If She's a Good Passer, Why Doesn't She Pass Good?

If you haven't noticed, big Moneyball guy over here.

Our issue here is not that there are rich teams and there are poor teams - but it's that we fail to evaluate Reception in reality.

We use metrics like Passer Rating because we think it's a good proxy for passing performance. To it's credit, it's a great proxy for how often you pass well. But I am here today to persuade you that passing well is not the point - the key is to not pass poorly.

And while you might be thinking, yeah no kidding...

I submit for your consideration: Expected Sideout

What passer ratings do is over-index on the good stuff, but seriously under-value the errors. I'll show you what I mean.

So the issue is that passer ratings assume each outcome is equidistant from the next.

3-2 = 2-1 = 1-0 and that 3-1 = 2-0.

In coaching terms, your player could pass 2, 2, 1, 1 or 3, 3, 0, 0 and both would be a 1.5 passer rating.

This doesn't pass the eye test.

Hypothetical - Two Passers

They'll both get 10 serves, with both passing a 1.5 passer rating

Player A: 3, 3, 3, 3, 3, 0, 0, 0, 0, 0 = 1.5 passer rating

Player B: 2, 2, 2, 2, 2, 1, 1, 1, 1, 1 = 1.5 passer rating

But what if we use reality, instead passer ratings?

Player A: 63.9, 63.9, 63.9, 63.9, 63.9, 0, 0, 0, 0, 0 = 32% Expected Sideout

Player B: 60.7, 60.7, 60.7, 60.7, 60.7, 54.2, 54.2, 54.2, 54.2, 54.2 = 57% Expected Sideout

Same 10 balls, huge difference.

Reception Error drives the difference.

I turn now to exhibit B, teams from the 2022 NCAA Women's season.

How does their Expected Sideout change as Reception Error increases?

Quite a bit seems to be the answer.

For those playing along at home, the R2 = 0.86. Meaning that 86% of the change in Expected Sideout is explained by Reception Error %.


Your Honor, I rest my case.

While the secret to passing is obviously: the shuffle step (@DannyKinda)

the key to passing, is clearly to:

not pass poorly...