Suggestions for top 30 bowlers countdown

[venkyrenga]

So I am going to start working on the bowlers ranking. I will be using the same formulas I used for batsmen with slight changes.

Bowling average and WPI with a weightage of 3:1
Instead of SR I will be using Econ. Bowlers who have their Econ above the mean will receive a penalty of 1% for every 4% above the mean. So if the mean is 3.2, a bowler with Econ of 3.33 will have 1% of his points deducted.

I am not using SR because it is almost always directly proportional to the average and average is already factored in. All the other criteria and methods will remain the same. Let me know if you guys think something should be changed.

And most importantly, I am unable to pull the data the way I want in cricinfo. I am trying to pull the bowlers' data against no.1-7 and no.8-11 separately and give more weightage for the record against no.1-7. Does anyone know if it is possible? Otherwise I will just have to use the combined data.

 

[TheJediBrah]

I don't think anything "should be changed", the whole point of this should be you doing things your way and seeing what results you get, you're the one doing the work, and as I'm sure you know you can't please everybody (least of all me).

Personally I think including WPI at all just makes the rankings less accurate than if you didn't. And unless I am really confusing my math, using ER instead of SR should make literally no difference to the final result because both are "directly proportional" to average and would rate the same way.

 

[Starfighter]

1. Glenn McGrath

/endthread

 

[Gob]

Suggestions

1.Do the spinners and fast bowlers together
2. I would like SR to be factored more than the econ as TJB said they are all proportionate to the average
3. Further break down of batsmen dismissed as 1 to 4, 5 to 7, tail with each category become less rewarding
5. Also I don't know if its possible since we are using the same formula but if possible more value to spinners in SENA conversely more value to fast bowlers in BISPU
6. Again if possible more value to spinner performances in the 1st and 2nd digs without detracting from their performances in 3rd and fourth. Opposition for the faster bowlers
7. Less value when performing under cloud cover

Glenn and Dale step forward

 

[trundler]

Average is just the product of economy and SR/6 so including either individually along with average doesn't work imo. You could do away with average altogether and work out SR and ER in your preferred ratio.

 

[Teja.]

Waqar and Steyn have the GOAT SR or close and an average ER so how you weigh those things will have an effect on how those bowlers are ranked, for example.

 

[The Sean]

I probably need to give this more thought, but while it's self-evident that average is a function of SR and ER, I don't - as far as my current thinking leads me - agree that they are of equal importance and that it's a case of "neither or both." While you could obviously argue that mathematically it should all even out, my opinion would be in Test cricket that wicket-taking ability - SR - is more important than containment - ER.

Of course, both being brilliant is the ideal, but if it's one or the other then I'd always tend to throw my preference behind the bowler who can get teams out more quickly and give his team more time to score the runs they need. Certainly, there is no way I would recommend the formula you suggested above that punishes a poor ER but doesn't reward a good SR. That's the exact opposite of where my thinking would lie.

 

[venkyrenga]

Lets not think in terms of specific bowlers as that may lead us to be biased. I know its difficult but lets try not to. Average will be the base and I don't think that will change. Now consider the below,

	Overs	Runs	Wks	Econ	SR
A	16​	60​	3​	3.8​	32​
B	20​	60​	3​	3.0​	40​
C	24​	60​	3​	2.5​	48​

Assuming that the quality of batsmen dismissed by all three is the same, if I had to rank them, I would go B>=C>A

B&C come off as dominant bowlers whereas A comes off as a bowler who took advantage of the pressure built by B&C. That is why I believe ER to be more important than SR. Thoughts?

 

[Gob]

I probably need to give this more thought, but while it's self-evident that average is a function of SR and ER, I don't - as far as my current thinking leads me - agree that they are of equal importance and that it's a case of "neither or both." While you could obviously argue that mathematically it should all even out, my opinion would be in Test cricket that wicket-taking ability - SR - is more important than containment - ER.

Of course, both being brilliant is the ideal, but if it's one or the other then I'd always tend to throw my preference behind the bowler who can get teams out more quickly and give his team more time to score the runs they need. Certainly, there is no way I would recommend the formula you suggested above that punishes a poor ER but doesn't reward a good SR. That's the exact opposite of where my thinking would lie.

Its interesting. Suppose you have have two bowlers averaging 25 each but one strikes at 40 but concedes 3.5 while the other strikes st 60 but concedes only 2.5.

If they both bowls to their stats for 20 overs, you've got one taking 2 for 50 while the other 3 for 70. I think most of the guys who tend to strike fast tend to have higher econs as they could be searching wickets bowling full.

 

[Teja.]

I think SR/ER is like batting SR in test cricket. Higher batting SR (for top order batsmen) is generally more helpful for strong teams with good bowling and batting lineups who want to push for a win and would rather not draw, especially on flatter tracks when the game is at risk for going on for five days otherwise. There's also a lot of other variables which affect this obviously.

I think usefulness of lower bowling SR has similar functions relating to the overall strength of the bowling lineup of the bowler, whether he has a batting lineup is good enough to push for a win to take advantage of his low bowling SR and how result oriented the nature of his home conditions where he will play half his matches in are.

The same would work in reverse for situations when lower SR for the same returns would be more valuable to draw more naturally losing games due to the other 10 players in the team apart from the bowler not being able to push for a win. Again, there's a lot of other variables which would affect how useful batting/bowling SR is.

As a general rule, the strength of the other players in the team is directly proportional to how useful a low bowling SR is IMO but overall, I think there are too many variables for this single SR/ER stat which means that ultimately, including it as general factor, either prejudicial or beneficial will probably not yield useful results

 

[Gob]

Lets not think in terms of specific bowlers as that may lead us to be biased. I know its difficult but lets try not to. Consider the below,

	Overs	Runs	Wks	Avg	Econ	SR
A	160​	600​	30​	20​	3.8​	32​
B	200​	600​	30​	20​	3.0​	40​
C	240​	600​	30​	20​	2.5​	48​

If I had to rank them, I would go B>=C>A

B&C come off as dominant bowlers whereas A comes off as a bowler who took wickets because of the pressure built by B&C. That is why I believe ER to be more important than SR. Thoughts?

You could actually pick 3 bowlers who suit this argument with more relatable stats (I mean no one strikes at 32) and then see if the end results justify our consensus as of who is better. But ofcourse most people here in CW may identify the bowler from the stats

 

[venkyrenga]

You could actually pick 3 bowlers who suit this argument with more relatable stats (I mean no one strikes at 32) and then see if the end results justify our consensus as of who is better. But ofcourse most people here in CW may identify the bowler from the stats

Ok, I have edited to make it seem more practical.

 

[Red_Ink_Squid]

Lets not think in terms of specific bowlers as that may lead us to be biased. I know its difficult but lets try not to. Consider the below,

	Overs	Runs	Wks	Econ	SR
A	16​	60​	3​	3.8​	32​
B	20​	60​	3​	3.0​	40​
C	24​	60​	3​	2.5​	48​

If I had to rank them, I would go B>=C>A

B&C come off as dominant bowlers whereas A comes off as a bowler who took wickets because of the pressure built by B&C. That is why I believe ER to be more important than SR. Thoughts?

Hard to consider in the abstract as the exact reverse of your argument could be made (bowlers B & C were able to bowl economically because bowler A was dismissing the set batsmen). In reality the gamestate would determine which return was more important to the team.

But extrapolating to a (much) higher sample size than 3 wickets, I'd say A > B > C.

Ideally you want your best averaging bowlers (who your countdown will be focusing on) to be the ones taking wickets. For a support bowler you might prefer them to be more economical to give your best bowlers a rest without letting the game get away. But if you've got a guy who can take wickets at 20 runs a piece you want them taking higher wickets per match so that they can impact the match the most.

 

[The Sean]

I don't think anything "should be changed", the whole point of this should be you doing things your way and seeing what results you get, you're the one doing the work, and as I'm sure you know you can't please everybody (least of all me).

Personally I think including WPI at all just makes the rankings less accurate than if you didn't. And unless I am really confusing my math, using ER instead of SR should make literally no difference to the final result because both are "directly proportional" to average and would rate the same way.

What's your argument against WPI? I'm not doubting you, I just don't think I've seen it articulated and I'm interested.

 

[Lillian Thomson]

I can’t be arsed with mucking about with decimal points, but if Graeme Swann makes the 30 start again.

 

[Engle]

I probably need to give this more thought, but while it's self-evident that average is a function of SR and ER, I don't - as far as my current thinking leads me - agree that they are of equal importance and that it's a case of "neither or both." While you could obviously argue that mathematically it should all even out, my opinion would be in Test cricket that wicket-taking ability - SR - is more important than containment - ER.

Of course, both being brilliant is the ideal, but if it's one or the other then I'd always tend to throw my preference behind the bowler who can get teams out more quickly and give his team more time to score the runs they need. Certainly, there is no way I would recommend the formula you suggested above that punishes a poor ER but doesn't reward a good SR. That's the exact opposite of where my thinking would lie.

AWTA. ER has greater value in ODI's, but SR has greater value in Tests

FWIW, an article I read many years ago gives equal weight to Test AVG and SR by multiplying the two and taking the square of that

 

[venkyrenga]

For a higher sample size we can consider this,

	Overs	Runs	Wks	Avg	Econ	SR
A	190	660	27	24.4	3.5	42.2
B	205	660	27	24.4	3.2	45.6
C	230	660	27	24.4	2.9	51.1

 

[Migara]

Lets not think in terms of specific bowlers as that may lead us to be biased. I know its difficult but lets try not to. Average will be the base and I don't think that will change. Now consider the below,

	Overs	Runs	Wks	Econ	SR
A	16​	60​	3​	3.8​	32​
B	20​	60​	3​	3.0​	40​
C	24​	60​	3​	2.5​	48​

Assuming that the quality of batsmen dismissed by all three is the same, if I had to rank them, I would go B>=C>A

B&C come off as dominant bowlers whereas A comes off as a bowler who took advantage of the pressure built by B&C. That is why I believe ER to be more important than SR. Thoughts?

Not really. It is usually A who carves through sides and opens up opportunities for others.

 

[ataraxia]

Lets not think in terms of specific bowlers as that may lead us to be biased. I know its difficult but lets try not to. Average will be the base and I don't think that will change. Now consider the below,

	Overs	Runs	Wks	Econ	SR
A	16​	60​	3​	3.8​	32​
B	20​	60​	3​	3.0​	40​
C	24​	60​	3​	2.5​	48​

Assuming that the quality of batsmen dismissed by all three is the same, if I had to rank them, I would go B>=C>A

B&C come off as dominant bowlers whereas A comes off as a bowler who took advantage of the pressure built by B&C. That is why I believe ER to be more important than SR. Thoughts?

In this scenario I'd think A the best but if instead we were to choose only one of them rather than having all three in one innings I'd pick C.

 

[TheJediBrah]

What's your argument against WPI? I'm not doubting you, I just don't think I've seen it articulated and I'm interested.

Basically because it is mostly reliant on factors outside the bowlers' control, eg. performance of other bowlers in team, type of bowler, overall strength of bowling attack. If you're Malcolm Marshall your WPI won't be anywhere near as high as a Murali because you are bowling a lower percentage of team's overs and the other bowlers in your team are taking more wickets.

It might shake the rankings up a bit, but not in a way that rewards the better bowlers.