WASP doesn't have much buzz

As an example, imagine that Zimbabwe were playing Australia and halfway through the second innings had done well enough to have their noses in front. WASP might give a winning probability of 55% for Zimbabwe; but based on past performances, one would still favour Australia to win the game. That prediction, however, would be using prior information about the ability of the teams; so it is not interesting as a statement about how a specific match is unfolding. Also, the winning probabilities are rounded off to the nearest integer; so WASP will likely show a probability of winning of either 0% or 100% before the game actually finishes, even though the result is not literally certain at that point.

Also, another novelty is in including an adjustment for the ease of batting conditions in the models. There is adjustment done for estimating ground conditions, here. Without that adjustment, the models would overstate the advantage or disadvantage a team would have based on a good or bad start respectively since those occurrences in the data would be correlated with ground conditions that apply to both teams. Using this novel technique, WASP is said to estimate ground conditions from historical games and so control for that confounding effect in our estimated models.

All said and done, I still feel there are some flaws with the WASP. Firstly, it does not take into account the nature of the batsmen and bowlers in the middle. For example, consider a match situation where Virat Kohli and MS Dhoni are batting, and India needs 40 runs from 5 overs. Given the previous track records of these two match winners, obviously the match will be in India’s favour in any condition, any ground against any opponent (although India may have poor records against that team in the past).

Now let us consider another match where Ravindra Jadeja and Mohammed Shami are batting, and India needs 40 runs from 10 overs against a relatively weaker side, say Bangladesh. Although Bangladesh has a poor record against India – under such a circumstance – the WASP should ideally favour Bangladesh to win the match even if the match is played on a placid track.

Similarly, South Africa defending 16 runs off the last two overs with Dale Steyn and Morne Morkel to bowl would be totally different when compared to Ishant Sharma and Ravichandran Ashwin bowling the last 5 overs with the opposition requiring 50 runs.

Secondly, WASP doesn’t take into account how teams fare in crunch matches. For example, South Africa has often grabbed defeat from the jaws of victory in knockout matches against not so formidable sides. So depending on the importance of a match, WASP should have a provision on taking into account the significance of a match (a semi-final or a final against a group league match) and how the team has fared in similar situations from previous matches.

Thirdly, the WASP, although, takes into account the past batting record of a batsman, what if that batsman bats in a different position on a particular day? Chris Gayle batting in the middle order as against batting as an opener can have two different outcomes. Unfortunately, WASP would average out Gayle’s performance as a batsman and given his audacious extravagance and great record as an opener, it might overestimate West Indies’ chances in a match where he plays as a middle order batsman.

Fourthly, the fielding dominance of a team should be taken into consideration. With a great fielding side, like South Africa in a big ground, say MCG, a team chasing even 240 may find it difficult owing to the great pressure implied by the South Africans.

Finally, the idea of strike rate for bowlers and batsman evolving over the course of a match is also not taken into account. A batsman would either accelerate at the start only to slow down in the middle, or would start slow and then explode. Similarly, a bowler’s ability to take wickets during different stages of a game is different, which should be taken into account while making such predictions.