DoG's Top 100 Test Bowlers Countdown Thread 100-1

[stephen]

Garner, Davidson, Walsh and Herath would make for a damn fine attack.

 

[Starfighter]

Nah, Sydney's as much a shithole as Melbourne, if not more so.

 

[stephen]

Melbourne: Not quite as bad as Brisbane.

Sydney: the dumping ground for those who don't know Brisbane and Melbourne are better places to live.

 

[Burgey]

Nah, Sydney's as much a shithole as Melbourne, if not more so.

This is empircally untrue.

 

[Starfighter]

This is empirically true.

ftfy

 

[morgieb]

Sydney: the dumping ground for those who don't know Brisbane and Melbourne are better places to live.

Yeah, posts like these are why we need the dislike button.

 

[TheJediBrah]

Sydney: the dumping ground for those who don't know Brisbane and Melbourne are better places to live.

Sydney's ok to visit. Wouldn't live there if you paid me though.

 

[Burgey]

Fortunately no one here wants you.

Melbourne thinks it's Paris because it has cafes. It's really Darwin with a river and winter clothing.

 

[Bolo]

Definitely. Wpm shouldn't really be a metric at all in this sort of analysis. It doesn't really reflect a bowler's ability or skills so much as how much many overs they bowled/how good the other bowlers in the team were.

I'm not saying it should be dumped. It's an extremely important metric. It's just a very flawed one as well. All of them are, but there are common ways of dealing with these like era/quality adjusting them, but it's difficult to do for WPM, and I don't think I've really seen it done before.

PPI is more advanced than WPM. A bowler’s performance (points per innings) takes in runs per wicket, balls per wicket, conditions, opposition, result, etc.

True, you lose points if you take less than 5 wickets and it favors bowlers who take a lot of 5-fers but nonetheless it is not a straight WPM analysis.

Not suggesting it is, but it is an important component. The WI quicks come out ahead on all of this after you have adjusted them, except in their ability to turn in great performances, which is to a large degree a function of competition and indirectly (but not completely) corresponds to wpm.

 

[TheJediBrah]

I'm not saying it should be dumped.

No, I am

 

[Starfighter]

I'm not saying it should be dumped. It's an extremely important metric. It's just a very flawed one as well.

Does not compute. If a metric is flawed then that's a very good reason for considering it not important.

 

[Bolo]

Does not compute. If a metric is flawed then that's a very good reason for considering it not important.

If you don't take 20 wickets you don't win. WPM is the most important metric there is in one sense. You can use SR, but it's not good enough- it doesn't give you a picture of ability to bowl a large number of overs and hence take more wickets. WPM gives a flawed picture of this as it depends on competition, but it's a whole lot better than nothing.

 

[Starfighter]

**** is not necessarily better than no ****

 

[Migara]

Definitely. Wpm shouldn't really be a metric at all in this sort of analysis. It doesn't really reflect a bowler's ability or skills so much as how much many overs they bowled/how good the other bowlers in the team were.

Agree with you. It should be replaced with Wickets per innings stat. WPM is dependent on the batting prowess of the team as well. A team putting up large totals will allow their bowlers to bowl at opposition twice on more occasions. In addition, the quality of the support attack as well as fitness of the said bowler all will determine WPM stat. Fitness of a player is a quality that we have to positively recognize, so WPI or WPM can be used to get an indirect idea about it.

 

[TheJediBrah]

If you don't take 20 wickets you don't win. WPM is the most important metric there is in one sense.

Wpm tells you nothing about whether you take 20 wickets or not. Strike rate does. Wpm is literally just a function of strike rate & number of overs bowled.

Strike rate is a way better metric in that sense. Say bowler A bowls 50 overs a game take 6 wickets a game because the rest of his team is ****. Bowler B bowls 25 overs a game and takes 5 wickets a game. Which bowler is better for taking 20 wickets? Bowler B. If Bowler B played in Bowler A's team he'd probably have 7-8 wickets a game.

The only possible value of wpm as a statistical measure is if you really think that bowling more overs per match significantly reduces your ability due to fatigue, which is very questionable in the context of how bowlers are compared. You're generally comparing, say, 30 overs a game to 25 overs a game. You really think the guy bowling an extra 5 overs a game is having his strike-rate and average significantly negatively affected as a result? Come on

 

[Migara]

@JediBrah: SR is important, if a bowler can keep on bowling. Let's assume a top bowler bowls 1/3 of the overs of the team. That is 180 deliveries. If the SR is 45 (which is insanely good), the bowler would take only 4 wickets in a day. To take 6 wickets he had to bowl 45 overs per innings. Now the question is whether the particular bowler is fit enough to do it. SR is important because the number of balls per test is limited. WPI is important because not all bowlers can bowl 30 overs or more per day.

 

[FBU]

Do people still think Rabada is gonna be in it? Not sure a place in the top 20 is more likely than a place just outside the top 100 tbh.

I was thinking the same about Anderson.

 

[TheJediBrah]

@JediBrah: SR is important, if a bowler can keep on bowling. Let's assume a top bowler bowls 1/3 of the overs of the team. That is 180 deliveries. If the SR is 45 (which is insanely good), the bowler would take only 4 wickets in a day. To take 6 wickets he had to bowl 45 overs per innings. Now the question is whether the particular bowler is fit enough to do it. SR is important because the number of balls per test is limited. WPI is important because not all bowlers can bowl 30 overs or more per day.

The only possible value of wpm as a statistical measure is if you really think that bowling more overs per match significantly reduces your ability due to fatigue, which is very questionable in the context of how bowlers are compared. You're generally comparing, say, 30 overs a game to 25 overs a game. You really think the guy bowling an extra 5 overs a game is having his strike-rate and average significantly negatively affected as a result? Come on

Migara if we have 2 hypothetical bowlers, say a Shaun Tait who struggles to bowl 10 overs a day and a Matthew Hoggard who bowls 25 overs a day, then sure, you may have a point. Tait would have a limited value even if he had a much higher strike rate. But that's an extreme example and generally not particularly relevant when making practical comparisons.

 

[TheJediBrah]

I was thinking the same about Anderson.

lol Anderson will be in it for sure. His career points are huge. He'll be top 10.

 

[Bolo]

Agree with you. It should be replaced with Wickets per innings stat. WPM is dependent on the batting prowess of the team as well. A team putting up large totals will allow their bowlers to bowl at opposition twice on more occasions. In addition, the quality of the support attack as well as fitness of the said bowler all will determine WPM stat. Fitness of a player is a quality that we have to positively recognize, so WPI or WPM can be used to get an indirect idea about it.

Yes. I tend to use them interchangeably without thinking because they are similar, but WPI is definitely better.

**** is not necessarily better than no ****

Discard the single most reliable measure in predicting results because it is an imperfect measure? Every stat is. May as well just discard stats altogether if going this route

Wpm tells you nothing about whether you take 20 wickets or not. Strike rate does. Wpm is literally just a function of strike rate & number of overs bowled.

Strike rate is a way better metric in that sense. Say bowler A bowls 50 overs a game take 6 wickets a game because the rest of his team is ****. Bowler B bowls 25 overs a game and takes 5 wickets a game. Which bowler is better for taking 20 wickets? Bowler B. If Bowler B played in Bowler A's team he'd probably have 7-8 wickets a game.

The only possible value of wpm as a statistical measure is if you really think that bowling more overs per match significantly reduces your ability due to fatigue, which is very questionable in the context of how bowlers are compared. You're generally comparing, say, 30 overs a game to 25 overs a game. You really think the guy bowling an extra 5 overs a game is having his strike-rate and average significantly negatively affected as a result? Come on

Simple rebuttal- spinner vs quick