How valuable is that wicket!

[vcs]

I wonder why Wasim and Waqar are so low down on the list of absolute average value of wickets despite playing in the same era as Ambrose and Walsh. Did they make hay against, on average, inferior opposition? Did they take a higher percentage of tail wickets?

 

[ankitj]

I wonder why Wasim and Waqar are so low down on the list of absolute average value of wickets despite playing in the same era as Ambrose and Walsh. Did they make hay against, on average, inferior opposition? Did they take a higher percentage of tail wickets?

Wasim definitely took lot of tail wickets. Not sure about Waqar.

 

[GotSpin]

Well, except that Miller had far shorter career. Nonetheless quite an excellent figure he notched up!

Still played for almost 11 years

 

[Prince EWS]

The real problem I have with this analysis is that it doesn't take into account the quality of the batsmen each bowler bowled to but didn't dismiss. One's bowling average is not merely a representation of one's effectiveness against batsmen one dismissed, but any batsmen one bowled to. Interesting all the same, however.

 

[GotSpin]

As for, Ambrose and Walsh,

Ambrose is better by a good margin.

However, If I had to have to pick between one of Waqar, Ambrose or Walsh starting his career with India right now. I'd take Ambrose's slightly lower quality 18 years over the other bowler's decade each of awesome bowling. Walsh over Akram, who played for 18 too, because Walsh remained awesome throughout his career while Akram was for only 8-9 years.

All three bowlers are better bowlers overall though.

Did you mix up Ambrose and Walsh? Ambrose played for 12 years

 

[Prince EWS]

Did you mix up Ambrose and Walsh? Ambrose played for 12 years

He meant to say Walsh instead of Ambrose in that bolded part, I'm sure. He was posting the other day about how he felt Ambrose was over-rated and Walsh under-rated.

 

[Daemon]

Taking a high number of tail end wickets would bring your 'Average value of wicket' down. This may allow for a better view or indication of a certain players bowling average. However, we can't use this value of wicket to compare players can we?

For example if you take a bowler who opens the bowling only and does not bowl with the older ball much, when tail enders are present, naturally he will take more top order wickets and thus have a higher average value of wicket. But if you have someone who bowls at the tail more often, taking almost 1-2 lower order wickets per game, he will have a lower average value per wicket as compared to the previous bowler, even if both took the same number of top order wickets. You can't really use this value to compare players imo, although it does give a better indication of where all the wickets of a player came from.

 

[ankitj]

Ankit,

If it's not too much trouble, I'd appriciate it if you could do this for Waqar from 1 jan 1990 to 31 dec 1994 and 1 jan 1990 to 31 dec 1999.

The latter would provide a more accurate account of Waqar as a bowler because he was a pretty average bowler for three uneccesary years after 2000 and the former is the greatest peak by a pacer ever. Thanks.

We could do that but that would be true for lot of bowlers. Hadlee, Imran and Murali had slow starts. Donald, Marshall and Pollock had bad last few years. So it will not be accurate to do it for only one bowler IMO.

 

[vcs]

Taking a high number of tail end wickets would bring your 'Average value of wicket' down. This may allow for a better view or indication of a certain players bowling average. However, we can't use this value of wicket to compare players can we?

For example if you take a bowler who opens the bowling only and does not bowl with the older ball much, when tail enders are present, naturally he will take more top order wickets and thus have a higher average value of wicket. But if you have someone who bowls at the tail more often, taking almost 1-2 lower order wickets per game, he will have a lower average value per wicket as compared to the previous bowler, even if both took the same number of top order wickets. You can't really use this value to compare players imo, although it does give a better indication of where all the wickets of a player came from.

That's why he divides the average value of wickets with the bowling average, because if a bowler is getting the opportunity to take more cheaper wickets, he should be taking them at a lower price.

 

[ankitj]

Taking a high number of tail end wickets would bring your 'Average value of wicket' down. This may allow for a better view or indication of a certain players bowling average. However, we can't use this value of wicket to compare players can we?

For example if you take a bowler who opens the bowling only and does not bowl with the older ball much, when tail enders are present, naturally he will take more top order wickets and thus have a higher average value of wicket. But if you have someone who bowls at the tail more often, taking almost 1-2 lower order wickets per game, he will have a lower average value per wicket as compared to the previous bowler, even if both took the same number of top order wickets. You can't really use this value to compare players imo, although it does give a better indication of where all the wickets of a player came from.

Thanks for not paying full attention to the opening post :P

 

[ankitj]

The real problem I have with this analysis is that it doesn't take into account the quality of the batsmen each bowler bowled to but didn't dismiss. One's bowling average is not merely a representation of one's effectiveness against batsmen one dismissed, but any batsmen one bowled to. Interesting all the same, however.

That's why once you take ratio of bowling average and average value, that problem is taken care of.

 

[Prince EWS]

That's why once you take ratio of bowling average and average value, that problem is taken care of.

Hmm no not really.

Say you get absolutely belted around by the top order in a game and have 0/130, then skittle the lower order for nothing and end up with 3/150. Your scorebook average for the innings is 50. Your system basically assumes that you've averaged 50 against prank-batsmen because that's who you got out, where in reality you've averaged about 7 against prank-batsmen and an infinite amount against good batsmen. It doesn't truly account for batsmen you've not not dismissed - it only really accounts for them on one side of the equation.

I don't mean to be too critical - I've always said that any system which provides a more accurate account than the standard statistics packages we have in place is worth looking at, and this one does that, so you should be commended for this thread. I'd sooner read into one of those figures than a scorebook average, so it was worthwhile. There are still significant, unresolvable flaws within it though.

 

[Daemon]

Thanks for not paying full attention to the opening post :P

haha tldr moment

 

[Migara]

Good work mate.

 

[vcs]

Hi Ankit,

Could you do the Indian spin quartet as well? I don't know enough about them, but it'd be interesting to see how they stack up statistically.

 

[ankitj]

Hi Ankit,

Could you do the Indian spin quartet as well? I don't know enough about them, but it'd be interesting to see how they stack up statistically.

I suspect they will do very poorly. But will do when I get chance. With all the requests I am getting, I should get paid for the work

 

[Teja.]

Did you mix up Ambrose and Walsh? Ambrose played for 12 years

I meant to say Walsh in the part you bolded. I was posting the other day about how I felt Ambrose is over-rated and Walsh, under-rated.

 

[Teja.]

We could do that but that would be true for lot of bowlers. Hadlee, Imran and Murali had slow starts. Donald, Marshall and Pollock had bad last few years. So it will not be accurate to do it for only one bowler IMO.

Nah, It was a personal request not a comparison one.

 

[Jono]

I meant to say Walsh in the part you bolded. I was posting the other day about how I felt Ambrose is over-rated and Walsh, under-rated.

The less people that know about that post the better tbh

 

[ankitj]

I'd be interested if you looked at the rivals of Lohmann and Barnes in their time and see how they place.

Here's how some of the bowlers of that era stack up:

[B]Rank	Bowler		Wickets	Bow Ave	 Avg value Discount Factor[/B]
1	J.J. Ferris	61	12.7	20.34	62.45%
2	Sydney Barnes	189	16.43	25.02	65.67%
3	George Lohman	112	10.75	15.92	67.53%
4	Charlie Turner	101	16.53	23.52	70.28%
5	Charlie Blythe	100	18.63	24.90	74.82%
6	Hugh Trumble	141	21.78	26.16	83.27%
7	Bobby Peel	101	16.98	19.61	86.58%
8	Fred Spofforth	94	18.41	20.01	92.00%
9	Johnny Briggs	118	17.75	17.42	101.89%
10	Wilfred Rhodes	127	26.96	26.21	102.85%
11	Monty Noble	121	25	24.13	103.62%
12	Aubrey Faulkner	82	26.58	23.02	115.49%

I put the number of wickets too because most of them had fairly short careers. Only Ferris could beat Barnes and Lohman there.