Who will win ProKabaddi 2016?

How do we predict win and loss projections?

Each raid, each tackle, each minute spent on bench will change the way the game unfolds. To predict match win probabilities we need to understand how player perform head to head. In kabaddi, this means understanding how a raider performs under different defense scenarios. For instance it might be much harder for raiders to perform with 3-4 tacklers in the defense compared to 6-7 member defense.

Using 2000+ raids and tackle digitized from Season 2, we predicted the relationship between Raid Strength, Tackle Strength and consequently the points scored in that raid.

Player Scores

Based on a player’s performance in the previous season and form in the current season – as measured by points scored for your team and points given to the opposition, every player gets a Tackle Strength and a Raid Strength. This calculation is based on a multi-linear regression model

Read more: Season 3 Predictions | Player Ratings

We simulated 5000 matches raid-by-raid between every team. For each raid, we estimated the points scored by bootstrapping the raid data from Season 2.

The match was simulated with the assumptions

  • Raiders with higher Raid Strength have a higher probability of conducting raids.
  • Defense players are out in successful raids are selected randomly.
  • On winning a match, the player strengths for all players in the team increased by 1 percentage point.

We have not accounted for home advantage or stress of matches played on a stretch.

In the end, match results are decided by the interaction between 14 players across 80 raids. But over a period of 56 matches, predictions are bound to make sense