Then Arjun Hemnani came out with a suggestion that I classify all bowling attacks into four groups and develop batting tables based on these groups. It seemed to be an excellent idea and I have created this article based on this idea. This is a quasi-rating work based on the most important of parameters, viz., the Bowling quality. I may do a similar exercise based on the Pitch conditions. Again some really tough work but at the end worthwhile.
I have summarized all relevant facts related to this analysis. First let me emphasize that this is not a Test innings Ratings analysis. There are many other relevant factors which would have to be considered in such an analysis. I have not done so in this analysis which is centred on Bowler quality. I would appreciate if the readers do not keep on repeating again and again that other relevant factors such as Pitch type, Innings status at entry, Result, Series status, Bowler recent form, Innings target et al, have not been included. That would be counter-productive.
1. As I have done in the Team strength calculations, I have considered only Tests played after 1900. It is impossible to fit in the Tests before 1900 because of uncovered pitches and many sub-20 averages. However we lose only 64 Tests.
2. The Bowling quality index (BQI) is based on Career-to-date values. This is the most dependable and accurate of the bowling measures. There is no situation where the Career-to-date figure is not the appropriate one. A bowler like Lee with a great start and tapering off towards the end or Muralitharan who had a poor start and wonderful finish will be taken care of equally well.
3. The BQI is based on the actual bowlers who bowled in the particular innings. This is very important. There is a Sri Lankan innings in which Wasim Akram and Waqar Younis bowled 14 overs each. That is all. This would have been a terrifying situation for the batsmen. Contrast this with an innings in which Akram and Younis bowled 50% of the overs.
4. The BQI is determined for each innings. However in order to reduce wild variations, I will apply the BQI of the first innings to the second innings also in case the number of balls bowled in the second innings is less than 60. This is commonsense. This is explained through an example. Readers should know that this would not have much of an impact since no batsman is likely to score even 25 within the 10 overs.
Test # 1962: Win 231, Saf 346 (Win BQI 41.68), Win 161, Saf 49/3 in 8.4 overs (Win BQI 50.67).
In the above Test, the Saf second innings will be evaluated at 41.68 since fewer than 10 overs were bowled.
5. The BQI is based on the Bowling average. However in order to recognize the importance of Strike rates in Test cricket is a special adjustment based on Strike rate. These concepts are explained in the examples.
6. The BQI is reduced by 5% for Home games and increased by 5% for away games. Reader should remember that the lower the BQI, the more potent the attack is. There have been suggestions on increasing this quantum and on making this dependent on the specific country. I feel 5% either way is ample and the later requires some tricky work since I am not sure how to make it work. So that is for a later day. In general this concept is fine and works well in most cases.
It is possible that the visiting team has the right bowlers and can exploit the "away" bowling conditions. However there is no denying that, in most cases, the home bowlers would have the advantage of familiarity with and knowledge of local conditions.
7. The BQI is further modified by the Period related factor. The concerned table is given below. If the period average is lower than the all-time average of 31.76, it is a bowler-friendly age and the bowler/team averages are adjusted upwards. On the other hand, if the period average is higher than the all-time average of 31.76, it is a batsmen-friendly age and the bowler/team averages are adjusted downwards. I have adjusted the factor at a bowler level than spin/pace level since the later would have required a completely different way of working, at a player level. Even checking of results would have become very cumbersome and difficult. I also do not think that there is that much of a change.
Test 226, Nzl (Hammond's 336), 4 "bowlers" had a ctd total of 26 wkts and a career total of 36 wkts, BQI value is 51.62.
There is no way I am going to put these into the same group. One is a good Test level attack and the other is one which Sydenham's College or Winchester School would dismiss as too weak. So the last group is necessary.
The average BQI for this huge sample is 37.13 and the median is at 37.13. This indicates a fairly balanced distribution of values. The Standard Deviation is 7.67. I have explained the whole concept of determining the BQI with the following examples.
First is MatchId 1267 between Sri Lanka and Pakistan, played at Kandy during 1994. In the first innings Wasim Akram (ctd 22.9) bowled 14.2 overs and Waqar Younis (ctd 19.3) bowled 14 overs and dismissed Sri Lanka for 71. The weighted BQI starts life at 20.99. This is multiplied by the Period index of 1.008 (this being a slightly bowler-friendly decade), 1.05 (this being away Test for Pakistan) and 0.93 (outstanding strike rate for Waqar and Wasim takes the weighted value to 45.1, an outstanding number. The final BQI value is 20.56 which puts this attack as one of the best ever. Any runs scored in this particular innings, say A de Silva's 7 will get into the highest classification.
The second is MatchId 1844 between Pakistan and South Africa, played during 2007. These were Steyn's early years. As everyone knows he had a fairly average start to his career. Steyn bowled 12 overs, Ntini 8 overs, Kallis 7 overs, Harris 20 overs and Nel 16 overs. A fairly pedestrian attack. The base BQI is 34.77. This is multiplied by the Period index of 0.909 (this being a strongly batsmen-friendly decade), 1.05 (this being away Test for South Africa), 1.05 (the weighted strike rate values coming to 75.6, a below average number). Remember these are early-Steyn days. The final BQI value is 34.55 which puts this attack as a fair one. Any runs scored in this particular innings, say Mohammed Yousuf's 25 will get into the second classification.
I have got into details here so as to give the readers a clear idea of the calculations.
There is so much data available that even the organization of the article is getting into trouble. I can only present in the article a certain amount of data. The serious reader should download the complete files and read the same. I have given below what I would be presenting within the article.
I have summarized all relevant facts related to this analysis. First let me emphasize that this is not a Test innings Ratings analysis. There are many other relevant factors which would have to be considered in such an analysis. I have not done so in this analysis which is centred on Bowler quality. I would appreciate if the readers do not keep on repeating again and again that other relevant factors such as Pitch type, Innings status at entry, Result, Series status, Bowler recent form, Innings target et al, have not been included. That would be counter-productive.
1. As I have done in the Team strength calculations, I have considered only Tests played after 1900. It is impossible to fit in the Tests before 1900 because of uncovered pitches and many sub-20 averages. However we lose only 64 Tests.
2. The Bowling quality index (BQI) is based on Career-to-date values. This is the most dependable and accurate of the bowling measures. There is no situation where the Career-to-date figure is not the appropriate one. A bowler like Lee with a great start and tapering off towards the end or Muralitharan who had a poor start and wonderful finish will be taken care of equally well.
3. The BQI is based on the actual bowlers who bowled in the particular innings. This is very important. There is a Sri Lankan innings in which Wasim Akram and Waqar Younis bowled 14 overs each. That is all. This would have been a terrifying situation for the batsmen. Contrast this with an innings in which Akram and Younis bowled 50% of the overs.
4. The BQI is determined for each innings. However in order to reduce wild variations, I will apply the BQI of the first innings to the second innings also in case the number of balls bowled in the second innings is less than 60. This is commonsense. This is explained through an example. Readers should know that this would not have much of an impact since no batsman is likely to score even 25 within the 10 overs.
Test # 1962: Win 231, Saf 346 (Win BQI 41.68), Win 161, Saf 49/3 in 8.4 overs (Win BQI 50.67).
In the above Test, the Saf second innings will be evaluated at 41.68 since fewer than 10 overs were bowled.
5. The BQI is based on the Bowling average. However in order to recognize the importance of Strike rates in Test cricket is a special adjustment based on Strike rate. These concepts are explained in the examples.
6. The BQI is reduced by 5% for Home games and increased by 5% for away games. Reader should remember that the lower the BQI, the more potent the attack is. There have been suggestions on increasing this quantum and on making this dependent on the specific country. I feel 5% either way is ample and the later requires some tricky work since I am not sure how to make it work. So that is for a later day. In general this concept is fine and works well in most cases.
It is possible that the visiting team has the right bowlers and can exploit the "away" bowling conditions. However there is no denying that, in most cases, the home bowlers would have the advantage of familiarity with and knowledge of local conditions.
7. The BQI is further modified by the Period related factor. The concerned table is given below. If the period average is lower than the all-time average of 31.76, it is a bowler-friendly age and the bowler/team averages are adjusted upwards. On the other hand, if the period average is higher than the all-time average of 31.76, it is a batsmen-friendly age and the bowler/team averages are adjusted downwards. I have adjusted the factor at a bowler level than spin/pace level since the later would have required a completely different way of working, at a player level. Even checking of results would have become very cumbersome and difficult. I also do not think that there is that much of a change.
Bowling average adjustment: AMF - Average multiplying factor Period BowAvg AMF 1877-1899 22.20 1.431 1900-1914 25.69 1.236 WW1-WW2 32.56 0.976 40s-50s 29.96 1.060 1960s 32.11 0.989 1970s 31.94 0.995 1980s 32.07 0.990 1990s 31.51 1.008 2000-2004 33.56 0.946 2005-2011 34.94 0.909 All Tests 31.76 1.000Finally the bowling attacks are classified into 5 groups, as described below. The fifth group was necessary to separate the REALLY weak bowling attacks. With the idea of short innings being tagged with the first innings, there have been 6837 qualifying innings until the West Indies - Pakistan Test which finished recently. The first cut-off has been fixed at 30 to have around 20% of the total innings into the top group. There may be a subjective element in this part of the exercise but that cannot be avoided. The basis on which we have decided that 30 will be the first cut-off point is not subjective. In fact Arjun's assertion that 20% means in a loose manner that at any time there are 2 really good bowling attacks makes eminent sense. The other cut-offs follow logically. The group cut-off details are given below.
Group B Q I # of Inns % (out of 6837) 5 19.03-29.99: 1313 19.2% Very good bowling attack. 4 30.00-34.99: 1632 23.9% Good bowling attack. 3 35.00-39.99: 1454 21.3% Average bowling attack. 2 40.00-44.99: 1600 23.4% Passable bowling attack. 1 45.00+: 838 12.2% Very poor bowling attack. Note: In group 5 only two innings were below 20.00. Test 830: England (19.03). Botham (17.3) and Willis (24.3) bowled 34 out of 37 overs. Test 237: Australia (19.89). Miller (22), Johnston (18.2) and Lindwall (19.1) bowled 22 overs.In fact Arjun questioned the necessity to have the last group. He felt that the last two could be combined into one. However I strongly feel it is essential for the reason cited below. Let us look at the following two bowling attacks.Test 1833, Bangladesh, Mortaza/Rafique/Shakib/Rasel/Sharif, BQI value is 40.38.
Test 226, Nzl (Hammond's 336), 4 "bowlers" had a ctd total of 26 wkts and a career total of 36 wkts, BQI value is 51.62.
There is no way I am going to put these into the same group. One is a good Test level attack and the other is one which Sydenham's College or Winchester School would dismiss as too weak. So the last group is necessary.
The average BQI for this huge sample is 37.13 and the median is at 37.13. This indicates a fairly balanced distribution of values. The Standard Deviation is 7.67. I have explained the whole concept of determining the BQI with the following examples.
First is MatchId 1267 between Sri Lanka and Pakistan, played at Kandy during 1994. In the first innings Wasim Akram (ctd 22.9) bowled 14.2 overs and Waqar Younis (ctd 19.3) bowled 14 overs and dismissed Sri Lanka for 71. The weighted BQI starts life at 20.99. This is multiplied by the Period index of 1.008 (this being a slightly bowler-friendly decade), 1.05 (this being away Test for Pakistan) and 0.93 (outstanding strike rate for Waqar and Wasim takes the weighted value to 45.1, an outstanding number. The final BQI value is 20.56 which puts this attack as one of the best ever. Any runs scored in this particular innings, say A de Silva's 7 will get into the highest classification.
The second is MatchId 1844 between Pakistan and South Africa, played during 2007. These were Steyn's early years. As everyone knows he had a fairly average start to his career. Steyn bowled 12 overs, Ntini 8 overs, Kallis 7 overs, Harris 20 overs and Nel 16 overs. A fairly pedestrian attack. The base BQI is 34.77. This is multiplied by the Period index of 0.909 (this being a strongly batsmen-friendly decade), 1.05 (this being away Test for South Africa), 1.05 (the weighted strike rate values coming to 75.6, a below average number). Remember these are early-Steyn days. The final BQI value is 34.55 which puts this attack as a fair one. Any runs scored in this particular innings, say Mohammed Yousuf's 25 will get into the second classification.
I have got into details here so as to give the readers a clear idea of the calculations.
There is so much data available that even the organization of the article is getting into trouble. I can only present in the article a certain amount of data. The serious reader should download the complete files and read the same. I have given below what I would be presenting within the article.