I thought it was interesting and therefore I should share it with you. Some of you might already know about this. A new metric has been introduced in cricket called the pressure index (PI). I came across this index in the rediff.com page while looking at the scores for the first World Cup 2007 match between WI vs Pak.
I was curious to know what it meant. I clicked on it and it opened up the pressure index for the Pakistan team as 134/200. Then it also gave me the link to the article about the index. It seems like a good index to see where the match is tilted at any given moment. Index ranges from 0-200. 0 would mean near absolute victory and 200 near absolute loss. It is calculated only when the second team comes into bat. So if a chasing team’s PI is 125 then conversely the defending team’s PI is 75. Professor Rajeeva L Karandikar, the well-known statistician and psephologist, has developed the mathematical theory of the pressure index. Professor M J Manohar Rao, who also developed the Rediff Test and ODI Rankings and Rediff's Most Valuable ODI Player Index, originally conceived the idea of a pressure index. Take a look at the following articles.
http://www.rediff.com/wc2007/2007/mar/08pressure.htm
http://www.rediff.com/wc2007/2007/mar/13pressure.htm
2 comments:
Somehow, I am not very enthused by this idea. First of all, I am not convinced abt the naming convention used. It surely does not represent the "pressure" felt either by the players or by the spectators.
Moreover, I feel that they are trying to take into account too many subjective factors to quantify the chances of a team.
1. The idea of asking experts for an opinion on what would represent a good chasing score is not a great start to quantifying anything. Though they say that the weight of that score reduces as game progresses, I am not sure if the experts will predict a close score all the time. There are many instances where the commentators look at the pitch, and say that its a 300 pitch, and all of a sudden they struggle to score 200.
2. I dont know if they take into account the form the current player is in. I feel that that should be a factor, and if it is, it cannot be quantified.
3. If Dhoni and Dravid are batting in the 10th over, the price of Dravids wicket is more than that of Dhoni, meaning Dravid's wkt would increase the P.I more than that of Dhoni's. However, in the 45th over, Dhoni's wkt is more important. This, again, is very subjective, and I am not sure how they quantify these things if they take them into account.
3. Do they take into account who's overs are left from the bowling side? How will they rate different part-time bowlers that are bowling? If a great bowler has given 60 runs in his 6 overs, and a part-timer has given 30 runs in 6 overs, how will the P.I change if one of these bowls the last over? Again, I feel these are all too subjective to be quantified.
Their comparison to D/L method is wrong, because D/L just goes by statistics and nothing subjective. If "P.I is not dissmilar to D/L" as they claim, then P.I is more like the probability to win. If it is so, then why not use just a fraction of probability (for e.g 60% chance of winning), instead of a weird 0-200 scale.
Finally, if 1 ball is remaining and India has to score 50 runs, the P.I would show 199. However, there is infact "no pressure" at that time on the batsmen to win. So, I think they have got their name wrong :-)
I feel its a waste of time... though I must confess, that I have looked at the P.I on rediff's page for all the games so far:-)
great analysis..
Post a Comment