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

[morgieb]

Standardised average = 17.90 on the lastest update 13/1/19

Since then he has taken 21/225 in three test matches at an average of 10.7 bringing his average down from 24.51 to 21.64 so I guess his standardised average would be even lower now.

Yeah saw it now. So h4x as ****, essentially.

 

[Burgey]

And maybe read the post after that which has Siddle 30 places ahead of Lillee. I mean could you imagine Lillee being dropped for Siddle in 1975 vs WI lol

Wait, this shitshow has Siddle ahead of Lillee at all? And by 30 places? FMD it must be demoralising to put a lot of work into something like this, which is worthy of a lot of credit tbf, only to discredit yourself by getting this sort of outcome. Fix it ffs. It's embarrassing.

 

[Red]

I literally can't believe how people in this thread are trying to justify Lillee having a low ranking.

Richie Benaud
Dickie Bird
Ian Chappell
The cricinfo staff
Graham Gooch
BBC Sport (Ashes XI)

all have him in their all time XIs

 

[MartinB]

Possible explanation for Lillee's low rating

Like others I think there is a problem with the Lillee and other Australian bowlers of that era.

I do have a possible explanation -

In the period 1969 --> 1985

* Australian (top 6) Batting averages where 25% higher when Lillee plays compared to when he did not. Its 20% if you exclude WSC. That's like replacing Steve Waugh with Don Bradman.
* Bowling averages of many of the Australian Fast bowlers (Thomson, Walker, Lawson, Hogg) are 15% higher when Lillee plays.

I do not know if the boost to the batting average was caused by flatter wickets or something else. This would mean the Country pitch analysis used in the calculation is wrong (to bowler friendly) for Lillee (and other Australian bowlers of the era).

A second factor is during World Series cricket, batting averages in Australia where lower than normal. That will make Australia seem more bowler friendly than it really was.

References:

Batting average when Lillee plays

Batting Average with Lillee is not playing

 

[Prince EWS]

Four of the top six bowlers as per this analysis played in the 70s and/or 80s. Ambrose was more of a 90s guy but even discounting him, half the bowlers in the top six from a 20 year period is still significant. That's actually a huge chunk for a small period of Test history.

All rating systems are flawed. I think this one is less flawed than either raw averages or the off the cuff opinions of people who have watched less than 1% of the balls bowled in Test cricket, and yes part of that will definitely be the fact that it reflects my values (values being actual values rather than preconceived notions) and the things people who rate Lillee really highly value -- wickets per innings, aesthetics, defining their childhood, Ian Chappell picking him in his World XI, whatever else - aren't the things I do. Playing 70% of your games at home in conditions favourable to your bowling style and still only finishing with an average comparable and in some cases worse than the other great bowlers of your era are things I raise an eyebrow at. As someone who has gone on the record as saying Imran was the greatest Test cricketer of all time, I don't think I have some inbuilt anti-70s/80s bias...

MartinB's entire point was to try and identify where the flaw here was, not to say it wasn't flawed. I think Lillee is vastly over-rated but I think this analysis also clearly under-rates him, and MartinB has offered a great theory as to why. Lillee seems to have missed a lot of the home games in which bowling was easier for various reasons, so it's treating the pitches he bowled on as being a bit friendlier to him than they actually were. Ultimately I think WSC broke things for him a little bit. I've got no dramas with people pointing out where the flaws are (this exact thing has made me tweak and in some cases massively overhaul things when it's been something I agree is a flaw and the data is there to correct for it) and I've done my best to explain the methodology. I don't even any problems with people coming in here to check where their ten favourite childhood players finished, declare the whole thing useless and ignore it if they don't come out favourably... but if you're going to accuse me of stacking the deck and putting in years or work for some weird agenda that I've actually done a terrible job of pushing if you look at where Imran, Hadlee and Marshall finished up then you can **** right off. You're the one who opens this thread with an agenda each time.

 

[trundler]

Yeah, the incredible work here deserves a lot more respect here. Been a source of great nerdgasming for me lately. I've been watching old footage, reading old articles and using the database as my myth-busting guide into yesteryear. Generally, it does a great job of adjusting for pitches and eras.

 

[MrPrez]

Four of the top six bowlers as per this analysis played in the 70s and/or 80s. Ambrose was more of a 90s guy but even discounting him, half the bowlers in the top six from a 20 year period is still significant. That's actually a huge chunk for a small period of Test history.

All rating systems are flawed. I think this one is less flawed than either raw averages or the off the cuff opinions of people who have watched less than 1% of the balls bowled in Test cricket, and yes part of that will definitely be the fact that it reflects my values (values being actual values rather than preconceived notions) and the things people who rate Lillee really highly value -- wickets per innings, aesthetics, defining their childhood, Ian Chappell picking him in his World XI, whatever else - aren't the things I do. Playing 70% of your games at home in conditions favourable to your bowling style and still only finishing with an average comparable and in some cases worse than the other great bowlers of your era are things I raise an eyebrow at. As someone who has gone on the record as saying Imran was the greatest Test cricketer of all time, I don't think I have some inbuilt anti-70s/80s bias...

MartinB's entire point was to try and identify where the flaw here was, not to say it wasn't flawed. I think Lillee is vastly over-rated but I think this analysis also clearly under-rates him, and MartinB has offered a great theory as to why. Lillee seems to have missed a lot of the home games in which bowling was easier for various reasons, so it's treating the pitches he bowled on as being a bit friendlier to him than they actually were. Ultimately I think WSC broke things for him a little bit. I've got no dramas with people pointing out where the flaws are (this exact thing has made me tweak and in some cases massively overhaul things when it's been something I agree is a flaw and the data is there to correct for it) and I've done my best to explain the methodology. I don't even any problems with people coming in here to check where their ten favourite childhood players finished, declare the whole thing useless and ignore it if they don't come out favourably... but if you're going to accuse me of stacking the deck and putting in years or work for some weird agenda that I've actually done a terrible job of pushing if you look at where Imran, Hadlee and Marshall finished up then you can **** right off. You're the one who opens this thread with an agenda each time.

Your averages are biased. Morne Morkel should be number 1.

 

[srbhkshk]

Your averages are biased. Morne Morkel should be number 1.

What a load of crap, deserves no higher than 2.

 

[Dan]

The system's crap. Scott Kremerskothen doesn't even get a ranking!

 

[Burgey]

Four of the top six bowlers as per this analysis played in the 70s and/or 80s. Ambrose was more of a 90s guy but even discounting him, half the bowlers in the top six from a 20 year period is still significant. That's actually a huge chunk for a small period of Test history.

All rating systems are flawed. I think this one is less flawed than either raw averages or the off the cuff opinions of people who have watched less than 1% of the balls bowled in Test cricket, and yes part of that will definitely be the fact that it reflects my values (values being actual values rather than preconceived notions) and the things people who rate Lillee really highly value -- wickets per innings, aesthetics, defining their childhood, Ian Chappell picking him in his World XI, whatever else - aren't the things I do. Playing 70% of your games at home in conditions favourable to your bowling style and still only finishing with an average comparable and in some cases worse than the other great bowlers of your era are things I raise an eyebrow at. As someone who has gone on the record as saying Imran was the greatest Test cricketer of all time, I don't think I have some inbuilt anti-70s/80s bias...

MartinB's entire point was to try and identify where the flaw here was, not to say it wasn't flawed. I think Lillee is vastly over-rated but I think this analysis also clearly under-rates him, and MartinB has offered a great theory as to why. Lillee seems to have missed a lot of the home games in which bowling was easier for various reasons, so it's treating the pitches he bowled on as being a bit friendlier to him than they actually were. Ultimately I think WSC broke things for him a little bit. I've got no dramas with people pointing out where the flaws are (this exact thing has made me tweak and in some cases massively overhaul things when it's been something I agree is a flaw and the data is there to correct for it) and I've done my best to explain the methodology. I don't even any problems with people coming in here to check where their ten favourite childhood players finished, declare the whole thing useless and ignore it if they don't come out favourably... but if you're going to accuse me of stacking the deck and putting in years or work for some weird agenda that I've actually done a terrible job of pushing if you look at where Imran, Hadlee and Marshall finished up then you can **** right off. You're the one who opens this thread with an agenda each time.

What are we to make of this, do you think? Purely a coincidence or do you think it's a conditions-related thing, albeit you've standardized things? I suspect it's coincidence tbh, in that there happened to be a stack of really good/ great bowlers who came around form the mid-70s until the late 80s. Starting with Lillee and running through Imran, all the WI blokes, Hadlee etc.

Keeping that in mind, do you factor in there's a heap of good bowlers around, or do you (a generic you, not a specific you in this instance) say it's a bad batting era/ conditions favoured bowlers? And if a combination of both, where do you draw the line between them and try to standardize the stats?

So, for example, you'd hear Slats on the radio when he was doing it, and he'd be grumbling saying he played in an era where most sides had good bowlers and scoring runs was tough etc etc. And personally I think he's right - the 90s was a tough era with really good bowlers cf the 2000s. But in the 2000s you also had an era where there were a lot of flat decks, not least because of a huge drought in the southern hemisphere for more than half the decade. So to what extent is the 2000s marked down for batting and up for bowling owing to conditions? And to what extent because of the lack of great bowlers compared with the decade before?

 

[Burgey]

Four of the top six bowlers as per this analysis played in the 70s and/or 80s. Ambrose was more of a 90s guy but even discounting him, half the bowlers in the top six from a 20 year period is still significant. That's actually a huge chunk for a small period of Test history.

All rating systems are flawed. I think this one is less flawed than either raw averages or the off the cuff opinions of people who have watched less than 1% of the balls bowled in Test cricket, and yes part of that will definitely be the fact that it reflects my values (values being actual values rather than preconceived notions) and the things people who rate Lillee really highly value -- wickets per innings, aesthetics, defining their childhood, Ian Chappell picking him in his World XI, whatever else - aren't the things I do. Playing 70% of your games at home in conditions favourable to your bowling style and still only finishing with an average comparable and in some cases worse than the other great bowlers of your era are things I raise an eyebrow at. As someone who has gone on the record as saying Imran was the greatest Test cricketer of all time, I don't think I have some inbuilt anti-70s/80s bias...

MartinB's entire point was to try and identify where the flaw here was, not to say it wasn't flawed. I think Lillee is vastly over-rated but I think this analysis also clearly under-rates him, and MartinB has offered a great theory as to why. Lillee seems to have missed a lot of the home games in which bowling was easier for various reasons, so it's treating the pitches he bowled on as being a bit friendlier to him than they actually were. Ultimately I think WSC broke things for him a little bit. I've got no dramas with people pointing out where the flaws are (this exact thing has made me tweak and in some cases massively overhaul things when it's been something I agree is a flaw and the data is there to correct for it) and I've done my best to explain the methodology. I don't even any problems with people coming in here to check where their ten favourite childhood players finished, declare the whole thing useless and ignore it if they don't come out favourably... but if you're going to accuse me of stacking the deck and putting in years or work for some weird agenda that I've actually done a terrible job of pushing if you look at where Imran, Hadlee and Marshall finished up then you can **** right off. You're the one who opens this thread with an agenda each time.

With respect, this is a classic mistake to make. To say the things you rely on in basing your rankings are the actual values, and everyone else's aren't. Naturally from your perspective that's right and tbf you set out the factors which you rely on. But it's no more empirically the case that your parameters are right and anyone else's are wrong. So, for example, you rate longevity as a massive plus mark, so Walsh might be right up there in your estimation compared with another bowler who averages and strikes at lower rates but over a shorter time. Why is that the value as opposed to what others put front and centre?

We see this sort of thing with the ODI batsman thread - these idiots rating Tendulkat, Kohli, de Villiers so high because they had a "peak". But they didn't do anything when it matters most, so they should be rated lower imo. You don't see these people saying their view is the definitive value to place on an ODI career.

Edit: TBF I do say that WCs are all that matter, and I've definitively proven that's the objective truth, so maybe this isn't the best analogy after all.

 

[Prince EWS]

With respect, this is a classic mistake to make. To say the things you rely on in basing your rankings are the actual values, and everyone else's aren't. Naturally from your perspective that's right and tbf you set out the factors which you rely on. But it's no more empirically the case that your parameters are right and anyone else's are wrong. So, for example, you rate longevity as a massive plus mark, so Walsh might be right up there in your estimation compared with another bowler who averages and strikes at lower rates but over a shorter time. Why is that the value as opposed to what others put front and centre?

Oh don't get me wrong, I'm not saying they're the only true values. It's totally valid to value different things. My rating system reflects what I value personally. My point was just to differentiate that I value weighted performance relative to one's peers and longevity, rather than just having some preconceived notion of who the best players are and stacking the deck to get them on top (which is straight up what I was being accused of). My results have actually changed my opinions on a couple of close calls between players a few times, and I'm always going to trust them more because I know they represent the things I value. There are still some flaws in it even from that perspective as there will be in any attempt at a ubiquitous method for rating all players.

 

[Prince EWS]

What are we to make of this, do you think? Purely a coincidence or do you think it's a conditions-related thing, albeit you've standardized things? I suspect it's coincidence tbh, in that there happened to be a stack of really good/ great bowlers who came around form the mid-70s until the late 80s. Starting with Lillee and running through Imran, all the WI blokes, Hadlee etc.

Keeping that in mind, do you factor in there's a heap of good bowlers around, or do you (a generic you, not a specific you in this instance) say it's a bad batting era/ conditions favoured bowlers? And if a combination of both, where do you draw the line between them and try to standardize the stats?

So, for example, you'd hear Slats on the radio when he was doing it, and he'd be grumbling saying he played in an era where most sides had good bowlers and scoring runs was tough etc etc. And personally I think he's right - the 90s was a tough era with really good bowlers cf the 2000s. But in the 2000s you also had an era where there were a lot of flat decks, not least because of a huge drought in the southern hemisphere for more than half the decade. So to what extent is the 2000s marked down for batting and up for bowling owing to conditions? And to what extent because of the lack of great bowlers compared with the decade before?

Yeah this is basically the first thing I tried to get this system to take into account. People often grumbled about how much easier run scoring was in the early 2000s so I decided to create a system that altered players averages based on how easy scoring was overall in their era. As time went on I made it more complicated - I looked at which teams players had played against and how easy run-scoring / wicket taking was against them at the time, put in a home / away split, narrowed down what an era was in a better way, split up pace and spin bowling, created a way to roughly measure pitch conditions in each country, etc etc. But the Slater point was more or less exactly what this thing was designed to address originally, and it still does that better than it deals with all those other things I mentioned.

 

[stephen]

I don't know why people are picking on Siddle being ranked bigger than Lillee. Srinath and Brett Lee are far more egregious examples in this list.

Still a worthwhile exercise given the constraints.

The main problem anyone has when rating cricketers against their peers is that there's no good, objective measure of the quality of their peers. Were the 80s Windies side actuality really bad but all their peers were amateur, or were they in fact the strongest side in what otherwise would have been the best era of batting in history? We can have an idea having watched them play but the reality is that it's impossible to say for sure.

But that's why any rating of cricketers is subjective - after all, the Australian hierarchy rate Bancroft highly while people outside that clique don't rate him at all.

 

[Burgey]

Yeah fair enough. I apologise if I sounded glib criticising what’s obviously take a lot of time and effort. It’s not dissimilar to DoGs innings/ bowling spell ratings. May not agree with everything it throws up but admire the effort and it promotes discussion, which can only be a good thing

 

[Burner]

Dennis Lillee at 83 seems about right to me. Great list.

 

[trundler]

Top 100 is generous

 

[GoodAreasShane]

Love how Lyon comes out a fair way ahead of Swann. Best thing about the whole list imo.

 

[Prince EWS]

Love how Lyon comes out a fair way ahead of Swann. Best thing about the whole list imo.

Haha he'd be even further ahead with a more recent update. Could end up in the top 20 by the time he's done at this rate.

 

[silentstriker]

I think it’s interesting. In real life, how much joy (or hatred) a player brings to you certainly has a big effect on how you will rate them. It’s human nature. On top of that, we all have biases and preconceived notions that will make us either over or underrate players based on various qualities (left handers, how “ugly” their technique is, whether they have a punchable face, whether they are being ranked against our favorite player, etc).

All automatic ranking systems have flaws, but people’s subjective rankings always have more. I’ve created ranking systems in the past (http://www.cricketweb.net/forum/cri...tings-statistical-analysis-silentstriker.html) - yours is definitely better. People like certain players and then create arguments to “prove” how they are better (or rig their rankings to do so) - these systems do have the benefit of objectivity (though not “truth” - as others have mentioned, because in the end, it is still based on assumptions).