Top 25 batsmen all time - ranked!

[Stonigut]

Must have played at least 16 tests in career. Combines test average and 1st class averages, as all eras of batsmen have played both types of cricket and also considers the ratio of test centuries scored per test played as the greatest test of mettle. Test average carries 65% weight and 1st class average 35% to drive the base average number.

Here are the results.

Rank Name Country 1 Bradman Aus 2 Headley WI 3 Walcott WI 4 Weekes WI 5 Pollock SA 6 Sutcliffe Eng 7 Sobers WI 8 Ponting Aus 9 Kallis SA 10 Tendulkar Ind 11 Hammond Eng 12 G. Chappell Aus 13 AD Nourse SA 14 Gavaskar Ind 15 Lara WI 16 Hutton Eng 17 Hobbs Eng 18 Barrington Eng 19 Sehwag Ind 20 Crowe NZ 21 Paynter Eng 22 Rahul Ind 23 Richards WI 24 Miandad Pak 25 Waugh Aus

[Gamma]

Surprised Sir I V A Richards, Lara and Sachin are so far down the line

[Islandman]

Well, we can clearly see that the results do not reflect reality.

[Stonigut]

Those are the hard facts, can't argue round it, can't deny it, can't do nothing but sey, that is that. At the end of the day that is what is left, all else is just hot air.

What is interesting to me is the number of west Indians on it, I was really surprised, especially at weekes and Walcott.

[Islandman]

Not arguing the maths, it is what it is. How you interpret it now, that is something else.

It certainly does not translate to a ranking of their value, ie "if I was picking a team, who would I choose first? "

[Islandman]

Viv did not surprise me that much. He had a lot of low scores especially in his last few years when his skills no longer matched his ego.

[Don1]

LOL... you regard this presentation as so definitive...why??

Bredrin di way yuh put yuh truss inna numba system smaddy invent yuh muss be one statistician.

Numbers can be manipulated to establish any claim or suit any agenda. Just change the all important "assumptions" slightly and the rankings are completely different. The assumptions are not sacrosanct

The assumptions should be the subject of critique as much as the rankings they produce

This is the identical issue I had with your player selection by statistics

[Stonigut]

Just the facts measuring across time in the same core areas.
This is one of the weaknesses of our football infrastructure little measurement and measurement is one of the definitive characteristics of successful entities whether companies, countries or any very successful organization,they have well defined goals,well defined measurement systems, well defined improvement systems to ensure continuity.

What it does it sets the table for all concerned to know what them must to do to move to the next level, as you say it is the performance at the next level which then defines stick ability.

Is it any wonder that the two sports we have had the most success in are measurement oriented in every way, track and cricket and we have actually even fully pulled in the system culture of measurement in these sports into the Wider culture, not so On the ball side.

As island an say, if him was picking a team then it would look different, all I am saying is based on the performances this is the hard ranking, no way around Bradman being number one no matter how you change the measuring points.

[Gamma]

that brad is # is indisputable using any criteria .... the order of the rest ...

[Islandman]

I do agree fully that in sports and in our society in general Jamaicans need to focus more on data and measurement.

Have you read the book Moneyball about the Oakland A's and how they used such techniques? Haven't read it yet myself, its on my short list though.

[Exile]

These rankings are all very subjective and depends on how different numbers are weighted. For example how did they arrive at the 65%:35% for Tests:first class? How did this bias this in favour of some players? Why not 80:20 or or why not rank Test players stats vs 1st class?
The other subjective thing is the opponents played and where.
Sorry, I will need some more detailed info on how the rankings were arrived at. What about players who scored double and triple centuries? A feat not easily repeated or even done.

[Stonigut]

I saw the movie, pretty good. For me I have been using stats for everything from improving processes at work, to measuring employee performance, using it to "incentivize" employees thus getting then to consistently beat goals, pretty much done that at every job for last 10 years always with excellent results. Led multiple teams to top region, top country, worldwide finals in process improvement competitions at a Fortune 100 company, so I fully believactuator power of measurement, competition and especially in shame, if you don't get them through competitiveness you get them through shame bdl! actually I was oe of those guys while at Calabar who would run the stats all the way through based on times, performance seeing what others were doing to predict if we would win champs or not, it worked well even from then.

[Exile]

Here are some Test Stats:

http://stats.espncricinfo.com/ci/con...ds/223646.html

[Exile]

..and highest career batting averages...for 20 Test min.:

http://stats.espncricinfo.com/ci/con...ds/282910.html

[Stonigut]

65 35 as the weighting of test should exceed that of 1st class, we can do it anyway you want, but bottom line test is more important than 1st class, what was interesting is that their averages for each area were not significantly different, so that is not an issue. In terms of opponent played that also is not really an issue over a career timeperiod, what we could throw in to make it interesting is the winning percentage of their team over the time they played but that is a ton of work as the stats don't lump that together nicely fitting one players career.