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

[Flem274*]

Cowie h4x is bigger than ever.

came just to check on this.

came to this.

 

[Athlai]

Vettori redeemed

 

[Coronis]

Always interesting to see how players stack up against their peers.

Bradman>>>>>>>>>Hammond>>Headley is no surprise but Gavaskar>Chappell>Viv is.

Chappell has by far the highest standardised average of that lot, just played about five less years. If we included WSC, which he dominated, he'd probably be ahead of Sunny.

Viv never comes out well on these things unless people isolate peaks.

I'll pick Chappell over Viv anytime.

 

[ankitj]

How much did Nufan pay you? Siddle at 60 while Lillee is only 83 lol

Why is this methodology so harsh on Lillee? Even Sobers is ahead of him.

 

[the big bambino]

Seems some guys whose careers were sawn off by the wars have dropped out altogether which I think is unfair. Minor quibble with the pitch metric too as it seems it lets current conditions speak for every era of pitch preparation. This unduly favours Amar Singh and Nissar imo as home pitches in their era were quite fast. The SA pitches then were quite dead and in both instances a reversal of their contemporary characters.

 

[SillyCowCorner1]

Thinking of Cabbage after seeing Cribbage.

Like wtf?

 

[morgieb]

Why is this methodology so harsh on Lillee? Even Sobers is ahead of him.

Might have something to with his lack of Tests in Asia/WI, so only really played in friendly conditions.

Still his average increasing by 5 is mental given the era he played in wasn't that bowler friendly.

 

[morgieb]

It's also interesting to compare it to DoG's, which had him 9th.

 

[Prince EWS]

Seems some guys whose careers were sawn off by the wars have dropped out altogether which I think is unfair.

Yeah it's a tough one deciding who to include on the list. I'd just include everyone if I was just posting a big list of standardised averages, but given I give players a rank, including players who played for less than three years can I think discredit it somewhat as you can get some weird results with them.

Minor quibble with the pitch metric too as it seems it lets current conditions speak for every era of pitch preparation.

Nah, it doesn't. It lets them change over time. It does technically take into account every game in history, but they're all weighted for proximity/recency. For example, when determining the conditions in 1932 when Amar made his debut, a game played in 1932 would count for 1,392,670,187,228,260 times more than a game played today, making whatever happened today an statistical irrelevance when determining it. Games played a long time away from the period in question don't end up counting for anything meaningful except in cases of teams not playing each other for a long time (for example, some old India v Pakistan are still playing a non-zero role in determining how friendly India and Pakistan's home conditions are now, because they just haven't played any more recent games to over-ride them).

It seems it just thinks the pitches they played on were less friendly to quicks than you do, for reasons other than modern cricket. You may be right and it may be wrong, but it's not for that reason. It may even just be that they conceded a lot of runs in away games in some countries that were really unfriendly to pace, and those runs have now been slashed. I don't really know much about the intricacies of their careers to say either way, but I can look into it for you if you like.

 

[Prince EWS]

Might have something to with his lack of Tests in Asia/WI, so only really played in friendly conditions.

Home and away cricket being treated so separately from right near the start of this project has always hurt him given he played almost twice as many home games as he did away games, as has the importance of longevity and the fact that WSC -- a not insignificant part of how he formed his reputation -- is not included. Even just missing the longevity from his couple of seasons away playing WSC hurts him in a way.

Pitch conditions being added on just made things worse for him though.

Ultimately though, while a couple of these factors are more or less due to the limitations of the algorithm -- I definitely do think it under-rates him -- some of them to me just show that an average of 23 or so actually overstates how well he performed in Test cricket.

 

[trundler]

Maybe there's something to it. Tayfield is far ahead of Adcock. Pitches in SA were definitely more spin friendly than now back then.

 

[the big bambino]

@PEWS. No that’s all fine. Mine was only a minor point. My understanding of pitches then came from Indian writers who saw both players and the conditions at the time. From memory I don’t think there is a big difference in their home and away record. It would be interesting to see a pitch rating per country as you have done specifically for the era but that would be very difficult if not impossible.

I don’t mean to under rate them either, as both are favourites of mine. So in a way I’m happy to see them there. The one I can’t see on your updated list is Bowes. Did I miss him? He was about 70 on an earlier list. If he drops down or off altogether now can you say why?

 

[Prince EWS]

@PEWS. No that’s all fine. Mine was only a minor point. My understanding of pitches then came from Indian writers who saw both players and the conditions at the time. From memory I don’t think there is a big difference in their home and away record. It would be interesting to see a pitch rating per country as you have done specifically for the era but that would be very difficult if not impossible.

Yeah I can see why you'd think that but I'm sure whatever reason it has for doing it, modern cricket isn't it. Showing you this stuff is easy enough, takes 5 minutes or so to get these queries run but I just have to type in a month number and press go so it's not hard. I'm always happy to answer questions like this to help people get more of a feel for how it works.

So lets take January 1934:

Pace-friendliness:
1. Australia - 1.18
2. New Zealand - 1.17
3. South Africa - 1.08
4. England - 0.94
5. West Indies - 0.88
6. India - 0.85

Spin-friendliness:
1. New Zealand - 1.32
2. England - 1.21
3. West Indies - 1.05
4. India - 0.98
5. South Africa - 0.82
6. Australia - 0.81

Now just for a proof of concept, if we tell it that almost all cricket after 1954 didn't actually happen and ask it to do it again for the same month (January 1934), we get this:

1. Australia - 1.18
2. New Zealand - 1.17
3. South Africa - 1.08
4. England - 0.94
5. West Indies - 0.88
6. India - 0.86

Spin-friendliness:
1. New Zealand - 1.32
2. England - 1.21
3. West Indies - 1.05
4. India - 0.98
5. South Africa - 0.82
6. Australia - 0.81

Which as you can see is almost the exact same set of numbers. I needed to leave a couple of games in because not all match-ups between those six teams had happened yet by 1954, but even with that I deleted 99% of cricket after that date, and you can see that wasn't affecting those stats. So however it's deciding Amar and Nissar played on roads at home, it's not doing it because of stuff that happened way later.

The one I can’t see on your updated list is Bowes. Did I miss him? He was about 70 on an earlier list. If he drops down or off altogether now can you say why?

Yeah Bowes missed the list because he was deemed to have not have played three years worth of cricket. It seems odd given how long his career spanned, even ignoring the war years, but it seems he missed a lot of games for England, so he's ended up being credited with the equivalent of 2.52 years of continuous cricket.

Year	Bowes	England	Credit

1932	2	3	0.67
1933	1	9	0.11
1934	3	7	0.42
1935	4	9	0.44
1938	2	6	0.33
1939	2	6	0.33
1946	1	5	0.20

Total:			2.52

I'll create a separate sheet in the Google Doc with every player listed some time today, in case people want to check players who were excluded from rankings. They all still have standardised averages, I just didn't want to rank them if they'd played so little.

 

[trundler]

I've got a query if you don't mind.

Spin friendliness of England and South Africa in the 1950s.

 

[Prince EWS]

I've added a sheet with a full list of every player, including those not in the main sheet anymore (or ever), in case anyone had some special cases to look up. Down the button next to "Rankings", you can now select "Full List". It's actually sorted by total value (bat value + bowl value) but it's not ranked per se.

https://docs.google.com/spreadsheet...RJyxqnl8WU_sZvK5SbDe109OA/edit#gid=2016770811

This shows that Bowes's bowling average standardised from 22.34 to 20.91, a similar standardised average to Malcolm Marshall. With his 2.53 longevity he would've been 73rd on the bowling list had he qualified for a rank.

 

[Coronis]

Always interesting to see how players stack up against their peers.

Bradman>>>>>>>>>Hammond>>Headley is no surprise but Gavaskar>Chappell>Viv is.

I've added a sheet with a full list of every player, including those not in the main sheet anymore (or ever), in case anyone had some special cases to look up. Down the button next to "Rankings", you can now select "Full List". It's actually sorted by total value (bat value + bowl value) but it's not ranked per se.

https://docs.google.com/spreadsheet...RJyxqnl8WU_sZvK5SbDe109OA/edit#gid=2016770811

This shows that Bowes's bowling average standardised from 22.34 to 20.91, a similar standardised average to Malcolm Marshall. With his 2.53 longevity he would've been 73rd on the bowling list had he qualified for a rank.

Bradman gets 0 in bowling?! Blasphemy! One of his two wickets was Hammond!

 

[Prince EWS]

I've got a query if you don't mind.

Spin friendliness of England and South Africa in the 1950s.

I'm going to take a bit of a break now but I'll do this soon. Do you have an exact month in mind? Or a few of them, even?

 

[Prince EWS]

Bradman gets 0 in bowling?! Blasphemy! One of his two wickets was Hammond!

Haha I think I made everyone everyone with less than five wickets get 0 in bowling to stop players who took 2/15 and never bowled again having astronomical bowling values.

 

[trundler]

I'm going to take a bit of a break now but I'll do this soon. Do you have an exact month in mind? Or a few of them, even?

Let's say Jan 1954. Whenever you can.

 

[Prince EWS]

Gah, found a bug in my code that slightly affects away bowling. Easy to fix but it'll mean I have to edit my bowling post and the Google Doc. God damn you, easy to miss errors.