+
☰
sports spectrum
✕
Cricket IPL Cricket ICC Football Kabaddi Olympics Athletics Sports History
AboutContactPrivacy Policy
✕

Raid success rate rewards the raider who does the least

The denominator is doing all the work in kabaddi's headline attacking statistic, and it is doing it dishonestly.

Raid success rate rewards the raider who does the least

Divide successful raids by total raids and you get a percentage that gets printed next to a player's face. It is the number agents quote at auction. It is also close to meaningless, and the problem is not the top of the fraction.

Every raid counts the same. A raid into a full seven man defence with the corners set and the cover players alert counts as one attempt. A raid in the last thirty seconds against three tired bodies also counts as one attempt. A do or die raid, where the raider is obliged to make contact and the defence knows it, counts as one attempt. These are not the same task. Grading them identically means a raider can lift his percentage by choosing his moments, and choosing your moments is not the same as being good.

The bonus point makes it worse. A bonus counts as a successful raid in most versions of the statistic. So a player who goes in, stretches over the line, collects one and retreats is building the same record as a player who goes in against seven and takes out a corner defender. One of those raids changes a match. The other keeps the scoreboard ticking.

What I would want instead is something like an expected value per raid, weighted by how many defenders were on the mat, whether the raid was compulsory, and the score margin at the time. That is not exotic. Every league already logs the defender count because it is needed to award super tackles. The data exists. Nobody publishes it in a form that would embarrass a well paid raider.

There is a defensive mirror of this problem. Tackle success rate for a corner is heavily shaped by how often his team sits at three defenders, since raiders play cautiously in that state and a cautious raider is an easier catch. Right corners and left corners do not face the same distribution of raid types either, which makes comparing them across teams roughly as useful as comparing a wicketkeeper to a slip fielder on catches taken.

The number is not lying to you exactly. It is answering a question nobody asked.

How did you feel about this article?