Test batsmen and bowlers peaks. Criteria: At least 5 years and at least 40 innings

[Days of Grace]

I have settled on a method to analyze test batsmen and bowlers peaks. The criteria is at least five years AND 40 innings. It can be longer, of course. The goal is to find the best averages for each batsman and bowler within the set criteria. Unlike 40 innings or five calendar years, which are easy to calculate, five years and 40 innings involves sifting through a lot of data on cricinfo. So far for batsmen I have:

Bradman (29/12/1928-24/6/1938): 103.23
Sobers (26/2/1958-6/6/1963): 78.53
Smith (1/3/2014-4/9/2019): 77.11
Richards (23/1/1976-27/3/1981): 74.72
Ponting (16/8/2001-1/12/2006): 74.01
Walcott (15/2/1952-13/3/1958): 72.78
Hutton (5/3/1949-30/3/1954): 69.88
Tendulkar (2/1/1997-21/2/2002): 68.02
Barrington (25/1/1963-8/2/1968): 67.87
Hobbs (1/1/1910-28/6/1924): 65.65
Lara (15/12/2000-9/3/2006): 61.32

Imran Khan (22/3/1982-4/6/1987): 16.21
Marshall (28/4/1983-30/6/1988): 18.29

My request is for some help in finding the best peaks for both batsmen and bowlers according to the above criteria. Post your findings below and thanks in advance! See if you can improve on my above findings. I will continue to post more batsmen and bowlers in this original post.

 

[Coronis]

Ok well the easy ones

Pollock (6/12/1963-5/3/1970): 62.55
Headley (11/1/1930-15/1/1954): 60.83

And some others (watch me be completely wrong about their peaks)

Sutcliffe (14/6/1924-15/6/1929): 64.67
Hammond: ;4/2/1928-31/3/1933): 69.58

 

[Burgey]

I think Ponting's 52-test peak average was 75 odd.

 

[Migara]

Kumar Sangakkara ( 04 Aug-2006 to 26-Aug-2009) 77.08

 

[ataraxia]

Kumar Sangakkara ( 04 Aug-2006 to 26-Aug-2009) 77.08

5 years really goes by so quick ey

 

[Migara]

5 years really goes by so quick ey

I counted 5 years or 40 innings BTW

 

[ZK$]

I counted 5 years or 40 innings BTW

I think it has to be both.

 

[Days of Grace]

That's right. It has to be both. I'm trying to make it as fair as possible for past and modern players.

 

[Prince EWS]

That's right. It has to be both. I'm trying to make it as fair as possible for past and modern players.

I reckon I could automate this entirely in an about an hour once I get back to Tea Gardens and have access to my database. The last 18 months or so of Tests wouldn't count though, and I probably won't be able to get back home for a couple of weeks.

 

[Bolo.]

What format do you have the data in? I think I have seen it kicking around in Excel on CW.

If Excel, can't you just filter by however you selected the best 5+ years, and then put in an AND inningsCOUNT>= 40? Should just return null if failing, filter by highest.

Not that I know dick about coding, but if you are running it in a programming language, it should work in a very similar way.

 

[Days of Grace]

All great suggestions but I want the best average as long as it is five years or more and 40 innings or more. For instance, a batsmen's best average may have been achieved over 7 years and 60 innings, although that would be rare.

 

[Prince EWS]

All great suggestions but I want the best average as long as it is five years or more and 40 innings or more. For instance, a batsmen's best average may have been achieved over 7 years and 60 innings, although that would be rare.

Yeah, I can definitely do that in an hour or two, but not for a couple of weeks.

 

[Prince EWS]

As you can see from this post, I basically already have the code for it:

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

Odd, Sutcliffe's overall standardised is 62.79 yet his peak average is 59.41? Hill and McCabe big (and welcome) surprises there. This could be due to a bug I haven't found yet, but it's possible for it to be true without that. For example, take this four number long list: 10, 10, 1, 10 The...

www.cricketweb.net

That was using standardised averages rather than scorebook ones, but the scorebook ones are in the database too. I'd just need to change one value in the code to use those, and also set it to ignore anything that was less than 40 innings (also probably like one line, lol). It would not be hard, and I could give you every player's peak ever (save for a few whose peak included the last 18 months I guess). Sutcliffe and Weekes on that second list prove it's capable of handling the issue you're worried about it missing, @Days of Grace.

 

[Patience and Accuracy+Gut]

As you can see from this post, I basically already have the code for it:

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

Odd, Sutcliffe's overall standardised is 62.79 yet his peak average is 59.41? Hill and McCabe big (and welcome) surprises there. This could be due to a bug I haven't found yet, but it's possible for it to be true without that. For example, take this four number long list: 10, 10, 1, 10 The...

www.cricketweb.net

That was using standardised averages rather than scorebook ones, but the scorebook ones are the database too. I'd just need to change one value in the code to use those, and also set it to ignore anything that was less than 40 innings (also probably like one line, lol). It would not be hard, and I could give you every player's peak ever (save for a few whose peak included the last 18 months I guess). Sutcliffe and Weekes on that second list prove it's capable of handling the issue you're worried about it missing, @Days of Grace.

Where would Steve Smith be?

 

[Days of Grace]

As you can see from this post, I basically already have the code for it:

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

Odd, Sutcliffe's overall standardised is 62.79 yet his peak average is 59.41? Hill and McCabe big (and welcome) surprises there. This could be due to a bug I haven't found yet, but it's possible for it to be true without that. For example, take this four number long list: 10, 10, 1, 10 The...

www.cricketweb.net

That was using standardised averages rather than scorebook ones, but the scorebook ones are the database too. I'd just need to change one value in the code to use those, and also set it to ignore anything that was less than 40 innings (also probably like one line, lol). It would not be hard, and I could give you every player's peak ever (save for a few whose peak included the last 18 months I guess). Sutcliffe and Weekes on that second list prove it's capable of handling the issue you're worried about it missing, @Days of Grace.

That would be brilliant, mate.

I’m also going to standardize the averages but I need to know where to look in a player’s career first.

 

[Migara]

All great suggestions but I want the best average as long as it is five years or more and 40 innings or more. For instance, a batsmen's best average may have been achieved over 7 years and 60 innings, although that would be rare.

Unlikely. Smaller the sample higher the variance. So high values are higher in smaller samples and low values are lower.

 

[Coronis]

Unlikely. Smaller the sample higher the variance. So high values are higher in smaller samples and low values are lower.

He literally said it would be rare in the post you quoted.

 

[Everton Seymour]

How about AB De Villiers?

10/1/2008 - 20/2/2014

90 Test Innings - 16 Test Hundreds , 22 Test 50's
5013 Test Runs @ 65.10

Also this peak coincides with one of the greatest periods ever in South African Test cricket

SA reach World No.1 in the rankings for the first time.
South Africa win their first ever test series in England in 08 ,AB got a brilliant 177 at Headingley.
First ever series win in Australia , they go on to win another series in England and Australia in 2012
SA's highest ever 4th innings successful chase 414 @ the WACA with AB getting a hundred ..
A maiden double hundred in India for De Villiers
Briefly holds the highest Test score by an SA batsman with 278 not out in Abu Dhabi in 2010
Unbeaten Away from home in this period aswell ..

It's quite a superb peak for a Test batsman that doesn't get talked about alot .

 

[Days of Grace]

Good find! Does that average become higher if you narrow the time period to five years instead of six?

 

[BazBall21]

Dravid (18/11/2000-04/07/2006): 66.22