IPL position predictor

As Ravi Shastri would say: “We are in the business end of the tournament”. The points table looks really crowded with a few teams on 12 points and a few bunched together on 10 points. There are 17 games left as of today (4/7/2010) and most teams have 4 games left. The range of points that cannot be used to predict who will make it to the top-4 because who plays who also becomes critical.

So I wrote up a simple simulation which looks at a 1000 possible result scenarios to determine the probability of each team ending up in positions 1 through 8. The  major assumptions in this simulation are

1. Each team has equal probability of winning a game. It might not be a great probability estimate for individual games but it is some starting point.

2. Net Run rate obviously has not been used to separate the teams on the same points. Thus, these probabilities reported below must be read as the probability that each team scores points that will put them in that position 1-8. Actual qualification may be more complicated with net run rates messing things up.

So here are the results:

Team 1 2 4 5 6 7 8
Mumbai Indians 0.803 0.124 0.044 0.021 0.007 0.001 0.000 0.000
Delhi Daredevils 0.232 0.343 0.203 0.135 0.055 0.028 0.004 0.000
Rajasthan Royals 0.113 0.283 0.213 0.187 0.124 0.076 0.004 0.000
Royal Challengers Bangalore 0.125 0.282 0.171 0.171 0.122 0.091 0.037 0.001
Chennai Super Kings 0.034 0.182 0.164 0.199 0.176 0.182 0.062 0.001
Kolkata Knight Riders 0.057 0.167 0.160 0.208 0.184 0.182 0.042 0.000

Deccan Chargers 0.001 0.017 0.037 0.067 0.128 0.216 0.418 0.116
Kings XI Punjab 0.000 0.000 0.000 0.002 0.021 0.048 0.266 0.663

Advertisements
  1. #1 by sriram subramanian on April 7, 2010 - 4:03 pm

    why not use the current win:played ratio as the probability of each team winning?

    • #2 by Kaushik on April 9, 2010 - 2:40 pm

      I used this years Home game win percentage as a new probability estimate. It is around .6 for home team instead of 0.5 for home team now. Not much difference in the results though!

  2. #3 by sachita on April 7, 2010 - 5:16 pm

    confused as to why csk and kkr have diff. nos!

    • #4 by Kaushik on April 7, 2010 - 6:46 pm

      because their remaining games are with teams in different halves of the table. CSK has games left afainst DC, KKR and Punjab while KKR have to play toppers like MI, RCB etc.

  3. #5 by Vinay on April 8, 2010 - 1:44 am

    Too good work, man! putting Game Theory into practice?? You could write up a conference paper for this.

    • #6 by Kaushik on April 9, 2010 - 2:41 pm

      What! There is no game theory in this. It is just the basic definition of probability.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: