Most Valuable Players

Okay, so I will get back to the 1000 words idea (at some point—I’m somewhat of a dilettante), but this is what has grabbed my interest at the moment.

The sport I’m most interested in is cricket. If you have no interest in the sport, feel free to ignore this post. 🙂

It’s another intellectual pursuit, I guess, as I was never a particularly good player, but I’ve long felt that the statistics and measurements used to define players leave a lot to be desired. The difference is particularly stark in comparison to its cousin, baseball (compare the Wikipedia pages for baseball stats and cricket stats, even just by length). For example, there are no records related to fielding beyond “number of catches taken”, which is only part of the story. Where were they fielding? How many catches did they not take? How many matches have they played?*

I’m encouraged that some new measures are emerging, like “control percentage” (e.g. how often did a batsman play the ball cleanly), but I also feel that some of the existing measures could use some adjustment. For example, a player’s batting average (total runs / number of times dismissed) can be warped by large numbers of “not-outs”; in extreme cases this can lead to the farcical situation of a player’s average score being higher than their high score. Additionally, all runs are not created equal—a score of 42 in a tight, low-scoring game may be more valuable to the team than a score of 67 in a run-fest that ends in a draw.

In order to better analyse, one must first have data, and fortunately there are a lot of publicly-available scorecards that could be used to delve into various parameters (though again, ignoring some useful measures, but it’s a start). For now, though I’ve examined “Player of the Match” awards. The existing data only gives a ranking for how many times a player has won the award (see here for example). As such, the “ranking” tells you more about longevity** than value, so rather than a raw count, I looked at matches per award for each player***.

While any informed fan (myself included) wouldn’t quibble with the calibre of the names atop the previously-linked list, my calculations yielded a different ordering (though, unsurprisingly, a lot of the same names). Here’s the top 20****:

  1. Vernon Philander (SA) 5.2
    (5 awards in 26 matches)
  2. Wasim Akram (Pak) 6.12
    (17 awards in 104 matches)
  3. Daryl Tuffey (NZ) 6.5
    (4 awards in 26 matches)
  4. Mitchell Johnson (Aus) 6.56
    (9 awards in 59 matches)
  5. Muttiah Muralitharan (SL) 7.0
    (19 awards in 133 matches)
  6. (Sir) Curtly Ambrose (WI) 7.0
    (14 awards in 98 matches)
  7. Jacques Kallis (SA) 7.22
    (23 awards in 166 matches)
  8. Irfan Pathan (India) 7.25
    (4 awards in 29 matches)
  9. Joe Root (Eng) 7.33
    (3 awards in 22 matches)
  10. Kumar Sangakkara (SL) 8.0
    (16 awards in 128 matches)
  11. Imran Khan (Pak) 8.0
    (11 awards in 88 matches)
  12. Stuart Clark (Aus) 8.0
    (3 awards in 24 matches)
  13. Malcolm Marshall (WI) 8.1
    (10 awards in 81 matches)
  14. Rangana Herath (SL) 8.14
    (7 awards in 57 matches)
  15. Dale Steyn (SA) 8.33
    (9 awards in 75 matches)
  16. Aravinda de Silva (SL) 8.45
    (11 awards in 93 matches)
  17. (Sir) Ian Botham (Eng) 8.5
    (12 awards in 102 matches)
  18. Shakib Al Hasan (Ban) 8.5
    (4 awards in 34 matches)
  19. Shane Warne (Aus) 8.53
    (17 awards in 145 matches)
  20. Dean Jones (Aus) 8.67
    (6 awards in 52 matches)

It’s not ground-breaking or anything, but I feel it’s an interesting start.

* Compare two players who have taken 10 catches. Player A has played 7 matches, and dropped 1 catch. Player B has played 40 matches and dropped 12 catches. Which seems to be the better fielder?

** Not to mention that the award was not given out for every match; any ranking of these awards is going to favour players in the modern era.

*** If all players were equally valuable, you would expect (on average) any given player would win an award every 22 matches (two teams of 11 in each match). Having a score lower than this indicates a more valuable player.

**** To prevent outliers I removed players who have played less than 22 matches.