Wednesday, May 30, 2012

Stas across eras 3 : Indian Batsmen - Tendulkar, Dravid, Gavaskar - a class apart

The eight part Statistics series by the author was published in Cricketcountry in April to May 2012 

In the first part of the series, we looked at how run making turned difficult or easier with different decades.

In the second episode, we analysed each decade of Test cricket and found that the most renowned batsmen always ended up at the peak among his contemporaries in terms of batting average.

Now, let us use the second finding and try to fit batsmen across eras into one scale. We start by evaluating the extraordinary riches of Indian batting.

Comparison between eras cannot be carried out through simple average analysis. Different eras did have different conditions, and different benchmarks, and the same average scores had different implications.

However, at the risk of repetition, “the best batsmen of a particular period always ended up with the best averages”. So, we will try and compare how different batsmen across time stood in comparison against their exact contemporaries.

To do this we first consider the playing days of a batsman – from the day he made his debut till his last Test.

Next, we find where he stood in terms of average among all the players during that exact period.

Three problems may arise due to this approach.

1. Too many lower order batsmen during an era can skew the figures.

2. A player who has played very few Tests can end up with a very high average.

3. In one period one batsman may rank say 4th among 30 men, and in another era a different batsman may rank 9th among 126. How do we compare them?

To circumvent the problems, we use the following counter approaches respectively.

1. We look only at batsmen who scored at more than the average runs scored during the period

2. We set a minimum criteria of 20 Tests for the modern (post 1950) era and 15 before that

3. We look at comparative index–among 100 exactly contemporary batsmen how many would the particular batsman be ahead of.

So, our algorithm is the following:

1. Select a batsman.

2. Find his first and last days in Test cricket.

3. Compute the average of all players during that period (global average).

4. Consider all players satisfying minimum Test criterion who averaged higher than global average during that period.

5. Find out the rank of the batsman among all the batsmen in (4).

6. Given the rank, find out how many batsmen he would be ahead of if the number of contemporary batsmen (4) had been exactly 100.

The positives of this calculation are that –

  • A player is only compared with his exact contemporaries, and this comparative index is computed to contrast players of different eras.
  • This eliminates the problems of fluctuations due to conditions, bowling quality etc.
For example,

i. Sachin Tendulkar has played between Nov 15, 1989 and Jan 28, 2012.

ii. The average of all batsmen (global average) during this period is 31.14.

iii. Between the dates in (i), there are 137 batsmen who played 20 or more Tests and scored above this average.

iv. Tendulkar’s average is 55.44, which ranks 3rd among the 137.

v. That gives him a comparative index figure of 99. For every 100 contemporary batsmen of his era, he would be ahead of 99.

For Sunil Gavaskar, the same computation yields 94. He is ahead of 94 among 100 contemporary batsmen who averaged more than the global average during his playing days.

Jacques Kallis and Kumar Sangakkara average higher than Tendulkar during his playing period, but therein is the sophistication of this approach. We cannot automatically conclude that Kallis and Sangakkara have a comparative index above Tendulkar, because it may very well happen, and indeed is, that during the period Sangakkara played Test matches (July 20, 2000 to April 7, 2012), Tendulkar averaged more than him. One is in staying clear of direct comparison of averages. Later in the series we will see how Kallis and Sangakkara score on comparative index with respect to their own exact contemporaries.

The table below shows some validation of what we have known for long. Tendulkar leads the charts, with a 99 comparative index. Dravid and Gavaskar end up in the 90s as well, after which there is a large gap.

NoNameAvgDebut TestLast TestGlobal avg

# Batsmen>   Global 

1SR Tendulkar55.4415.11.8928.1.12   31.14   137399
2R Dravid52.6320.6.9628.1.12   31.42   110894
3SM Gavaskar51.126.3.7117.3.87   30.43    78694
4M Azharuddin45.0331.12.846.3.00   29.83    952179
5V Sehwag50.913.11.0128.1.12   32.63    811978
6M Amarnath42.5024.12.6915.1.88   30.44    832373
7VVS Laxman45.9720.11.9628.1.12   31.43   1103470
8DB Vengsarkar42.1324.1.765.2.92   30.09    732665
9PR Umrigar42.229.12.4818.4.62   28.69    572164
10NS Sidhu42.1312.11.836.1.99   30.05    893660
11GR Viswanath41.9315.11.694.2.83   30.28    592460
12SC Ganguly42.1720.6.9610.11.08   30.94    954059
13G Gambhir45.263.11.0428.1.12    32.61    612658
14VL Manjrekar39.1230.12.522.3.65   29.44    472157
15VS Hazare47.6522.6.464.4.53   31.49    261352
16DN Sardesai39.231.12.6125.12.72   30.96    432738
17SM Patil36.9315.1.8017.12.84   29.9    352335
18SV Manjrekar37.1425.11.8723.11.96   30.03    604034
19RJ Shastri35.7921.2.8129.12.92   30.62    634726
20MS Dhoni37.322.12.0515.1.12   32.98    544125
21CG Borde35.5928.11.589.11.69   30.35    463524
22MAK Pataudi34.9112.12.6129.1.75   31.14    574521

*Criteria for minimum Tests played is 20 for all batsmen other than 15 for Hazare.

The only major surprises maybe finding Mohammad Azharuddin high up and Vijay Hazare quite low down in the ranking. Hazare may have suffered from playing in an era of Don Bradman, Len Hutton, Denis Compton, the 3 Ws, Neil Harvey and a roll call of great batsmen belonging to more established sides. Azharuddin’s ranking on the other hand is perhaps an indication that we often fail to recognise greatness when it is right in front of us.


i. Vijay Merchant, the other great Indian batsman, played only 10 Tests and has to be omitted because of low sample size.

ii. Home/Away, first innings/second innings etc. have not been considered here. However, this method can be refined for those specific analyses as well

No comments:

Post a Comment