DoG's top 100 test bowler countdown: Adjust for wickets per innings

[Days of Grace]

I have one final question to the forum before commencing the top 100 test bowler countdown. It is whether to adjust the wickets per innings along with the average and strike-rate.

The weights are as follows (Mirroring that of test batsmen):
Average x6
Wickets per innings x3
Strike-rate x1

It is clear that Malcolm Marshall and Glenn McGrath had to compete more for their share of the 10 wickets available per innings compared to bowlers such as Mutiah Muralidaran and Richard Hadlee.

If you think wickets per innings should be adjusted, do you have any thoughts about how they could be refined? I don't have time to go through an innings-by-innings adjustment for every bowler but perhaps a career adjustment can be done.

Please vote over the next 24 hours.

 

[ataraxia]

The weights do not mirror those of test batsmen. The task of the bowler is to thwart the batsman, and vice versa. If scoring quickly is preferable, that means that bowlers should prevent quick scoring. So the weight on strike rate should be changed to economy.

 

[The Sean]

The weights do not mirror those of test batsmen. The task of the bowler is to thwart the batsman, and vice versa. If scoring quickly is preferable, that means that bowlers should prevent quick scoring. So the weight on strike rate should be changed to economy.

Completely agree with this. And not just because it helps Keith Miller.*








* It also helps Bill O'Reilly

 

[Days of Grace]

The weights do not mirror those of test batsmen. The task of the bowler is to thwart the batsman, and vice versa. If scoring quickly is preferable, that means that bowlers should prevent quick scoring. So the weight on strike rate should be changed to economy.

I agree with this for limited overs cricket. But in test cricket, wickets win matches, and the time is limited, so a bowler who takes wickets quickly is better than a containing bowler. In addition, economy-rate is reflected in a bowler's average, the runs they concede per wicket.

 

[ataraxia]

I agree with this for limited overs cricket. But in test cricket, wickets win matches, and the time is limited, so a bowler who takes wickets quickly is better than a containing bowler. In addition, economy-rate is reflected in a bowler's average, the runs they concede per wicket.

The idea that turning draws into wins is more valuable than turning losses into draws, and therefore batsmen and bowlers should cooperate to score and take wickets as quickly as possible, is a modern one. It does not reflect what I assume this countdown will concern, the entirety of test history.

Strike rate and economy are equally reflected in bowling average: they combine to create it.

 

[Randomfan]

I agree with this for limited overs cricket. But in test cricket, wickets win matches, and the time is limited, so a bowler who takes wickets quickly is better than a containing bowler. In addition, economy-rate is reflected in a bowler's average, the runs they concede per wicket.

I accidentally posted in other thread, copying it here.

----------------

In the test format, bowlers main job is to take wickets cheaply(Avg) and quickly(SR). Taking cheaply directly helps the team and taking quickly allows other bowlers to bowl to a new batsman. Low ER can put pressure and also gets you wickets, but it's less important in the test format than limited overs. Now if we were talking about limited overs then SR becomes second fiddle to ER. You can win limited overs by simply containing runs without picking a single wicket but you can't do it in the test format. In the test format, goal is to pick 20 wickets.

Also, avg has input from ER and SR. They are mathematically linked. Avg = (ER * SR)/6 I think, the most important is Avg and then SR and then ER in the test format.

 

[Johan]

bowling strike rates hold value because they can demonstrate wicket-taking-ability of a bowler, many times it's likely the player with the lower strike rate would end up with higher wickets accross career, for example Steyn and Donald take more an innings than McGrath and Ambrose, now obviously you have your Hadlee or Barnes who utterly and completely dwarf Steyn or Donald in wicket taking and their SR just doesn't matter as they make up for it by bowling way more balls. However, batting strike rate has nothing to do with demonstrating how many runs one makes, that's RPI, it's all about the pace of getting things done.

While bowling strike rate often influences a factor that actually matters (number of wickets taken), batting SR and ER are completely related to pace of the game and have no value in reality. Bowling SR should hold more value than either, bowling ER and batting SR are equivalent as they are entirely related to the pace of the game.

 

[honestbharani]

I think it should be adjusted but also keeping in mind some teams may have better bowlers in their line up so may not use certain bowlers. If we can also add in good weightage for bowling strike rate as well, I think it might make sense together.

 

[Line and Length]

Wickets per innings is detrimental to bowlers who were part of a strong team attack. Hence Hadlee and Murali are given an advantage over McGrath or Marshall. Economy rate should be taken into consideration, Bowlers who tie up an end while attack come from the other end often play an important roll.

 

[Randomfan]

It is clear that Malcolm Marshall and Glenn McGrath had to compete more for their share of the 10 wickets available per innings compared to bowlers such as Mutiah Muralidaran and Richard Hadlee.

If you think wickets per innings should be adjusted, do you have any thoughts about how they could be refined? I don't have time to go through an innings-by-innings adjustment for every bowler but perhaps a career adjustment can be done.

Please vote over the next 24 hours.

Wickets per inning can be higher if there is no competition and you bowl more overs. You can have worst average and worst SR, two primary criterion for bowers, but end up with lot more wickets per inning simply because you bowled more. That does not make a better bowler but It shows that the team is dependent on you. Having said that, if you have great support from other end then you are likely to pick wickets due to pressure applied from the other end. So wickets per inning is not meanigless but not sure if it should, have 3 times weight of SR and half as much as avg. Not sure what should be the ratio but giving wickets per ining half weight as avg seems very high to me. It should be less. Perhaps giving the same weight to wickets per inning and SR. Avg is anyway main criterion and rest is support.

 

[DrWolverine]

1. WPI/WPM shouldn’t adjusted.

2. No need for economy rate. It is anyway reflected in the bowling average.

3. Strike rate should be given more importance. A bowler who takes more wickets in less balls gives the team more time to win

 

[Johan]

1. WPI/WPM has to be adjusted, like everything else.

2. If strike rates for batsmen were counted (runs scored per ball), inverse of that is economy rate for bowlers and that should also be counted (runs conceded per ball)

3. Strike rates, no opinion on bowling strike rate.

 

[Johan]

WPI is also a thing to be era adjusted too. Certain eras, the more defensive ones, produce naturally lower WPI even for the best bowlers.

 

[Two short legs]

Proportion of victims that are top-7 batsmen ? Not every wicket is equal. Although I accept that there is value in being able to remove tailenders quickly. And opening bowlers have the advantage of bowling more often to a new batsman, with a new ball, on a fresh pitch.

Only 65% of Akram's wickets were top-7, falling to 61% when he captained himself. In contrast the figures were 78%-81% for Larwood, Lawson, Thomson, Miller and Hall. Which makes me think that they were often kept in cotton-wool for the next innings.

 

[DrWolverine]

Top six were the main batsmen.
Pre 2000s, No. 7 was usually a WK with limited batting

 

[shortpitched713]

@Days of Grace I honestly think you should disregard this poll, and just do it however you feel like doing it.

Sure people cribbed a bit about strike rate in the batsmen ranking, but in the end it was still a vastly appreciated effort, and the process of putting out the list, debates, predictions, etc. was a load of fun. Back yourself and whatever your gut is feeling, and I'm sure this next bowlers list will be the same!

 

[Thala_0710]

I agree with this for limited overs cricket. But in test cricket, wickets win matches, and the time is limited, so a bowler who takes wickets quickly is better than a containing bowler. In addition, economy-rate is reflected in a bowler's average, the runs they concede per wicket.

I agree with this but the bowling avg already reflects the SR as well.
As others have pointed out Bowling avg is basically = (SR*Eco)/6, so it's statistically just a metric which accounts for both.
If I had to do it, I'd give Eco and SR the same weights. SR demonstrates the pure wkt taking ability, Eco demonstrates the "unhittability" of a bowler, and also the pressure one creates through it. There's lots of times, less so than other formats but still, where we've seen that a bowler like Hazlewood is just keeping it tight at one end, Starc comes up at the other end and the batter tries to attack him and gets out in doing so. This doesn't appear in Hazlewood's stats but he has a crucial role in the wicket.
So overall I'd give both the same weight.

 

[Thala_0710]

Also yeah, every metric including WPI could be adjusted imo, at least relative to era if not team strength ideally

 

[capt_Luffy]

I agree with this but the bowling avg already reflects the SR as well.
As others have pointed out Bowling avg is basically = (SR*Eco)/6, so it's statistically just a metric which accounts for both.
If I had to do it, I'd give Eco and SR the same weights. SR demonstrates the pure wkt taking ability, Eco demonstrates the "unhittability" of a bowler, and also the pressure one creates through it. There's lots of times, less so than other formats but still, where we've seen that a bowler like Hazlewood is just keeping it tight at one end, Starc comes up at the other end and the batter tries to attack him and gets out in doing so. This doesn't appear in Hazlewood's stats but he has a crucial role in the wicket.
So overall I'd give both the same weight.

Just count Avg, it already does that

 

[Thala_0710]

Just count Avg, it already does that

Yup that's what I meant, Avg and WPI/WPM. If you're giving another point to SR, I'd give it to Eco too