May 17, 2014

The fascinating topics of maidens and innings progressions in T20Is

An analysis of scoring rates across over groups and maidens in T20Is
19

Rangana Herath's 5 for 3 - which included two maidens - against New Zealand in the recent World T20 is one of the greatest T20I bowling efforts © Getty Images

I will pose 11 questions as an introduction to this article that covers two fascinating areas of T20 matches, viz Over-groups and Maiden overs.

Which two bowlers bowled maidens to Chris Gayle?
What is the highest score in the first six overs?
Which bowlers have bowled the maximum number of maidens in their career?
Which match featured the quietest last-five overs in a completed T20 innings?
Who played out 24 dot balls spread across six maiden overs?
What has been the most even match?
Maidens to Brendon McCullum and Quinton de Kock: Are you kidding?
How many maidens have been bowled in the overs 18-20?
Which is the most topsy-turvy of all T20 matches?
Has any bowler bowled three maiden overs in his spell?
Has the middle-over group been as low-key as normally perceived?

By now the readers will have a clear idea of what I expect to cover in this article. So without much ado we will move on to the article. You will find answers for these and other equally intriguing questions in the article, which is both analytical and anecdotal. This is one of the most interesting articles I have done because of the varied types of insights that are being offered.

First, a brief comment on the innings exclusion criteria.

T20 Internationals #26 & #68 were abandoned (four innings).
There is no ball-by-ball data for #9 & #335 (four inns)
Since this is an article on Over groups, I have excluded all innings in which fewer than 30 balls were bowled. This is anyhow the minimum number of balls in an interrupted match to constitute a result.
#318: Both innings excluded (two innings).
#119, #160, #260, #273 & #290: Second innings excluded (five innings).
Thus a total 15 innings have been excluded. This leaves us with 785 innings.

A. Innings over group analysis

The over groups do exist, although there is no formal separation. Overs No. 1 to No. 6 form the first group. Let us call this the "Start over group". The second group can be defined as the overs No. 7 to No. 15. This can be called the "Consolidation over group". The last five overs form the "Finish over group". Some people might argue for 15-20 instead of 16-20, but I have observed that the real finish is normally planned for five overs, rather than six.

First, let me present a simple table that displays all relevant data relating to the three over groups. In the table, the over group is the unit of measure, so that we are able to compare the results directly. In the same table, the over-group data is taken to the next level, i.e. overs, so that a different perspective can be obtained.

I would like readers to recall my first article on T20, published a few weeks back. In that article I had worked on the same over-group concept. I had mentioned that one-third of the total ball resource is normally utilised in each of the three over-groups. At that time I did not have the ball-by-ball data and this was a reasonable assumption. I was projecting that the final score would be around three times the score at the end of sixth over and 150% of the score at the end of the 15th over. In this analysis let us see whether that postulate can be verified and how far off I was. I am quite sure that my estimates that the scoring rate would drop off drastically during the overs 7 and 15 will take a big hit.

A1. Summary of Innings: by Over Groups
Og-Desc Runs Wickets DotBalls OverGroup Overs
Number Runs/Og Wkts/Og DBs/Og Number Runs/Ov Wkts/Ov DBs/Ov
1: Overs 1-634437129113937784.743.91.6517.847077.320.27 3.0
2: Overs 7-1548327197414403756.463.92.6119.068057.100.29 2.1
3: Overs 16-20280991743 5727634.944.32.75 9.031768.850.55 1.8

First, a brief explanation of the number of over groups. The 785 has already been explained. Two of these over groups had five overs and the exact total works to 784.67. I went to this level of accuracy so that the averages will come out accurately. The average score in the first over group is 44 for 1.6 (I suggest that the readers get used to looking at wickets in decimals). This should open a few eyes. Nowadays if we see a score of 45 for 1, we think that the batting team is lagging behind. That does not seem to be the case. They are ahead on both measures. The key resource up to over No. 15 is the number of wickets that have fallen. It is possible that 42 for no loss is better than 50 for 1 which is better than 60 for 2. Almost half of the balls bowled in this over group are dot balls.

The average score in the second over group is 64 for 2.6. Certainly I am in for a surprise here. This is not necessarily the consolidation phase I had portrayed it to be. The scoring rate is almost the same and the number of wickets fallen has increased significantly. It is probably correct to say that scoring at six per over during the middle overs will certainly set the team back barring a peculiar situation of 80 for 2 in 6, to start with. What is the average score at the end of the 15th over? It is 108 for 4.3. It is certainly below our current perception of a good score. Teams are expected to be around 120 for 3/4 at this point. The dot balls drop slightly to just over a third.

The third phase is a tough phase to analyse. Many innings end during this phase. I have taken the trouble to get this done accurately. An average score of 44 for 2.7 leads to an average final score of 152 for 7. This seems fine and is acceptable because it also includes unsuccessful second-innings chases.

The average Runs-per-Over (RpO) for the three phases is 7.32, 7.10 and 8.85 respectively. This means my perception of the middle over group is off the mark. There is only 3% drop from the first over group. So I would say that a 30-40-30 split between the three over groups seems to be acceptable and realistic. Since the overs are split 30-45-25, this means only a slight lowering in the middle phase, to be made up in the third phase.

A2. Most even matches

In match No. 163, Pakistan, playing against England, scored 44, 58 and 45 in the three over groups, leading to 147. This is the most even T20 match of all time. It is the closest to the 30-40-30 split.
In match No. 191, Pakistan, playing against South Africa, scored 35, 49 and 36 in the three over groups, leading to 120. This is nearly as even as the earlier Pakistan innings.
In match No. 7, England, playing against Sri Lanka, scored 49, 65 and 47 in the three over groups, leading to 161. This is very close to the second match.

I had earlier used a variation index based on the 33.3-33.3-33.3 split across the groups. Then I decided that I could as well implement the 30-40-30 split. The index sums the three absolute ratio values for the groups.

A3. Most topsy-turvy matches

In match No. 344, Kenya, playing against Netherlands, scored 10, 69 and 22 in the three over groups, leading to 101. There has never been a match with such wild changes. The amazing change is in the second over group.
In match No. 73, Zimbabwe, playing against Pakistan, scored 54, 43 and 10 in the three over groups, leading to 107.
In match No. 322, West Indies, playing against Pakistan, scored 13, 56 and 25 in the three over groups, leading to 124. Somewhat similar to the first match.

A4. Start over-group (1-6)

Mat Year Bat    Bow R   Score   Final Score
377 2014 HOL vs ire W 91 for 1 193 for 4 91 2009 NZL vs sco W 90 for 3 90 for 3 147 2010 AUS vs win W 83 for 0 142 for 2 ... 75 2008 CAN vs zim 10 for 3 75 for 10 344 2013 KEN vs hol 10 for 5 101 for 9 64 2008 IRE vs ken W 13 for 2 72 for 6

The Netherlands blitz is of recent vintage. They had to deliver, in spades, on that day, against a very good team, and they delivered, and how! Surely one of the greatest batting displays ever. New Zealand's performance is less impressive. It was a seven-over rain-shortened match and they won with an over to spare. Australia's effort was a low total chase, done with disdain.

The low scores have been scored by the Associates. Canada and Kenya lost. Ireland made a heavy weather of a simple chase of 67 and took over 19 overs to score 72. But they won.

A6. Consolidation over-group (7-15)

Mat Year Bat    Bow R   Score   Final Score
27 2007 SLK vs ken W 131 for 2 260 for 6 125 2009 SAF vs eng W 126 for 2 241 for 6 328 2013 AUS vs eng W 122 for 1 248 for 6 ... 330 2013 KEN vs afg 21 for 4 56 for 10 131 2010 BNG vs nzl 25 for 6 78 for 10 19 2007 KEN vs pak 27 for 5 92 for 10

Sri Lanka's demolition job of Kenya in the middle overs, to the tune of 131 for 2, took them to a record score of 260. The next-two highest totals have been against England's bowling attacks. All these led to scores over 240, and comfortable wins.

Surprisingly, the lowest totals in the middle overs have not followed the three lowest totals in the first over group. Kenya followed a miserable 21 for 5 with an equally miserable 21 for 4. Bangladesh scored a respectable 39 for 2 in the first group and then collapsed to 64 for 8, scoring a forgettable 25 for 6 in the middle 9 overs. Kenya scored 34 for 3 and then scored 27 for 5 in the middle overs.

A7. Finish over group (16-20)

Mat Year Bat    Bow R   Score   Final Score
268 2012 ENG vs afg W 87 for 3 196 for 5 94 2009 SAF vs sco W 83 for 2 211 for 5 397 2014 WIN vs pak W 82 for 1 166 for 6 ... 73 2008 ZIM vs pak 10 for 5 107 for 8 67 2008 BER vs can 17 for 3 70 for 10 332 2013 AFG vs ken 19 for 3 115 for 8

Let me say that I have only looked at teams that played out 20 overs for the third over-group computations. Zimbabwe were reasonably placed at 97 for 3 at the end of 15th over, and then lost five wickets for ten runs. Bermuda, at 53 for 7 in 15, slid, predictably, to 70 all out. Afghanistan, in a reasonable position at 96 for 5, did badly in the last five overs, scoring 19 for 3.

B. Maiden Over analysis

First let me reiterate that the definition of dot balls and maidens are as defined in my previous article. A dot ball is one in which no run is added to the opposing team. A maiden comprises of six such dot balls. There are 152 maidens as per this definition. As per the traditional definition, there are 193 maidens. This means that there is a maiden once every five innings or so. So I can say that an "Ananth-maiden" is rarer than a "ESPNcricinfo-maiden"!!!.

B1. Bowlers who have bowled three or more maidens in their career

Only ten bowlers have bowled three or more maidens in their career. Four of these have bowled four maidens. The list is given below.

Bowler         Ctry Mat Overs Mdns %Mdns  M/M
RW Price Zim 16 61.3 4 6.50 0.25 Harbhajan Singh Ind 24 90.0 4 4.44 0.17 KMDN Kulasekara Slk 40 139.1 4 2.87 0.10 BAW Mendis Slk 39 147.3 4 2.71 0.10 MR Gillespie Nzl 11 35.0 3 8.57 0.27 SJ Benn Win 17 59.0 3 5.08 0.18 Shapoor Zadran Afg 20 59.0 3 5.08 0.15 DT Johnston Ire 29 99.0 3 3.03 0.10 AD Mathews Slk 53 121.3 3 2.47 0.06 GP Swann Eng 39 135.0 3 2.22 0.08

The interesting fact is that six of these are spinners. Ray Price and Mark Gillespie are the outstanding bowlers in this lot of accurate bowlers, having bowled more than a quarter of their overs as maidens. Let us also raise our collective hats for Shapoor Zadran and Trent Johnston, the only bowlers from the Associate countries.

B2. Bowlers who have bowled 2 maidens in an innings

The astute reader might recall one of the questions asked at the beginning of the article. "Has any bowler bowled three maiden overs in his spell?". The answer is no. The highest number of maidens bowled in an innings is two. It is an exclusive list of eight bowlers and is given below.

Mat Year Player Name     For  Vs  O  M  R W
395 2014 HMRKB Herath Slk Nzl 3.3 2 3 5 150 2010 SJ Benn Win Zim 4.0 2 6 4 75 2008 RW Price Zim Can 4.0 2 6 2 263 2012 BAW Mendis Slk Zim 4.0 2 8 6 137 2010 Shapoor Zadran Afg Sco 4.0 2 8 1 334 2013 LL Tsotsobe Saf Pak 4.0 2 9 2 272 2012 Harbhajan Singh Ind Eng 4.0 2 12 4 268 2012 GP Swann Eng Afg 4.0 2 22 2

The interesting fact is that seven of these eight bowlers are spinners. Rangana Herath, Price, Graeme Swann and Sulieman Benn bowled two consecutive maiden overs. Considering the quality of opposition, I would put Herath's spell as the greatest of these outstanding efforts.

B3. Maximum maidens in an innings/match

The highest number of maidens bowled in an innings is three, achieved by six teams. The list is given below.

64 : Kenya       vs Ireland
134: Australia   vs Pakistan
150: West Indies vs Zimbabwe
268: England     vs Afghanistan
272: India       vs England
330: Afghanistan vs Kenya.

The highest number of maidens bowled in a match is four, achieved in three matches. The list is given below.

64 : Kenya-3       & Ireland-1
150: West Indies-3 & Zimbabwe-1
268: England-3     & Afghanistan-1.

B4. Perfect Maiden overs

By a perfect maiden over I mean one in which six balls were bowled by the bowler to the same batsman, i.e. not dismissing the batsman in the first five deliveries. There are 50 such instances. I have selected the interesting and out-of-the-way performances by bowlers. 248/1/3 means Match #248, Innings 1 and Over 3.

248/1/3  KD Mills        to CH Gayle
105/2/6  Harbhajan Singh to CH Gayle
165/1/1  Harbhajan Singh to SR Watson
365/1/1  MA Starc        to Q de Kock
 56/1/4  JM Anderson     to BB McCullum
161/2/18 SW Tait         to Abdur Razzak

Do we believe our eyes when we read these lines? Kyle Mills bowling six dot balls at a stretch to the marauder Gayle. And this happened at the unlikely location of Lauderhill, Florida, USA. It is another thing that Gayle still scored 53 in 39, took his team to 177 and they won by 61 runs.
This time it was the turn of Harbhajan Singh to bowl a bewildering maiden to Gayle in the sixth over. He managed to restrict Gayle to 22 in 28. But a Dwayne Bravo special took West Indies to a comfortable win.
Harbhajan is the only top-flight bowler to do this twice. The second time was to the equally devastating Shane Watson. This was the first over of the innings. Still, Watson scored 54 in 32 and Australia won comfortably by 49 runs.
The rampaging Quinton de Kock was kept quiet by Mitchell Starc in the first over of the innings. de Kock could only score 41 in 40 and South Africa lost quite badly to Australia. Very recent vintage, this match is.
Shaun Tait bowled six dot balls to Abdur Razzak in the 18th over. That helped Australia defend a low total of 141.
Shoaib Malik, was the main batsman in six maiden overs. Two overs, he faced completely (against Robin Petersen and Lonwabo Tsotsobe). And in the other four maidens, he faced 12 balls. (35/1/6, 134/2/6, 226/1/11, 378/1/9).

B5. Super maiden overs

I define a super maiden over as one in which a bowler dismissed three or more batsmen. Taking two or more wickets gives me more than 30 such overs, so I have to move the bar up to three. Some of these overs were covered in Part 1. I have repeated the descriptions for completeness of this article.

Mohammad Amir-156/1/20 vs Australia: I consider this to be the most incredible over in T20 history. I feel a move from 191 for 5 to 191 all out, during the course of a single over, gets the nod. I would always take the side of the bowlers anyhow. The opponents were Australia and established batsmen were at the crease. The sequence was Brad Haddin caught, Mitchell Johnson bowled, Michael Hussey run out, Steven Smith run out, Shaun Tait dot ball and Tait bowled. Five dismissals, three wickets to the bowler, six dot balls, this was Houdini at work.
Tim Southee-193/1/8 vs Pakistan: Younis Khan dot ball, Younis caught, Mohammad Hafeez caught, Umar Akmal lbw, Abdul Razzaq dot ball and Razzaq dot ball. A hat-trick to Southee was the icing on the cake.
Jerome Taylor-50/1/2 vs South Africa: Van Wyk lbw, JP Duminy bowled, AB de Villiers dot ball, de Villiers dot ball, de Villiers dot ball and de Villiers bowled. All three wickets to Taylor.
Ashok Dinda-255/2/18 vs Sri Lanka: Dinesh Chandimal dot ball, Chandimal caught, Shaminda Eranga dot ball, Eranga caught, Lasith Malinga dot ball and Malinga caught. Again, all three wickets to bowlers. A lovely symmetrical over: {. W . W . W}.

B6. Effective maiden overs

These are the maiden overs bowled at the business end of the innings, which means the overs 18-19-20. Five such maidens have been bowled till date.

Mohammad Amir-156/1/20 vs Australia: This over has already been covered.
Jeetan Patel-79/2/20 vs West Indies: Denesh Ramdin caught, Kieron Pollard dot ball, Pollard caught, Benn dot ball, dot ball and dot ball.
Wayne Parnell-323/2/19 vs Sri Lanka: Sachithra Senanayake dot ball, dot ball, dot ball, caught, Malinga dot ball and Malinga caught.
Tait-161/2/18 vs Pakistan: Perfect 18th over. Abdul Razzaq 6 dot balls.
Dinda-255/2/18 vs Sri Lanka: Already covered.

B7. Consecutive maiden overs

In match No. 150, Benn (Over 1), Kemar Roach (Over 2) and Benn (Over 3) bowled consecutive maiden overs against Zimbabwe. This is the only instance of three consecutive maiden overs and no run on the board at the end of 2/3 overs. Let us not forget that Zimbabwe also lost three wickets. Three wickets down for no score in 3.0: some start that was. They "recovered" to reach 105.
In match No. 64, Tony Suji (Over 15) and Jimmy Kamande (Over 16) bowled consecutive maiden overs against Ireland.

B8. Maiden overs: by bowling team

Australia:   17
Sri Lanka:   17
New Zealand: 15
India:       13
South Africa:13
Pakistan:    11
Ireland:     10
England:      9
West Indies:  9
Afghanistan:  8 

Australia and Sri Lanka have bowled the maximum number of maidens. Surprisingly, despite the presence of Saeed Ajmal, Umar Gul et al, Pakistan have only 11 maidens. I expected more.

B9. Maiden overs: by batting team

Pakistan:    27
Kenya:       16
West Indies: 13
England:     12
Afghanistan:  9 
New Zealand: 11
South Africa: 9
Australia:    7
Ireland:      6
Sri Lanka:    5
India:        3

Yes, I assure you guys that I cross-checked. Pakistan have indeed conceded 27 maidens; eight of these to South Africa. I get the feeling that when playing with Misbah-ul-Haq, Malik, Hafeez et al, even the strokemakers go strokeless. And these, in just 81 innings: once every three innings. Kenya have been quite strokeless. West Indies are also there at the top. India and Sri Lanka are brilliant, conceding very few maidens: India once every 17 innings and Sri Lanka, once every 12 innings.

I have created a veritable treasure house of information, this time in the form of a text file. Those who are interested can download and create the Excel sheet themselves. This contains the runs conceded, wickets captured and dot balls bowled by over-group for each innings. To download/view the text file, please CLICK HERE.

I am so confident of the coverage that I have not even checked whether I have answered all the 11 questions posed at the beginning of this article. I am sure I have done it. Please revert if I have missed something.

I have rectified the lacuna in my analyses, i.e. absence of coverage of T20 internationals, with a bang, with a series of three articles. Now I will revert to Test matches. My next article will feature a very intriguing topic: Test batsmen consistency. Let me thank Robert Eddings for providing the spark for this.

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

  • drinks.break on May 23, 2014, 7:45 GMT

    It's interesting that all three "most even" innings (A2) ended up on the losing side. Is this a case of mediocrity getting what it deserves?
    [[
    Yes, David, in this case it could be a question of not doing more in the third phase of the innings. There is probably no justification for those teams scoring fewer than 50 runs in the third phase.
    Ananth
    ]]

  • Snambidi on May 20, 2014, 2:25 GMT

    I have not an article on cricket as informative& accurate as this one from Mr.Anantha Narayanan. I am very much thankful to him for analyzing the T20 version so nicely that a few doubts I had in mind disturbing me since a long time.

    The best part of this article is the division of the 20 overs into " start overs",Consolidation overs"&"Finishing overs". Though the consolidation overs later on enables a team to do their best in Finish overs,some players deliver the goods by shining from Consolidation overs to finish overs. Such playersvare treat to watch. YuvarajSingh is one of them. Sir ,although I like the fast game like T20 ,somehow or other ,I do not incline to appreciate the way of Auction etc for it. Always Matches between Nations give the real pleasure to watch. Sir,I do not appreciate the third umpire & the latest technical method of D/W method. Th Umpires who were kings of the game have lost their identity now. Do you agree partially sir
    [[
    I am quite firm on the following.
    The T20 International game is clean and these are real contests. I am not comfortable with the other leagues. Too many attempts are made to scrap things under the carpet.
    I do not approve of the auctions which are like cattle markets. I think teams should make closed bids for players and allotment would be based on these.
    The D/L rule is fine for ODIs. But is very poorly implemented for T20s. The dynamics of the two formats are very different and this difference has not been factored in.
    Ananth
    ]]

  • Rufus_Fuddleduck on May 18, 2014, 3:12 GMT

    As usual, well conceived and well executed. And as usual some nitpicking since you specifically asked ;). You have given a superb hypothesis and then left it (and us souls too) hanging. You'd said that 40 for no loss as a start is better than 50 for one which is better than 60 for two. Now - would the stats throw up a list of anomalies in this hypothesis too? For keeping wickets in hand, there can indeed be too much of a good thing ... witness the England chase in WC final 1979 and some of India's efforts when Ravi Shastri got inspired the wrong way by Horatio. Against that you have the Tunbridge Wells match where the top order neither scored nor preserved. Talking 50-overs matches here is unfair, so I desist, but I hope you find the idea appealing enough to delve for a list of exceptional matches where the late charge either didn't materialise (40 for no loss and match still lost) or was delivered in style despite a flurry of wickets up top.
    [[
    Pankaj, I have partly covered this in my response to Arnab. My statements were general in nature and applicable to T20s only. ODIs with 250% of balls-resource available are totally different. T20 would keep on surprising us forever. Maxwell is rewriting the laws. 69 for 2 in 10 overs would be deemed below-par. 45 minutes later the score was 231. What analysis is possible?
    I promise you that I will do an analysis or two to either silently say "I told you so" or loudly say "I was probably wrong". My statements were off-the-cuff and were meant to emphasize the first over group average score of 43 for 2.
    Ananth
    ]]

  • kentjones on May 17, 2014, 10:30 GMT

    Anantha, this is really intriguing statistics.I am still coming to terms with some of this data.It will take me some time to really analyse all you have presented but some things have already stood out.!.The impact of the spinner on T20, spinners have certainly come to the fore in this format and they are an important weapon for every captain.Dividing the T20 innings into 3 phases, which is quite logical and then analysing each phase's impact is really great.Also the impact of the maiden on the T20 game is quite helpful.You can see that the top teams are the one who have utilized the maiden to full effect. Anantha, this analysis propels the discourse of the T20 game into theoretical focus.It is time to accept it as an evolution of cricket for the modern day fan who wants quick excitement capsuled in a couple hours. Analyzing in detail as you have done allows the strategist to plan and deliver the perfect game for both batting and bowling sides, and the fans of the future.Thanks Anantha
    [[
    Thank you. My personal take is that a maiden over is more important than a wicket and a late order maiden over is as important as a top order wicket. The change of 24 in 3 to 24 in 2, after a 18th over maiden, is amazing.
    Ananth
    ]]

  • drinks.break on May 23, 2014, 7:43 GMT

    Greatest ever sportsperson? It has to be Heather McKay. (As a professional squash player on the world tour she was undefeated for almost 20 years, and besides this was an Australian representative hockey player and a professional racquet-ball player who has been inducted into the USA racquet-ball hall of fame. Federer's still in achievement elementary school in comparison!)
    [[
    David, I know everything there is to know about Heather McKay from Queanbeyan, New South Wales. I have conducted extensive conversations with Australians in my previous blogspace in Cricinfo. I respect her achievements just as I respect Magic Johnson's, Bradman's, Pele's, Bolt's, Phelps' and so on. Having never seen a single smash of Heather, even on television, I would vote for Roger.
    Ananth
    ]]

  • on May 22, 2014, 6:30 GMT

    Hi Ananth

    Couldn't quite understand your comment" Ray Price and Mark Gillespie are the outstanding bowlers in this lot of accurate bowlers, having bowled more than a quarter of their overs as maidens." The percentage of maidens are 6.50 % and 8.57 % for Price and Gillespie respectively. Sri lanka and India have outstanding records when it comes to conceding maiden overs. Also from the Excel in your previous article I noted that India have been bowled out on only 3 occasions. Regards

    Santosh
    [[
    Santosh
    Again my wordings have gone awry. It should be a maiden in a quarter of their matches. I saw 4 maidens in 16 matches for Price and 3 maidens in 11 matches for Gillespie. The table from which I picked up this data had Mdns/Match as 0.25 and 0.27. That is what I wanted to highlight. Unfortunately it went wrong.
    My aplogies.
    Ananth
    ]]

  • MNA1974 on May 20, 2014, 0:54 GMT

    Your diehard fan from Lahore (Pakistan) sir. Your work always fascinates since I also have a lifelong interest in cricket statistics. I just wanted to know why you limited your definition of a maiden over. An analysis of 40 or so more overs would not have been that difficult, I assume. I mean that a wide or no ball are understandable, but if a bye or leg bye is scored in an over, the bowler does not need to be discredited for that. If allowed, I also want to share ideas for some future pieces with you. One of them that I have always wanted, though you would have to greatly extend the scope of your research for that, is on batsmen and bowlers who performed better in tests than in first class cricket. Best wishes from you with the hope that you will continue to amaze us with your wonderful work.
    [[
    I will repeat a statement from my previous article that I have not discredited the bowler at all. I am only saying that, if any run accrues to the batting team in any ball, do not call it a dot ball. I would not even have a problem if the current bowler analysis methodology is retained.
    I do not have access to FC records.
    Ananth
    ]]

  • shakeena on May 19, 2014, 20:06 GMT

    WOW .. I never knew those stuff.. It is like reading a fairy tale.. Just an one question dear sir.. Who do you think the greatest T20 bowler of all time considering both spin & fast bowling ? ? ?
    [[
    Very difficult to answer considering that many nuances have to be studied. I would say that mantle probably rests between Ajmal, Mendis, Steyn, Mlainga and Gul. Badree and Narine have not played enough matches.
    Ananth
    ]]

  • getsetgopk on May 19, 2014, 19:16 GMT

    Nobody will ever dispute Heraths 5 for 3 effort in t20's but Fedrer the greatest sportsman to have walked the earth? Did not see that one coming LOL, a very nice article as usual. Thanks.
    [[
    That is my view and I have everything I need to convince knowledgeable sports followers. But I will not do it in this forum. Someone questioned me on that. I in turn asked him who was his greatest sportsman ever. He said Maradona. I told him that I had all data to prove that Maradona was not even the greatest footballer who ever lived, leave alone sportsman. When I make that statement let me assure you that I have in my mind Pele, Jordan, Bradman, Bolt et al and of course Laver and Nadal.
    Ananth
    ]]

  • ww56 on May 19, 2014, 2:36 GMT

    [[
    You will have the ignominy of having your comment(s) published sans the words. I hope that, at least now, you will go off this blog, and, maybe go to the glades and watch crocodiles.
    Thanks.
    Ananth
    ]]

  • drinks.break on May 23, 2014, 7:45 GMT

    It's interesting that all three "most even" innings (A2) ended up on the losing side. Is this a case of mediocrity getting what it deserves?
    [[
    Yes, David, in this case it could be a question of not doing more in the third phase of the innings. There is probably no justification for those teams scoring fewer than 50 runs in the third phase.
    Ananth
    ]]

  • Snambidi on May 20, 2014, 2:25 GMT

    I have not an article on cricket as informative& accurate as this one from Mr.Anantha Narayanan. I am very much thankful to him for analyzing the T20 version so nicely that a few doubts I had in mind disturbing me since a long time.

    The best part of this article is the division of the 20 overs into " start overs",Consolidation overs"&"Finishing overs". Though the consolidation overs later on enables a team to do their best in Finish overs,some players deliver the goods by shining from Consolidation overs to finish overs. Such playersvare treat to watch. YuvarajSingh is one of them. Sir ,although I like the fast game like T20 ,somehow or other ,I do not incline to appreciate the way of Auction etc for it. Always Matches between Nations give the real pleasure to watch. Sir,I do not appreciate the third umpire & the latest technical method of D/W method. Th Umpires who were kings of the game have lost their identity now. Do you agree partially sir
    [[
    I am quite firm on the following.
    The T20 International game is clean and these are real contests. I am not comfortable with the other leagues. Too many attempts are made to scrap things under the carpet.
    I do not approve of the auctions which are like cattle markets. I think teams should make closed bids for players and allotment would be based on these.
    The D/L rule is fine for ODIs. But is very poorly implemented for T20s. The dynamics of the two formats are very different and this difference has not been factored in.
    Ananth
    ]]

  • Rufus_Fuddleduck on May 18, 2014, 3:12 GMT

    As usual, well conceived and well executed. And as usual some nitpicking since you specifically asked ;). You have given a superb hypothesis and then left it (and us souls too) hanging. You'd said that 40 for no loss as a start is better than 50 for one which is better than 60 for two. Now - would the stats throw up a list of anomalies in this hypothesis too? For keeping wickets in hand, there can indeed be too much of a good thing ... witness the England chase in WC final 1979 and some of India's efforts when Ravi Shastri got inspired the wrong way by Horatio. Against that you have the Tunbridge Wells match where the top order neither scored nor preserved. Talking 50-overs matches here is unfair, so I desist, but I hope you find the idea appealing enough to delve for a list of exceptional matches where the late charge either didn't materialise (40 for no loss and match still lost) or was delivered in style despite a flurry of wickets up top.
    [[
    Pankaj, I have partly covered this in my response to Arnab. My statements were general in nature and applicable to T20s only. ODIs with 250% of balls-resource available are totally different. T20 would keep on surprising us forever. Maxwell is rewriting the laws. 69 for 2 in 10 overs would be deemed below-par. 45 minutes later the score was 231. What analysis is possible?
    I promise you that I will do an analysis or two to either silently say "I told you so" or loudly say "I was probably wrong". My statements were off-the-cuff and were meant to emphasize the first over group average score of 43 for 2.
    Ananth
    ]]

  • kentjones on May 17, 2014, 10:30 GMT

    Anantha, this is really intriguing statistics.I am still coming to terms with some of this data.It will take me some time to really analyse all you have presented but some things have already stood out.!.The impact of the spinner on T20, spinners have certainly come to the fore in this format and they are an important weapon for every captain.Dividing the T20 innings into 3 phases, which is quite logical and then analysing each phase's impact is really great.Also the impact of the maiden on the T20 game is quite helpful.You can see that the top teams are the one who have utilized the maiden to full effect. Anantha, this analysis propels the discourse of the T20 game into theoretical focus.It is time to accept it as an evolution of cricket for the modern day fan who wants quick excitement capsuled in a couple hours. Analyzing in detail as you have done allows the strategist to plan and deliver the perfect game for both batting and bowling sides, and the fans of the future.Thanks Anantha
    [[
    Thank you. My personal take is that a maiden over is more important than a wicket and a late order maiden over is as important as a top order wicket. The change of 24 in 3 to 24 in 2, after a 18th over maiden, is amazing.
    Ananth
    ]]

  • drinks.break on May 23, 2014, 7:43 GMT

    Greatest ever sportsperson? It has to be Heather McKay. (As a professional squash player on the world tour she was undefeated for almost 20 years, and besides this was an Australian representative hockey player and a professional racquet-ball player who has been inducted into the USA racquet-ball hall of fame. Federer's still in achievement elementary school in comparison!)
    [[
    David, I know everything there is to know about Heather McKay from Queanbeyan, New South Wales. I have conducted extensive conversations with Australians in my previous blogspace in Cricinfo. I respect her achievements just as I respect Magic Johnson's, Bradman's, Pele's, Bolt's, Phelps' and so on. Having never seen a single smash of Heather, even on television, I would vote for Roger.
    Ananth
    ]]

  • on May 22, 2014, 6:30 GMT

    Hi Ananth

    Couldn't quite understand your comment" Ray Price and Mark Gillespie are the outstanding bowlers in this lot of accurate bowlers, having bowled more than a quarter of their overs as maidens." The percentage of maidens are 6.50 % and 8.57 % for Price and Gillespie respectively. Sri lanka and India have outstanding records when it comes to conceding maiden overs. Also from the Excel in your previous article I noted that India have been bowled out on only 3 occasions. Regards

    Santosh
    [[
    Santosh
    Again my wordings have gone awry. It should be a maiden in a quarter of their matches. I saw 4 maidens in 16 matches for Price and 3 maidens in 11 matches for Gillespie. The table from which I picked up this data had Mdns/Match as 0.25 and 0.27. That is what I wanted to highlight. Unfortunately it went wrong.
    My aplogies.
    Ananth
    ]]

  • MNA1974 on May 20, 2014, 0:54 GMT

    Your diehard fan from Lahore (Pakistan) sir. Your work always fascinates since I also have a lifelong interest in cricket statistics. I just wanted to know why you limited your definition of a maiden over. An analysis of 40 or so more overs would not have been that difficult, I assume. I mean that a wide or no ball are understandable, but if a bye or leg bye is scored in an over, the bowler does not need to be discredited for that. If allowed, I also want to share ideas for some future pieces with you. One of them that I have always wanted, though you would have to greatly extend the scope of your research for that, is on batsmen and bowlers who performed better in tests than in first class cricket. Best wishes from you with the hope that you will continue to amaze us with your wonderful work.
    [[
    I will repeat a statement from my previous article that I have not discredited the bowler at all. I am only saying that, if any run accrues to the batting team in any ball, do not call it a dot ball. I would not even have a problem if the current bowler analysis methodology is retained.
    I do not have access to FC records.
    Ananth
    ]]

  • shakeena on May 19, 2014, 20:06 GMT

    WOW .. I never knew those stuff.. It is like reading a fairy tale.. Just an one question dear sir.. Who do you think the greatest T20 bowler of all time considering both spin & fast bowling ? ? ?
    [[
    Very difficult to answer considering that many nuances have to be studied. I would say that mantle probably rests between Ajmal, Mendis, Steyn, Mlainga and Gul. Badree and Narine have not played enough matches.
    Ananth
    ]]

  • getsetgopk on May 19, 2014, 19:16 GMT

    Nobody will ever dispute Heraths 5 for 3 effort in t20's but Fedrer the greatest sportsman to have walked the earth? Did not see that one coming LOL, a very nice article as usual. Thanks.
    [[
    That is my view and I have everything I need to convince knowledgeable sports followers. But I will not do it in this forum. Someone questioned me on that. I in turn asked him who was his greatest sportsman ever. He said Maradona. I told him that I had all data to prove that Maradona was not even the greatest footballer who ever lived, leave alone sportsman. When I make that statement let me assure you that I have in my mind Pele, Jordan, Bradman, Bolt et al and of course Laver and Nadal.
    Ananth
    ]]

  • ww56 on May 19, 2014, 2:36 GMT

    [[
    You will have the ignominy of having your comment(s) published sans the words. I hope that, at least now, you will go off this blog, and, maybe go to the glades and watch crocodiles.
    Thanks.
    Ananth
    ]]

  • on May 18, 2014, 5:33 GMT

    why no ipl?
    [[
    Why Ipl???
    Ananth
    ]]

  • on May 18, 2014, 0:56 GMT

    Thanks .... Nice Article ....

  • SLSup on May 17, 2014, 19:22 GMT

    I apologize for staying off-topic but I did take a peak on Idea Ranking - SL tops ALL THREE FORMATS. Under methodology used there's ref to various factors considered but don't list them. I am not a fan of giving extra points to a team/country on series wins. Games by themselves ought to be the primary factor cos Series wins tend to negate games lost in a series, just a feelgood deal.
    [[
    I consider matches only. But give more weight to important matches, especially in the limited over formats.
    Ananth
    ]]
    On topic, I cannot but help ref to PRAGMATISM: the ability to reason within context. A dot ball CLEARLY has more relevance in a shorter format, I recall a local 6-a-side tournament where a dot ball was murder! It's good to see so much focus on t20s though my interest in them is entirely on T20Is where I feel the QUALITY of cricket is so much better and the MEANING is more relevant. This is another post that everyone can benefit from, two great posts Anantha!

    My ONLY objection is referencing to overs 7-15 as a 'consolidation' period. Overs 1-15 seem to differ SOMEWHAT EVENLY from overs 16-20.
    [[
    "Moving ahead" period, perhaps.
    Ananth
    ]]

  • crikcrazy on May 17, 2014, 18:53 GMT

    Unless I missed something, there appears to be a minor calculation mistake in the following statement, "In match No. 322, West Indies, playing against Pakistan, scored 13, 56 and 25 in the three over groups, leading to 124." 13, 56 and 25 adds up to 94.
    [[
    Thanks. It should be 13, 56 and 55. I will ask Cricinfo to correct. And of interest is that the 13 was 13 for 3 (Charles, Gayle and Samuels). And the 55 included a 9-ball-17 by Best.
    Ananth
    ]]

  • its.rachit on May 17, 2014, 16:01 GMT

    Hehe ..you are welcome Ananth .... but just a humble request from my side .. I would love to be a part of it, contributiing anything if you would like that ... BTW I could not find the IDEA rating on the home page ... can u share the link pls ...
    [[
    http://www.espncricinfo.com/ci/content/zones/cricketrating
    Ananth
    ]]

  • on May 17, 2014, 14:06 GMT

    Dear Ananth, This is an off-topic comment, my apology for this. I am interested to know that what are the next say 5 or 10 upcoming topics in your To Do List? Also, with the retirement of some big names (in Tests), perhaps only Jayawardene, Chanderpaul, Sangakkara (for batting) & Harbhajan, Steyn, Vettori, Anderson (for bowling) are still playing as leading players. Sehwag, Harbhajan & Vettori's further test careers are doubtful and Clarke & Cook as batsman and Zaheer & Johnson as bowler are still way behind in run & wicket tally. In this scenario, would you repeat / revisit some of your earlier topics once again covering the complete career stats for recently retired all time greats? And when are you going to do your all encompassing Batsman & Bowler analysis for Tests & ODIs? Thanks, Arnab Mallick, Kolkata
    [[
    Yes, I agree, Arnab. The first step is the revisit of the Batsman Consistency, a topic I looked two years back. Now I have a totally different and possibly a lot more acceptable thought process. For Tests and Odis I have only part ball-data. that is where I see the T20s as providing me a lot more opportunities.
    I also expect to re-do the Test HSI concept incorporating the Sqrt measure and GM, as sugegsted by Milind.
    Ananth
    ]]

  • kentjones on May 17, 2014, 13:39 GMT

    Yes. A maiden is more important. At the final 17 to 20 over stage, which is 18 balls and a possible 50 to 60 runs, a maiden reduces the probability of such a score by 33 percent and thus to 30 to 40 runs instead. A prime wicket taken cam also impact, but depending on the state of the game, if for instance the team has wickets in hand. it may not be as critical. Anantha I am also pursuing the possibility of six or more dot balls in the final 3 three overs as possessing the same level of significance as a maiden over. What you think?
    [[
    I have only looked at the ball-by-ball data from the point of view of an over. Not yet from the batsman's and bowler's points of view. Should be fascinating. I had already started work on evaluating the T20 bowler's spells. There I had valued the dot balls based on match context. Now I have moved the analysis back for some time so that I can do some Test/Odi analysis.
    Ananth
    ]]

  • its.rachit on May 17, 2014, 11:43 GMT

    Great analysis as usual Ananth ... talking of the worst performance by a team in the Finish Overs, it has got to be India in this Year's T20 final .. even though Sri Lanka were fantastic in the over's, India were horrible ... got to be the worst finish to an innings ever ... if not statistically, but in an anti-climactic way atleast ... only Pakistan's 125 all out in the 99 WC final (the whole innings was a disaster) can compare to this ...
    [[
    Worst finish to an innings ever...You guys are amazing. Here is a topic for me to look at. As I write this mail I can visualize my doing a projection at the end of each ball, say after 15 overs and compare that with what was achieved. For a rainy day. Thanks.
    Ananth
    ]]

  • pradeep_dealwis on May 17, 2014, 10:07 GMT

    I think the fact that SL bowl the highest number of maidens and conceded the 2nd lowest shows a great amount of discipline. It also shows why they are the best T20 team in the world (forget the nonsensical ranking that puts India ahead) even without brutal and explosive batsman.
    [[
    Nothing is nonsensical. Sri Lanka and India have traded places at the top over the past month. And you should not forget that the first WC win for Sri Lanka came only two months back. But I can and have devised a much better Team Ranking system. If you look at the Idea Ratings, which is featured on the Cricinfo main page and is done by me, the following are the top-5. I give more weight to the important matches.
    1, 820.9, Sri Lanka
    2, 819.2, India
    3, 798.6, Pakistan
    4, 782.4, South Africa
    5, 730.9, England
    But it can be seen that Sri Lanka and India are quite close. After the WC win, they were separated a bit. But the dropping of some of the early matches reduced the gap.
    Ananth
    ]]

  • pradeep_dealwis on May 17, 2014, 10:07 GMT

    I think the fact that SL bowl the highest number of maidens and conceded the 2nd lowest shows a great amount of discipline. It also shows why they are the best T20 team in the world (forget the nonsensical ranking that puts India ahead) even without brutal and explosive batsman.
    [[
    Nothing is nonsensical. Sri Lanka and India have traded places at the top over the past month. And you should not forget that the first WC win for Sri Lanka came only two months back. But I can and have devised a much better Team Ranking system. If you look at the Idea Ratings, which is featured on the Cricinfo main page and is done by me, the following are the top-5. I give more weight to the important matches.
    1, 820.9, Sri Lanka
    2, 819.2, India
    3, 798.6, Pakistan
    4, 782.4, South Africa
    5, 730.9, England
    But it can be seen that Sri Lanka and India are quite close. After the WC win, they were separated a bit. But the dropping of some of the early matches reduced the gap.
    Ananth
    ]]

  • its.rachit on May 17, 2014, 11:43 GMT

    Great analysis as usual Ananth ... talking of the worst performance by a team in the Finish Overs, it has got to be India in this Year's T20 final .. even though Sri Lanka were fantastic in the over's, India were horrible ... got to be the worst finish to an innings ever ... if not statistically, but in an anti-climactic way atleast ... only Pakistan's 125 all out in the 99 WC final (the whole innings was a disaster) can compare to this ...
    [[
    Worst finish to an innings ever...You guys are amazing. Here is a topic for me to look at. As I write this mail I can visualize my doing a projection at the end of each ball, say after 15 overs and compare that with what was achieved. For a rainy day. Thanks.
    Ananth
    ]]

  • kentjones on May 17, 2014, 13:39 GMT

    Yes. A maiden is more important. At the final 17 to 20 over stage, which is 18 balls and a possible 50 to 60 runs, a maiden reduces the probability of such a score by 33 percent and thus to 30 to 40 runs instead. A prime wicket taken cam also impact, but depending on the state of the game, if for instance the team has wickets in hand. it may not be as critical. Anantha I am also pursuing the possibility of six or more dot balls in the final 3 three overs as possessing the same level of significance as a maiden over. What you think?
    [[
    I have only looked at the ball-by-ball data from the point of view of an over. Not yet from the batsman's and bowler's points of view. Should be fascinating. I had already started work on evaluating the T20 bowler's spells. There I had valued the dot balls based on match context. Now I have moved the analysis back for some time so that I can do some Test/Odi analysis.
    Ananth
    ]]

  • on May 17, 2014, 14:06 GMT

    Dear Ananth, This is an off-topic comment, my apology for this. I am interested to know that what are the next say 5 or 10 upcoming topics in your To Do List? Also, with the retirement of some big names (in Tests), perhaps only Jayawardene, Chanderpaul, Sangakkara (for batting) & Harbhajan, Steyn, Vettori, Anderson (for bowling) are still playing as leading players. Sehwag, Harbhajan & Vettori's further test careers are doubtful and Clarke & Cook as batsman and Zaheer & Johnson as bowler are still way behind in run & wicket tally. In this scenario, would you repeat / revisit some of your earlier topics once again covering the complete career stats for recently retired all time greats? And when are you going to do your all encompassing Batsman & Bowler analysis for Tests & ODIs? Thanks, Arnab Mallick, Kolkata
    [[
    Yes, I agree, Arnab. The first step is the revisit of the Batsman Consistency, a topic I looked two years back. Now I have a totally different and possibly a lot more acceptable thought process. For Tests and Odis I have only part ball-data. that is where I see the T20s as providing me a lot more opportunities.
    I also expect to re-do the Test HSI concept incorporating the Sqrt measure and GM, as sugegsted by Milind.
    Ananth
    ]]

  • its.rachit on May 17, 2014, 16:01 GMT

    Hehe ..you are welcome Ananth .... but just a humble request from my side .. I would love to be a part of it, contributiing anything if you would like that ... BTW I could not find the IDEA rating on the home page ... can u share the link pls ...
    [[
    http://www.espncricinfo.com/ci/content/zones/cricketrating
    Ananth
    ]]

  • crikcrazy on May 17, 2014, 18:53 GMT

    Unless I missed something, there appears to be a minor calculation mistake in the following statement, "In match No. 322, West Indies, playing against Pakistan, scored 13, 56 and 25 in the three over groups, leading to 124." 13, 56 and 25 adds up to 94.
    [[
    Thanks. It should be 13, 56 and 55. I will ask Cricinfo to correct. And of interest is that the 13 was 13 for 3 (Charles, Gayle and Samuels). And the 55 included a 9-ball-17 by Best.
    Ananth
    ]]

  • SLSup on May 17, 2014, 19:22 GMT

    I apologize for staying off-topic but I did take a peak on Idea Ranking - SL tops ALL THREE FORMATS. Under methodology used there's ref to various factors considered but don't list them. I am not a fan of giving extra points to a team/country on series wins. Games by themselves ought to be the primary factor cos Series wins tend to negate games lost in a series, just a feelgood deal.
    [[
    I consider matches only. But give more weight to important matches, especially in the limited over formats.
    Ananth
    ]]
    On topic, I cannot but help ref to PRAGMATISM: the ability to reason within context. A dot ball CLEARLY has more relevance in a shorter format, I recall a local 6-a-side tournament where a dot ball was murder! It's good to see so much focus on t20s though my interest in them is entirely on T20Is where I feel the QUALITY of cricket is so much better and the MEANING is more relevant. This is another post that everyone can benefit from, two great posts Anantha!

    My ONLY objection is referencing to overs 7-15 as a 'consolidation' period. Overs 1-15 seem to differ SOMEWHAT EVENLY from overs 16-20.
    [[
    "Moving ahead" period, perhaps.
    Ananth
    ]]

  • on May 18, 2014, 0:56 GMT

    Thanks .... Nice Article ....

  • on May 18, 2014, 5:33 GMT

    why no ipl?
    [[
    Why Ipl???
    Ananth
    ]]

  • ww56 on May 19, 2014, 2:36 GMT

    [[
    You will have the ignominy of having your comment(s) published sans the words. I hope that, at least now, you will go off this blog, and, maybe go to the glades and watch crocodiles.
    Thanks.
    Ananth
    ]]