 % correct according to the numbers "ON PAPER", but if u go with the other four ie Australia, England, Pakistan and Srilanka, their performances have swung like a yo-yo of late, Australia might have won the series to englnad 6-1, however england were a tired lot, and they wld continue to be so in the world cup, Srilanka have won against westindies, who are not in the best in form, their bowling except for malinga and murli lacks bite, Pakistan, well they are the ultimate jigsaw puzzle since their entry into cricket, they are the ones who can PULL OUT DEFEAT from the jaws of opponents, but still they have a better bowling line up against the other three. but wait there;s something more to ruin their worldcup, THE SRILANKAN RAIN, it comes down at the slightest moment, and with pakistan playing all their matches in srilanka, their qualification also depends on the skies.

    over to all the other guys n ananth. [[ Yes, rain is another huge X factor. However while I could have built in rain into my simulation I did not do that since that is a very unpredictable thing. Let us hope that key teams in Group A are not hit badly by rain. Ananth: ]]

  • Santhana Krishnan on February 11, 2011, 9:44 GMT

    In spite of being number one and three time defending champions, most people are not backing Australia. I think they will be thought to beat simply because they are Australia. My ranking for this tournament 10. Zimbabwe 9.West Indies 8.New Zealand 7. Bangladesh 6. Pakistan 5. England 4. South Africa 3. Australia 2. India 1, Sri Lanka . I am putting Sri Lanka ahead of India because of their all-round bowling strength. The teams which I would prefer to win 1. India 2.South Africa 3. England 4.Bangladesh 5. New Zealand 6. West Indies 7.Sri Lanka 8. Pakistan 9.Zimbabwe 10. Please anyone but Australia

  • Amer Hussain on February 11, 2011, 9:44 GMT

    A wonderful analysis - and probably a very close to the mark prediction - thing is its based on statistics which I can understand, but players are always more than stats - there is emotion, passion, desire and unpredictability to take into account - and you have allowed for this - with your statement that 3 great days would win the cup. But doesn't this take away the beauty of the World Cup - would you have been able to predict that Pakistan would have won in '92 or Sri Lanka in '96 using a similar statistical approach. In my opinion this would have been unlikely and thats why the World Cup in the previous eras were better than this one will be or the last one was. By the way, I agree with your last statement - they should be going to win the Cup for themselves and their country!

  • Rachit Gupta on February 11, 2011, 9:41 GMT

    Hi Anantha,

    Brilliant analysis! But for me the real dark horses of this tourney are Zimbabwe. They have an excellent spin bowling attack and my guess is that based on current form NZ could be a Zimbabwean casualty. There is a very real chance of Zim winning 3 games in the round robin phase. I know you talk about their batting being shite but NZ can be equally bad on their day! Thats my two pence worth. Rest I enjoyed your analysis thoroughly and do hope that your predictions materialize. On a cautionary note however, stock markets have already taught us that past performance is not an indicator of future returns ;-) [[ The one difference between NZL and ZIM is that NZL has match-winners. However what you said could happen. In my simulation they have quailfiied reasonable number of times for the QF. But then, three great days, almost zero chance. Ananth: ]]

  • Chris Brydges on February 11, 2011, 9:32 GMT

    Hi Anantha,

    Great article, as always. I have one question though, about the recent form part of the analysis - was it calculated based on the last 10 ODIs that the players had played in, or was it the last 10 games of any kind? The example you give of Tendulkar, who hasn't played many ODIs in the past 2 years, suggests that taking innings/bowling figures from other forms of the game (e.g. test matches) that have been played more recently might make a difference. [[ That would be very bad. How can we compare the form shown in Tests with ODIs. That is the reason I had to go back to 2009. Ananth: ]]

  • No featured comments at the moment.

  • Chris Brydges on February 11, 2011, 9:32 GMT

    Hi Anantha,

    Great article, as always. I have one question though, about the recent form part of the analysis - was it calculated based on the last 10 ODIs that the players had played in, or was it the last 10 games of any kind? The example you give of Tendulkar, who hasn't played many ODIs in the past 2 years, suggests that taking innings/bowling figures from other forms of the game (e.g. test matches) that have been played more recently might make a difference. [[ That would be very bad. How can we compare the form shown in Tests with ODIs. That is the reason I had to go back to 2009. Ananth: ]]

  • Rachit Gupta on February 11, 2011, 9:41 GMT

    Hi Anantha,

    Brilliant analysis! But for me the real dark horses of this tourney are Zimbabwe. They have an excellent spin bowling attack and my guess is that based on current form NZ could be a Zimbabwean casualty. There is a very real chance of Zim winning 3 games in the round robin phase. I know you talk about their batting being shite but NZ can be equally bad on their day! Thats my two pence worth. Rest I enjoyed your analysis thoroughly and do hope that your predictions materialize. On a cautionary note however, stock markets have already taught us that past performance is not an indicator of future returns ;-) [[ The one difference between NZL and ZIM is that NZL has match-winners. However what you said could happen. In my simulation they have quailfiied reasonable number of times for the QF. But then, three great days, almost zero chance. Ananth: ]]

  • Amer Hussain on February 11, 2011, 9:44 GMT

    A wonderful analysis - and probably a very close to the mark prediction - thing is its based on statistics which I can understand, but players are always more than stats - there is emotion, passion, desire and unpredictability to take into account - and you have allowed for this - with your statement that 3 great days would win the cup. But doesn't this take away the beauty of the World Cup - would you have been able to predict that Pakistan would have won in '92 or Sri Lanka in '96 using a similar statistical approach. In my opinion this would have been unlikely and thats why the World Cup in the previous eras were better than this one will be or the last one was. By the way, I agree with your last statement - they should be going to win the Cup for themselves and their country!

  • Santhana Krishnan on February 11, 2011, 9:44 GMT

    In spite of being number one and three time defending champions, most people are not backing Australia. I think they will be thought to beat simply because they are Australia. My ranking for this tournament 10. Zimbabwe 9.West Indies 8.New Zealand 7. Bangladesh 6. Pakistan 5. England 4. South Africa 3. Australia 2. India 1, Sri Lanka . I am putting Sri Lanka ahead of India because of their all-round bowling strength. The teams which I would prefer to win 1. India 2.South Africa 3. England 4.Bangladesh 5. New Zealand 6. West Indies 7.Sri Lanka 8. Pakistan 9.Zimbabwe 10. Please anyone but Australia

  • QRK on February 11, 2011, 10:22 GMT

    Well Ananth,

    India and South Africa having good chances are about 90-95 % correct according to the numbers "ON PAPER", but if u go with the other four ie Australia, England, Pakistan and Srilanka, their performances have swung like a yo-yo of late, Australia might have won the series to englnad 6-1, however england were a tired lot, and they wld continue to be so in the world cup, Srilanka have won against westindies, who are not in the best in form, their bowling except for malinga and murli lacks bite, Pakistan, well they are the ultimate jigsaw puzzle since their entry into cricket, they are the ones who can PULL OUT DEFEAT from the jaws of opponents, but still they have a better bowling line up against the other three. but wait there;s something more to ruin their worldcup, THE SRILANKAN RAIN, it comes down at the slightest moment, and with pakistan playing all their matches in srilanka, their qualification also depends on the skies.

    over to all the other guys n ananth. [[ Yes, rain is another huge X factor. However while I could have built in rain into my simulation I did not do that since that is a very unpredictable thing. Let us hope that key teams in Group A are not hit badly by rain. Ananth: ]]

  • Gans on February 11, 2011, 12:01 GMT

    I am curious about the rules for Q/F.. is it going to be based on the ranking after the group matches or as per the seedings at the start of tournament?

    That would be a useful one to simulate who is likely to play at that stage and who is likely to be there in the S/Fs [[ This time ICC (probably prodded by BCCI) has made aure that India gets all assistance in coming through. Only recently I learnt the following. If any of the Asian teams do meet in the QF, the higher ranked team (in this case, India) will host the match. So India will host any QF match against the two Group A teams, Pakistan and Sri Lanka if such a match comes through. Of course if Sri Lanka and Bangladesh meet, the match will be held in Slk. Whatever happens, Pakistan will travel. It certainly helps to have clout !!! Ananth: ]]

  • Gohar on February 11, 2011, 12:09 GMT

    Hi,

    Brilliant analysis, well i would say advantage in the knock out stage would go to the Team batting second because of the Dew factor and playing under lights. So for a match to be won; captain needs to win the toss as well. Also, home advantage for India will only counts when they are playing against Pakistan, in other cases Pressure will be on Team India because of high expectations, so my suggestion to all Subcontinental team is that start your campaign as a Low as possible. My Semi finalist are India Australia South africa and Pakistan, I am feeling srilanka will face india in quarters!! [[ And India will have the advantage. Pl see the response to the porevious comment. Ananth: ]]

  • Dr.Limple on February 11, 2011, 12:51 GMT

    Its sad to see that you spent so much time on something that is utterly useless. Whichever team has a bit of momentum going into quarter final and semis, is going to win, and surely you would agree that sudden shifts of momentum are very hard to predict "statistically". So all of your "analysis" would carry little meaning once the world cup gets underway, because only then the clear favorites will emerge. [[ Only shows that you have not read the article completely. What you see is not even 5% of what I have done. Unfortunately I cannot help it. If you do not like predictions, would you accept these if they come from writers (ex-players and others) with very little analytical background and who talk about potential winners from the top of their heads. Ananth: ]]

  • Ludd on February 11, 2011, 13:03 GMT

    Hi, Thanks for another great article. For what its worth I think your rankings table looks pretty accurate; I find it very hard to decide what order to put the top 4 in, but you too have them very close. My question is - does this mean that the statistics support me, or that you agree and I agree? You say that you decided the numbers after a lot of trial runs, so to what extent are you picking the figures to match your opinions? Is that clear? If the cold hard stats say that Kenya will win, you would say the numbers were wrong and rework them, but what if they said England would win, or Australia? Are the numbers 'right' when they give the right result? No criticism meant, just curioisty from a statistical incompetent. (and - whilst I'm being picky - in the second para below the group tables you say SAf play only one match away. For a moment I thought I'd missed something about this WC, but I think you mean Sri Lanka) [[ My hands are still tied. However you have raised a valid point in a nice manner and I will still try and answer it. I have simulated the WCs in units of 100, many many times. Each time I have run a correlation analysis of the team win table with the team strength table and finally selected one which had the highest correlation coefficient. Within the 100, Bangladesh could still win, probably once or maybe twice, out of 100. While the top four teams would win quite a few times. At no stage would I ever predict a single result. That would be stupid and allow my subjective bias to come through. Not here in the manner described. Thanks for pointing out the mistake. Will be corrected. Ananth: ]]

  • Alex on February 11, 2011, 13:19 GMT

    Ananth - SL's has high bowling indices are surprising: Oz has a better bowling unit. Haven't seen all SL matches the last 2 years but Indians hammered their bowling home & away convincingly.

    The hype around Team India reminds me of '87 which was supposed to produce an Ind-Pak final but the two dark horses took them both out in semis. In this edition, Oz might still have the best shot if their #3 & #4 click big time. [[ India's scoring off Sri Lankan bowlers is not relevant. Please look at Sri Lankan bowlers (Murali, Mendis, Malinga, Mathews, Kulasekara). That will explain their high number. You could say Mendis would be hit but his career figures are still exceptional. The numbers are 4.3/4.5/3.9/4.6/4.4 (Adj RpO) and 42/50/55/31/46 (Adj BpW). Absolutely top-drawer figures. I personally think Australia should have waited on Hussey and replaced only Hauritz. Ananth: ]]