Cribbage's Standardised Test Averages (UPDATED November 2018 - posts 753-755)

[Flametree]

Would it be fair to say players get penalised if they play in a successful side, and promoted if they play in a less good side? Seems to me some Aussies (Chappell G, Walters, for instance) are far lower in the list than I would have expected given they played in an era against good bowlers. Is it just because with Redpath, Chappell I, Edwards, Stackpole, McCosker, Border etc in their team, their own contributions were less "important" towards team success?

I'm a kiwi and Sir Richard is of course near God to us, but even I wouldn't have him the second best bowler ever. He does well on the longevity of course. But Marshall, Ambrose, McGrath, Trueman played in sides with far better bowling teammates, so their own contribution was more likely to be less significant.

 

[Howe_zat]

Is the fact that Hadlee often had to carry the attack not to his credit, though?

What this essentially gives us is a ranking of the most valuable players, not necessarily the best - though Cribb iirc views those as more or less synonymous - and one would be hard pushed to think of bowlers more valuable to their sides than Hadlee and Murali.

 

[HeathDavisSpeed]

Peter Such criminally low in the bowling ranking.

 

[Prince EWS]

Would it be fair to say players get penalised if they play in a successful side, and promoted if they play in a less good side? Seems to me some Aussies (Chappell G, Walters, for instance) are far lower in the list than I would have expected given they played in an era against good bowlers. Is it just because with Redpath, Chappell I, Edwards, Stackpole, McCosker, Border etc in their team, their own contributions were less "important" towards team success?

I'm a kiwi and Sir Richard is of course near God to us, but even I wouldn't have him the second best bowler ever. He does well on the longevity of course. But Marshall, Ambrose, McGrath, Trueman played in sides with far better bowling teammates, so their own contribution was more likely to be less significant.

It doesn't really work like you think it does, especially with the batting. There's absolutely no reason at all why being in a good batting side would hurt you in this analysis. Chappell actually has a higher standardised average than Tendulkar but is relatively low down on the list due to a lack of longevity, owing partly to World Series Cricket not counting. It's purely a set of Test rankings, so a player who goes off and misses part of his career to play in a rival league, no matter how noble the motives, is going to be of less use to a Test side than one who does not.

Hadlee is high on the bowling list for the opposite reason; his longevity was excellent. In real terms there's pretty little difference in what Hadlee brought to the table with the ball and what Marshall did, even if one is to concede that Marshall's average performance was a tiny fraction better when he was on the park (which this analysis would agree with, matter of fact), but doing it for 15 years rather than 9 is quite an advantage. If I was the captain of a team and was offered Hadlee's bowling career or Marshall's bowling career from start to finish I'd take Hadlee's and that's what the analysis is based around. Not that he was more valuable to New Zealand than Marshall was to West Indies, or Heath Streak would feature in the top 20, but this his career would be more valuable outright to any unknown team than Marshall's due to its longevity.

 

[Flametree]

Is the fact that Hadlee often had to carry the attack not to his credit, though?

True, though you can probably argue it both ways. He got preferential treatment (not as much as he'd have liked, of course!) So he got to choose ends for instance. And when they got to the tail he was thrown the ball a lot more than some bowlers in sides with more options.

Prince's explanation makes sense however, Hadlee played continuously for 18 years missing only a couple of series, and was pretty much excellent for about 16 of those.

And back to Prince - sorry, you've probably answered this question (more than once) in the earlier posts in the thread, but I can't face reading through the other 440 posts to find it. How come Herbie (was it?) Taylor rates so highly then despite an average of 40? Is it not in some way because he was a good player in a bad team?

 

[Flem274*]

So how did you get around Jack Cowie in this one Cribblett? (Where did he come btw? I couldn't find him even with cntrl + f.)

Lets hope cricinfo doesn't release an article using that formula now.

Well now you've written it, copyright laws dictate it's yours, so in a perfect world they can't.

In the real world they have more expensive lawyers than you.

Lastly

503	JS Patel (NZ)		13	40	48.40	47.09	1.56	0.276

 

[Prince EWS]

So how did you get around Jack Cowie in this one Cribblett? (Where did he come btw? I couldn't find him even with cntrl + f.)

Getting around Cowie was easy; I solved that problem several editions ago.

New bowling rankings (qual 10 Tests and 20 wickets):

...

Cheating a bit, but there you go.

(I actually don't think he'd be a problem with weldone's new longevity formula anyway)

 

[Prince EWS]

True, though you can probably argue it both ways. He got preferential treatment (not as much as he'd have liked, of course!) So he got to choose ends for instance. And when they got to the tail he was thrown the ball a lot more than some bowlers in sides with more options.

Prince's explanation makes sense however, Hadlee played continuously for 18 years missing only a couple of series, and was pretty much excellent for about 16 of those.

And back to Prince - sorry, you've probably answered this question (more than once) in the earlier posts in the thread, but I can't face reading through the other 440 posts to find it. How come Herbie (was it?) Taylor rates so highly then despite an average of 40? Is it not in some way because he was a good player in a bad team?

It's not. It's because he played in a bowler-dominated era and scored lots of runs against good bowling attacks. His averages standardises out to 47.

 

[Flametree]

It's not. It's because he played in a bowler-dominated era and scored lots of runs against good bowling attacks. His averages standardises out to 47.

Mmm. The 1920s wasn't any more bowler-dominated than any other decade. In fact, I'd suggest batsmen averaged more in that decade than pretty much any other till recent times (ignoring the 1940's). Not very evenly spread of course - England and Australia scored a lot, South Africa not so much, and the Windies even less.

 

[Prince EWS]

Mmm. The 1920s wasn't any more bowler-dominated than any other decade. In fact, I'd suggest batsmen averaged more in that decade than pretty much any other till recent times (ignoring the 1940's). Not very evenly spread of course - England and Australia scored a lot, South Africa not so much, and the Windies even less.

Well then it'll be down to him scoring all his runs against the good England/Australia attacks, probably away from home.

It's not about how good the rest of his team was, but how good the bowlers he was facing were.

 

[HeathDavisSpeed]

Bill O'Reilly's a fair chunk ahead of Curtley Ambrose despite having a lower longevity factor and higher averages. Find that a little odd.

 

[Prince EWS]

Bill O'Reilly's a fair chunk ahead of Curtley Ambrose despite having a lower longevity factor and higher averages. Find that a little odd.

That'll just be wickets per match.

 

[HeathDavisSpeed]

That'll just be wickets per match.

Seems interesting to use wickets per match when Ambrose's Strike Rate is far statistically significant in its superiority. I also find it odd that the 3 day 'Test' Match vs. NZ in 1946 is included in his stats when so many similar games are excluded from the stats of players such as Richie Benaud, Bert Sutcliffe et. al.

 

[Flametree]

Well then it'll be down to him scoring all his runs against the good England/Australia attacks, probably away from home.

It's not about how good the rest of his team was, but how good the bowlers he was facing were.

I'm not trying to pick an argument, seriously I'm genuinely curious. The bowlers he was facing would have played around 1/3 of their matches against South Africa. So I guess the bad performance of his teammates did in a roundabout way improve his rating, though I don't know how you can escape the circular loop.

For instance, Taylor would have played in matches against George Geary for England. Geary averaged 36 against Australia, and 20 against South Africa. Overall 29. Scoring runs against an attack containing him presumably improves Taylor's adjusted average, slightly since Geary's average is less than the 31 "norm". But if it weren't for the inadequacies of his teammates, it might reduce the adjusted average.

I guess this is only a problem back in the earlier years when there were so few test playing nations.

 

[Prince EWS]

Seems interesting to use wickets per match when Ambrose's Strike Rate is far statistically significant in its superiority. I also find it odd that the 3 day 'Test' Match vs. NZ in 1946 is included in his stats when so many similar games are excluded from the stats of players such as Richie Benaud, Bert Sutcliffe et. al.

Wickets per match is needed to stop a theoretical batsman who plays for 15 years and takes 20 wickets behind credited as one of the best bowlers of all time. Tendulkar's longevity and not-absolutely-awful standardised average would put him in batting allrounder territory if not for that qualifier.

As far as what's included - if cricinfo thinks it's a Test, then it's a Test, unless it wasn't between two Test nations (like the SuperTest).

 

[Prince EWS]

I'm not trying to pick an argument, seriously I'm genuinely curious. The bowlers he was facing would have played around 1/3 of their matches against South Africa. So I guess the bad performance of his teammates did in a roundabout way improve his rating, though I don't know how you can escape the circular loop.

For instance, Taylor would have played in matches against George Geary for England. Geary averaged 36 against Australia, and 20 against South Africa. Overall 29. Scoring runs against an attack containing him presumably improves Taylor's adjusted average, slightly since Geary's average is less than the 31 "norm". But if it weren't for the inadequacies of his teammates, it might reduce the adjusted average.

I guess this is only a problem back in the earlier years when there were so few test playing nations.

Everyone else who played against Geary would be credited the same though. Fact is, England and Australia were better than South Africa, so scoring runs/taking wickets against them was more difficult and it has been credited as such. The fact that they were batting at both batting and bowling may slightly exaggerate how much better they were at either individually, I suppose, but I don't think it's a big factor.

 

[HeathDavisSpeed]

Wickets per match is needed to stop a theoretical batsman who plays for 15 years and takes 20 wickets behind credited as one of the best bowlers of all time. Tendulkar's longevity and not-absolutely-awful standardised average would put him in batting allrounder territory if not for that qualifier.

As far as what's included - if cricinfo thinks it's a Test, then it's a Test, unless it wasn't between two Test nations (like the SuperTest).

Well, maybe the *impact* is slightly too high. It seems odd to me that Ambrose, who played more tests, with a better average and better strike rate would be behind O'Reilly. Not to disparage O'Reilly, but it just doesn't 'feel' right to me.

 

[Prince EWS]

Well, maybe the *impact* is slightly too high. It seems odd to me that Ambrose, who played more tests, with a better average and better strike rate would be behind O'Reilly. Not to disparage O'Reilly, but it just doesn't 'feel' right to me.

Yeah, fair enough.

I do like that it rewards bowlers who can bowl large percentages of their team's overs though; great spinners typically have slightly higher averages than great quicks but can contribute by bowling more overs and having a greater overall impact on their team's attack. Ambrose was in a bit of a unique position where if he was able to bowl a larger percentage of his teams overs it wouldn't really have improved matters even though he was an ATG, due to the quality of the other bowlers, but that's not usually the case.

Either way I'll have a look at making that count for less. Square rooting or something, perhaps.

 

[Jarquis]

Have noted some personal observations, which I found quite awesome for rankings based on cricket statistics. Will share that tommorrow when I get some time.

 

[The Sean]

Seems interesting to use wickets per match when Ambrose's Strike Rate is far statistically significant in its superiority. I also find it odd that the 3 day 'Test' Match vs. NZ in 1946 is included in his stats when so many similar games are excluded from the stats of players such as Richie Benaud, Bert Sutcliffe et. al.

I imagine it's included because it is actually an official, recognised Test.

Though its inclusion is also probably one reason why O'Reilly's standardised average is fractionally higher than his official Test average - which seems somewhat counter-intuitive given the batting-friendly era that the Tiger played in.