THE CORDON HOME

BLOGS ARCHIVES
SELECT BLOG
March 18, 2010

ODIs

Top ODI performers in each position: a quick follow-up

Anantha Narayanan

MS Dhoni has an excellent ODI batting index, which is next only to that of Viv Richards © AP
This is a follow-up to the article published a few days back. Alex had suggested that I do this based on the strike rates as the defining measure. I was not very comfortable with that since I think the batting average is a very important measure. Then Mareeswaran made the excellent suggestion that I use the combination of batting average and strike rate.

The ODI Batting Index (OBI), which is a product of batting average and strike rate, was used by me as part of television analysis during 2002/3. Afterwards it has undergone many transformations, Strike rate remaining common but multiplied by batting average, runs per innings and even extended batting average. However the original idea is still the best. The batting average is the most accepted of all measures.

First I am going to present the top-10 batsmen, based on OBI, based on their career figures. This has been given to let the readers have a perspective. An OBI of 50.00 has not been reached so far !!!

1   Hussey M.E.K         Aus  115   38   4136   53.71   88.4   47.46
2   Dhoni M.S            Ind  143   37   5420   51.13   89.9   45.95
3   Richards I.V.A       Win  167   24   6721   47.00   90.2   42.40
4   Zaheer Abbas         Pak   60    6   2572   47.63   84.8   40.39
5 ~ Bevan M.G            Aus  196   67   6912   53.58   74.2   39.74
6   Tendulkar S.R        Ind  431   41  17598   45.12   86.3   38.92
7   Pietersen K.P        Eng   88   15   3220   44.11   86.7   38.24
8   de Villiers A.B      Saf   92   13   3333   42.19   88.9   37.52
9 ~ Klusener L           Saf  137   50   3576   41.10   89.9   36.96
10   Symonds A            Aus  161   33   5088   39.75   92.4   36.75
As per request of some readers I have given also the OBIdx based on the eminently acceptable Runs per innings measure. This removes the anamolies of excessive not outs. However the main tables are still based on batting average since the not outs impact there is minimal. Position no.3 will always have lower number of not outs than no.7 and is applicable to all.
1   Zaheer Abbas         Pak   60   2572   42.87   84.8   36.35
2   Richards I.V.A       Win  167   6721   40.25   90.2   36.30
3   Tendulkar S.R        Ind  431  17598   40.83   86.3   35.22
4   Sehwag V             Ind  215   7091   32.98  103.5   34.14
5   Dhoni M.S            Ind  143   5420   37.90   89.9   34.06
6 ~ Gilchrist A.C        Aus  279   9619   34.48   96.9   33.42
7   de Villiers A.B      Saf   92   3333   36.23   88.9   32.22
8   Hussey M.E.K         Aus  115   4136   35.97   88.4   31.78
9   Pietersen K.P        Eng   88   3220   36.59   86.7   31.72
10 ~ Smith G.C            Saf  147   5613   38.18   83.1   31.73
First I worked out the all-match ODI Index for each batting position. In this case the OBI will be appropriate since the same methodology is used to determine the individual batsmen figures. Since the comparisons are across all batsmen at the same position the impact of not outs is minimised. The Strike rates are for that position. In order to ensure that flashes in the pan do not spoil the comparisons, a minimum limit of 1000 runs is set for Opening, no.3, no.4, no.5 and no.6 positions. For the position 7, the bar is set at 700 runs.

The OBI of the batsman in the relevant position is divided by the all-match OBI for that position and the ratio is arrived at. The tables are ordered on this ratio and the top-10 shown. Let us now look at the tables.

Analysis of Opening position

ODI Index for all matches: 22.50

No Batsman Cty Inns No Runs Avge Balls S/R OBI AllIdx-%

1.Dilshan T.M Slk 25 2 1263 54.91 1238 1.020 56.02 249.0% 2.Tendulkar S.R Ind 319 23 14482 48.93 16431 0.881 43.12 191.7% 3.Watson S.R Aus 48 5 1986 46.19 2264 0.877 40.51 180.1% 4.Sehwag V Ind 182 5 6336 35.80 6100 1.039 37.18 165.3% 5.Gilchrist A.C Aus 259 7 9200 36.51 9386 0.980 35.78 159.0% 6.Gayle C.H Win 193 14 7510 41.96 8901 0.844 35.40 157.3% 7.Turner G.M Nzl 29 5 1197 49.88 1688 0.709 35.37 157.2% 8.Hayden M.L Aus 147 14 5891 44.29 7486 0.787 34.86 154.9% 9.Lara B.C Win 52 5 2166 46.09 2871 0.754 34.77 154.5% 10.Smith G.C Saf 146 9 5598 40.86 6724 0.833 34.02 151.2%

Dilshan retains his position at the top. However, Tendulkar, with his excellent Strike rate has moved into the second position. The mountain of runs at an outstanding OBI of 43.12 is testament to the greatness of Tendulkar in this position. Any comment will be an understatement.

The pleasant surprise is the presence of two great attacking players, Sehwag and Gilchrist in the top 5. They had missed out in the earlier analysis.

Analysis of no. 3 position

ODI Index for all matches: 23.04

No Batsman Cty Inns No Runs Avge Balls S/R OBI AllIdx-%

1.Richards I.V.A Win 51 9 2418 57.57 2891 0.836 48.15 209.0% 2.Mohammad Yousuf Pak 43 7 1988 55.22 2521 0.789 43.55 189.0% 3.Ganguly S.C Ind 32 4 1476 52.71 1952 0.756 39.86 173.0% 4.Lara B.C Win 106 9 4447 45.85 5167 0.861 39.46 171.3% 5.Zaheer Abbas Pak 47 3 2009 45.66 2485 0.808 36.91 160.2% 6.Gambhir G Ind 31 4 1161 43.00 1403 0.828 35.58 154.4% 7.Ponting R.T Aus 305 30 11978 43.56 14779 0.810 35.30 153.2% 8.Kallis J.H Saf 176 29 6898 46.93 9455 0.730 34.23 148.6% 9.Chanderpaul S Win 25 2 1125 48.91 1635 0.688 33.66 146.1% 10.Hick G.A Eng 58 9 2182 44.53 2891 0.755 33.61 145.9%

In the pivotal position of no.3, there is no one to beat the great Viv Richards. His OBI is an amazing 48.15. The well-known no.3 batsmen, Md Yousuf, Ganguly and Lara follow. It is interesting to note that Ponting has scored nearly 12000 runs at an OBI of 35.30.

It may be of interest to note that Dhoni has scored 993 runs at an OBI of over 75.00.

Analysis of no. 4 position

ODI Index for all matches: 25.40

No Batsman Cty Inns No Runs Avge Balls S/R OBI AllIdx-%

1.Richards I.V.A Win 81 12 3373 48.88 3593 0.939 45.89 180.7% 2.de Villiers A.B Saf 42 8 1740 51.18 1967 0.885 45.27 178.2% 3.Sarwan R.R Win 43 12 1707 55.06 2172 0.786 43.28 170.4% 4.Bevan M.G Aus 53 15 2265 59.61 3232 0.701 41.77 164.5% 5.Jadeja A Ind 29 10 1008 53.05 1391 0.725 38.45 151.4% 6.Crowe M.D Nzl 53 14 1899 48.69 2436 0.780 37.96 149.4% 7.Boon D.C Aus 35 12 1255 54.57 1811 0.693 37.81 148.9% 8.Twose R.G Nzl 44 5 1829 46.90 2410 0.759 35.59 140.1% 9.Ranatunga A Slk 36 6 1272 42.40 1540 0.826 35.02 137.9% 10.Kallis J.H Saf 73 17 2635 47.05 3636 0.725 34.10 134.3%

The change has meant that Richards moves to the top position in this position instead of Bevan whose scoring rate is a pedestrian 0.7. Richards is the only batsman to have finished on top in two batting positions. de Villiers has shown his potential greatness by getting into the second position with a 45+ OBI. A surprise in this position is the high placement of Ajay Jadeja.
Analysis of no. 5 position

ODI Index for all matches: 22.77

No Batsman Cty Inns No Runs Avge Balls S/R OBI AllIdx-%

1.Dhoni M.S Ind 38 9 1560 53.79 1832 0.852 45.81 201.2% 2.Flintoff A Eng 48 10 1749 46.03 1854 0.943 43.42 190.7% 3.Hussey M.E.K Aus 25 6 1003 52.79 1221 0.821 43.36 190.4% 4.Symonds A Aus 96 18 3473 44.53 3780 0.919 40.91 179.7% 5.Yuvraj Singh Ind 81 13 2878 42.32 3268 0.881 37.27 163.7% 6.Collingwood P.D Eng 74 16 2621 45.19 3213 0.816 36.86 161.9% 7.Rhodes J.N Saf 90 23 2734 40.81 3302 0.828 33.79 148.4% 8.Cronje W.J Saf 43 7 1451 40.31 1745 0.832 33.51 147.2% 9.Dravid R Ind 69 13 2459 43.91 3341 0.736 32.32 141.9% 10.Inzamam-ul-Haq Pak 105 22 3473 41.84 4559 0.762 31.88 140.0%

This is Dhoni's position. He is now batting more and more at no.5. He again has a very high OBI of 45+. Flintoff jumps over Hussey into the second position. What a loss Flintoff's is to the game. Hussey just about gets in at a 43+ OBI. It is a surprise that Symonds, while scoring the same runs as Inzamam, has an OBI value of 40.91, which is about 8 more than Inzamam. Dravid retains his top-10 position with a respectable ODI of 32.32.
Analysis of no. 6 position

ODI Index for all matches: 19.91

No Batsman Cty Inns No Runs Avge Balls S/R OBI AllIdx-%

1.Raina S.K Ind 32 10 1087 49.41 1171 0.928 45.86 230.4% 2.Bevan M.G Aus 87 34 3006 56.72 3871 0.777 44.04 221.2% 3.Younis Khan Pak 28 5 1012 44.00 1108 0.913 40.19 201.8% 4.Hussey M.E.K Aus 51 14 1607 43.43 1811 0.887 38.54 193.6% 5.Arnold R.P Slk 59 21 1703 44.82 2273 0.749 33.58 168.6% 6.Cronje W.J Saf 45 16 1235 42.59 1567 0.788 33.56 168.6% 7.Dhoni M.S Ind 47 11 1395 38.75 1718 0.812 31.46 158.0% 8.Yuvraj Singh Ind 57 8 1727 35.24 2032 0.850 29.95 150.5% 9.Jadeja A Ind 43 8 1324 37.83 1743 0.760 28.73 144.3% 10.McMillan C.D Nzl 39 5 1058 31.12 1244 0.850 26.47 132.9%

The change in measure has allowed Suresh Raina, a faster scoring batsman, to jump over Michael Bevan, the finisher extraordinary. Both have very high OBI values of around 45.
Analysis of no. 7 position

ODI Index for all matches: 15.86

No Batsman Cty Inns No Runs Avge Balls S/R OBI AllIdx-%

1.Hussey M.E.K Aus 20 14 706 117.67 706 1.000 117.67 741.9% 2.Shahid Afridi Pak 40 9 718 23.16 527 1.362 31.56 199.0% 3.Abdul Razzaq Pak 79 21 1848 31.86 2076 0.890 28.36 178.8% 4.Hopes J.R Aus 38 6 896 28.00 911 0.984 27.54 173.6% 5.Pollock S.M Saf 81 26 1633 29.69 1836 0.889 26.41 166.5% 6.Boucher M.V Saf 44 16 846 30.21 991 0.854 25.79 162.6% 7.Chigumbura E Zim 38 6 995 31.09 1222 0.814 25.32 159.6% 8.Streak H.H Zim 40 12 864 30.86 1175 0.735 22.69 143.1% 9.Arnold R.P Slk 36 12 707 29.46 984 0.718 21.17 133.5% 10.O'Donnell S.P Aus 38 12 717 27.58 950 0.755 20.81 131.2%

Hussey has numbers which are beyond imagination. Granted he has scored only 700+ runs but what a finishing job he does. The next best is Shahid Afridi with 31.56. I am happy that Afridi is in this list because he is an outstanding talent. Before any negative comments are made on the high number of not outs, please do not forget that each not out instance indicates that the batsman has stayed on and finished his job, maybe not always successfully.

The candidates for the top-7 positions in an all-time ODI team, again my choice, are given below. Since this analysis incorporates the Strike rates it is possible to select a team. I have not just gone on the numbers.

Op Tendulkar S.R       Ind  319 23 14482  48.93 16431 0.881 43.12  191.7%
Op Gilchrist A.C       Aus  259  7  9200  36.51  9386 0.980 35.78  159.0%
3 Lara B.C            Win  106  9  4447  45.85  5167 0.861 39.46  171.3%
4 Richards I.V.A      Win   81 12  3373  48.88  3593 0.939 45.89  180.7%
5 Symonds A           Aus   96 18  3473  44.53  3780 0.919 40.91  179.7%
6 Hussey M.E.K        Aus   51 14  1607  43.43  1811 0.887 38.54  193.6%
7 Shahid Afridi       Pak   40  9   718  23.16   527 1.362 31.56  199.0%
Now add 4 top bowlers and we have a team the Gods would stop and watch. It is unfortunate Dhoni misses out but Gilchrist wins for many reasons, his numbers and the balance he brings by taking the opening positions. Hussey or Bevan is a tough call and a personal one.

Arjun Hemnany has done some additional work on the Not outs % by position. This is quite relevant to the discussions on Batting average vs Runs per innings. I have presented this table below.

% of Not outs out of all innings

Openers - 4.74 % no.3 - 7.84 % no.4 - 13.14 % no.5 - 16.09 % no.6 - 19.83 % no.7 - 23.69 % no.8 - 27.39 % no.9 - 33.56 % no.10 - 41.98 % no.11 - 59.74 %

At a later date I will come out with the Batting position analysis incorporating the figures above so that we would see a "normalized" Batting average figure. Many thanks to Arjun.

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

RSS Feeds: Anantha Narayanan

© ESPN Sports Media Ltd.

Posted by Abhi on (March 30, 2010, 6:23 GMT)

As an aside: Fascinating all time world xi out by colin croft. It's on the net : " Bradman, Headley out Crofty’s World XI " "

Posted by Abhi on (March 30, 2010, 2:35 GMT)

Nick, All sounds quite gruesome! Pardon my ignorance- but what exactly is the difference between “Runs/out” and the normal calculation of average (where Total runs are divided by Total innings minus N.O innings)?

Xolile, As Alex has so rightly said , it all depends on what spectacles you’ve got on. With your Pink specs- you are going to see everything in Pink. So,for fear of landing right smack bang back on square 1- we’ll let it pass.

Posted by Abhi on (March 30, 2010, 2:34 GMT)

Nick, All sounds quite gruesome! Pardon my ignorance- but what exactly is the difference between “Runs/out” and the normal calculation of average (where Total runs are divided by Total innings minus N.O innings)? Xolile, For fear of landing right smack bang back on square 1- we’ll let it pass.

Posted by pharmacy tech on (March 29, 2010, 19:05 GMT)

nice post. thanks.

Posted by Nick on (March 29, 2010, 14:14 GMT)

Just a quick word on the Runs/Out versus Runs/Innings debate over methods for calculating averages, and what is a fair reflection. There is strong mathematical evidence to suggest that Runs/Out is indeed best. I am currently studying applied statistics, and one topic we cover is to do with death rates. The case we study involves a number of patients, some who we observe to die at a certain age, and some who all we know is that they are still alive and at a certain age at the end of the observation period. It is known as Right-Censoring. If we make simple assumptions about the distribution of ages (namely an exponential distribution, which cricket innings do mirror very well) It turns out that the expected age of death is calculated by total life observed, divided only by number of deaths. in a cricket scenario, this translates to total runs scored divided by number of times out. I would gladly provide the relevant calculations if you wish to post it. [[ Nick That is very interesting. The only quick observation I can make is that death is final, consequently, it is only a matter of when not whether. There is no unbeaten life. However an innings could end undefeated. That is more philosophical than mathematical. But I like your comparison. I would like to have the observations. I will contact you directly so that you could mail me also directly. Thanks Ananth: ]]

Posted by Sarosh on (March 28, 2010, 3:02 GMT)

Considering the incredible firepower the Indian ODI batting lineup has had in the last couple of years it is scary to think of the totals they would have put up if they played mostly in New Zealand or such similar places. It just so happened that they played a lot in India.Once the kind of batsmen they possess fire, then no ground is safe.

Posted by boll on (March 27, 2010, 18:45 GMT)

I look back on some of the ridiculous hyperbole regarding Ponting/Lara/Sachin and who has been the greatest batsman and remember being reminded by an Indian mate of mine that they`d reached 10,000 test runs in 196/195/195 innings respectively. Argue about that until the cows come home. Fastest to 6000 runs was of course The Don, in 68 innings, no-one else under 110, and the 3 men in question all around 120. Now, that`s a stat that can`t be argued with.

Posted by boll on (March 27, 2010, 18:37 GMT)

I always appreciate that Anantha doesn`t attempt to draw spurious conclusions from these statistics. It`s refreshing to see them presented for what they are - a statistical representation of a player`s career. Too often, however, I think readers make claims based on these statistics that are simply false. People are often far to quick to laud a statistical analysis which agrees with their emotional perception of a cricketer, and either pass over, or accuse the author of bias, when it doesn`t. SRT fans (and if you love your cricket, who isn`t?) often seem to have issues with this. We all have our favourites, sometimes based on nationality, hopefully more often based on a style of cricket we`d love to emulate. When we do make claims, in this era of readily accessible statistics, we should however be able to support them, as I believe most writers on this site attempt to do.

Posted by Abhi on (March 27, 2010, 14:00 GMT)

Alex, Well put. In fact Ananth’s assertion that over a long career a top batsman will likely get a cheap run for every two tough ones is probably correct. Some arguments we hear are similar to ppl trying to devalue Federer’s achievements vis a vis Sampras. The argument now is that Sampras faced tougher opposition ! (Since Fed is piling ‘em on ,this seems to be the only argument remaining to the poor Sampras faithful!) What do you want Fed to do? Manufacture opponents?.He is bound to get some easy draws and opponents here and there over a long enough career. A lot of people went on about how he won the French since he didn’t have to play Nadal. He had already proven that he was the 2nd best clay court player in the world and he was due some “luck of the draw”.Poetic justice. Another thing ppl forget is the advantage lefties get over righties. The primary reason Nadal has (had?) the whip over Fed was because he is a leftie. If Fed too was a leftie (or vice versa) I don’t see Nadal winning a single match. This applies to cricket too- I rem. Reading an article by Frank Tyson on how he hated bowling to lefties because of the tremendous advantages they had( cant find it now)…So,that’s the way it goes…How can you “fault” someone for being a leftie! As you correctly mention - Depending on the spectacles you put on, you can always , in any circumstance, find a flaw in either an individual innings or a career. A scorecard reveals as much as it hides. It is extremely rare to find a flawless, perfect innings or career- as usual with the exception of the Don’s career.

Posted by Gavs on (March 27, 2010, 11:29 GMT)

@Xolile, considering career stats tends to naturally even out a lot of imbalances for players in the same era. For instance, India & SL play a lot of ODIs against each other on the sub-continent, but then Aus play a lot of ODIs against lowly-ranked WI and NZ home and away (and many WI, NZ grounds are are even smaller than in Asia). And South African batsmen too had their merry times in the '90s against weaker sides like India and SL (both are higher-ranked today), not to mention Zim.

That is why looking at career stats is a fair way to go as lots of things get evened out. Whereas you were simply looking at one year's stats in India and trashing the entire exercise. (Don't forget that the first 400+ game - actually 800+ game - came on African soil. And SA had already feasted on unexceptional Zim, India, SL, WI bowlers for countless 300+ totals in SA before that)

Comments have now been closed for this article

ABOUT THE AUTHOR

Anantha Narayanan
Anantha spent the first half of his four-decade working career with corporates like IBM, Shaw Wallace, NCR, Sime Darby and the Spinneys group in IT-related positions. In the second half, he has worked on cricket simulation, ratings, data mining, analysis and writing, amongst other things. He was the creator of the Wisden 100 lists, released in 2001. He has written for ESPNcricinfo and CastrolCricket, and worked extensively with Maruti Motors, Idea Cellular and Castrol on their performance ratings-related systems. He is an armchair connoisseur of most sports. His other passion is tennis, and he thinks Roger Federer is the greatest sportsman to have walked on earth.

All articles by this writer