step 3.5 – attack efficiency does explain the past

To be fair, predicting the past is easier. It’s already happened. We’re just looking for which metrics help explain something that has already happened. To it’s credit, attack efficiency differences between two teams in a single set is wildly indicative of who won that set. So yes, we do care about attack efficiency, definitely don’t ignore it…but due to set by set variation even amongst the top teams, we cannot accurately rely on this metrics as means of predicting the future.

Here’s how we explain the past, using attack efficiency.

The basic setup to get the data in the right form. Correlation looks promising.
Linear Regression – doesn’t make as much sense since winning and losing is not a continuous variable. Strong R2 value relative to others we’ve seen so far.
Logistic Regression – makes the most sense since we are trying to classify the likelihood of winning the set. Strong explanatory power indicated by McFadden psuedo-R2 value – nearly 70% of variance explained with this model.
Here’s the dual-probability + histogram view. You’ll notice pretty clean separation between W/L where if your opponent out-hits you, you typically lose, lot of data to the left of 0 on the bottom histogram. And then naturally the opposite at the top.

So that’s really all I have to say about that. Don’t ignore attack efficiency – it’s clearly an important indicator when you look back and reflect on why you’ve won or lost, but don’t make the mistake of thinking your team’s attack efficiency going into your next set has the same ability to predict the future as it does to explain the past.

1 Comment

Comments are closed.