venkyrenga
U19 12th Man
I have been working on a method extensively for the past few months to determine the top 30 test batsmen from the last 30 years as a fun exercise. I have also been reading the thoughts of members on here in similar threads and taken some input. The method I used is very different from the ones we have seen here before. I will explain the method first and then start posting the results. Let me know what you guys think.
The measures I used:
Batting average adjusted for not outs - I have adjusted the average by giving 0.75*avg for each not out. So it would be, (0.75*avg*not outs)+Runs/Innings. I have just deducted 1/4th of avg for each not out. This is done to reward batsmen who scored more runs with the same average.
Strike rate - I have taken 60 as the optimal SR with a 10% leeway. So it would be 54-66. (This range changes slightly for different periods. For example in the 90s it is 50-64.) And for every 4 away from that range 1% of points will be deducted. So, if the SR is 50 or 70 the batman will have 1% of his points deducted. For 46 or 74, it is 2% and so on. The idea is that a batsman with an SR in the optimal range is capable of accelerating or decelerating depending on the situation unlike batsmen with a very high or low SR. Since SR is not be a very important measure in test matches I have given very less weightage.
All average is adjusted for the era (every 4-year period).
Weightage for different categories:
Overall record 60%
Overseas record 20%
Peak record (Best 4 years) 10%
Record against best bowling attack 10%
To determine the best bowling attack, I have used runs conceded per wicket data and taken top 2 opponents or sometimes 3 if runs conceded is very close(difference of less than 2).
The main difference here is that I have not considered the entire career of the batsmen. For me it makes very less sense to do that when one batsman has an 18 year career and other has just 10 years or so. Some batsmen start very early or retire very late which results in a longer career with a depreciated average. Consider this, if batsmen A and B have parallel careers for 12 years and average 54 and 50 in that period. Now B retires but A continues to play a few more years or started a few years earlier hence ends up with lesser avg., say 49. Who is the better batsmen? I would say A. For me it is a no brainer. Better yet, if Bradman continued to play another 10 years and brought down his average to 60 would it change the fact that he was the best ever? Absolutely not.
To address this, I have considered the best 12-year period of a batsman's career. Because 12 years seems to be the mean career length of the top batsmen of the last 30 years. I have also not completely ignored their career outside that 12 years. If one continues a quality run beyond 12 years it will be included with a longevity bonus of 15% for twice the no. of years. For example if I am considering 16 years for a batsman he will get 5% additional points. And if a batsman has a career of less than 12 years, points will be deducted at the same rate. And 8 years is the minimum career length.
Points are calculated for every 4-year period and then averaged. But it has to be continuous 12 years, in other words three continuous 4-year periods. And all records against minnows are excluded completely. Records against the standard eight teams are considered throughout and minnows are determined based on runs conceded per wicket data for each 4-year period.
Note: Apart from calculating points for 4-year periods, points are also calculated for 1-year period in case it needed to be factored in after the 8 years or 12 years in consideration. Also, while calculating points for a given period a batsman with double the innings count(compared to the mean) will get additional 5% points. This is done to reward batsmen who maintained the same quality with higher quantity. Likewise, a batsman with lesser innings count will have points deducted.
The measures I used:
Batting average adjusted for not outs - I have adjusted the average by giving 0.75*avg for each not out. So it would be, (0.75*avg*not outs)+Runs/Innings. I have just deducted 1/4th of avg for each not out. This is done to reward batsmen who scored more runs with the same average.
Strike rate - I have taken 60 as the optimal SR with a 10% leeway. So it would be 54-66. (This range changes slightly for different periods. For example in the 90s it is 50-64.) And for every 4 away from that range 1% of points will be deducted. So, if the SR is 50 or 70 the batman will have 1% of his points deducted. For 46 or 74, it is 2% and so on. The idea is that a batsman with an SR in the optimal range is capable of accelerating or decelerating depending on the situation unlike batsmen with a very high or low SR. Since SR is not be a very important measure in test matches I have given very less weightage.
All average is adjusted for the era (every 4-year period).
Weightage for different categories:
Overall record 60%
Overseas record 20%
Peak record (Best 4 years) 10%
Record against best bowling attack 10%
To determine the best bowling attack, I have used runs conceded per wicket data and taken top 2 opponents or sometimes 3 if runs conceded is very close(difference of less than 2).
The main difference here is that I have not considered the entire career of the batsmen. For me it makes very less sense to do that when one batsman has an 18 year career and other has just 10 years or so. Some batsmen start very early or retire very late which results in a longer career with a depreciated average. Consider this, if batsmen A and B have parallel careers for 12 years and average 54 and 50 in that period. Now B retires but A continues to play a few more years or started a few years earlier hence ends up with lesser avg., say 49. Who is the better batsmen? I would say A. For me it is a no brainer. Better yet, if Bradman continued to play another 10 years and brought down his average to 60 would it change the fact that he was the best ever? Absolutely not.
To address this, I have considered the best 12-year period of a batsman's career. Because 12 years seems to be the mean career length of the top batsmen of the last 30 years. I have also not completely ignored their career outside that 12 years. If one continues a quality run beyond 12 years it will be included with a longevity bonus of 15% for twice the no. of years. For example if I am considering 16 years for a batsman he will get 5% additional points. And if a batsman has a career of less than 12 years, points will be deducted at the same rate. And 8 years is the minimum career length.
Points are calculated for every 4-year period and then averaged. But it has to be continuous 12 years, in other words three continuous 4-year periods. And all records against minnows are excluded completely. Records against the standard eight teams are considered throughout and minnows are determined based on runs conceded per wicket data for each 4-year period.
Note: Apart from calculating points for 4-year periods, points are also calculated for 1-year period in case it needed to be factored in after the 8 years or 12 years in consideration. Also, while calculating points for a given period a batsman with double the innings count(compared to the mean) will get additional 5% points. This is done to reward batsmen who maintained the same quality with higher quantity. Likewise, a batsman with lesser innings count will have points deducted.
Last edited:
