Now here is a set of coach rankings I can get behind. While the annual summer lists of “here’s where we rank the Power 5 coaches” are easier to understand (and argue with), I like a little more rigor. Why? Because even the best human minds are unreliable. So you build a model and try to remove some of the uncertainty.
Are models perfect? Rarely, but at least they make an additional effort towards objectivity. About a month ago, I took one of those coach rankings, pitted it against some numbers over a seven-year stretch and came away with different conclusions. It was a quick check, but Bill Connelly of Football Study Hall has been on that over/underrated coaches corner for years.
He recently released his updated numbers based on wins versus win expectancy for every coach that has coached at least three seasons since 2005. Here’s the idea in a nutshell:
For preview purposes, I note when teams strayed pretty far from their win expectations, one way or the other, and in 2016 two teams strayed really far from expectations: Idaho overachieved by 2.3 wins, and Notre Dame underachieved by 3.2 wins, the fourth-highest (lowest?) in 12 years.
From year to year, this is a sign of randomness. The teams on the extreme ends are all but guaranteed to regress (or progress) toward the mean the next year. Notre Dame was the fourth team to underachieve its second-order win total by at least three games; the other Power 5 team on that list: 2013 TCU, which improved from 4-8 to 12-1 the next season.
Win Expectancy mostly addresses the randomness in results. But over time, it can tell us a little bit about certain coaches.
The list includes 224 coaches, ranked by wins above or below expected wins per year. If a coach annually wins nine games with stats that indicated an eight win team, that coach scores 1.0.
So where do you think Mike Riley ranks on this list?
Over Riley’s 12 seasons since 2005, he has averaged 0.24 more actual wins than expected wins per year. That’s overachieving (or maximizing resources if you prefer) to a slight degree and ranks in the 73rd percentile, meaning 73 percent of coaches over that span have been worse. Including Nick Saban, who ranks 69th with a 0.21 win differential per year.
Saban, of course, illustrates a key point. If you’re a good coach at a place like LSU or Alabama (where the talent level is high), you almost become capped. You almost can’t overachieve because your teams put up great stats and only lose once or twice a season. It’s not impossible, however. Urban Meyer’s post-2005 career includes only Florida and Ohio State, but he ranks in the top 10. (If you care to engage in a Saban-Meyer debate, this is a pretty good trump card in Meyer’s favor.)
That said, it’s easier to do this at a place like Kansas State. Bill Snyder, in his nine seasons since 2005, has been one win better than expectations per season, which ranks second on the list. Pat Fitzgerald, another coach who always seems to come up in these conversations, is fifth. On the other end of the spectrum, Kirk Ferentz ranks 169th (-0.31/year, 28th percentile) and only one coach with at least 10 seasons ranks lower.
But back to Riley. His win differential is on just the right side of even. What does that tell us about his Nebraska tenure? A couple of things.
It supports at least a little bit of the “if you can win and keep your job at Oregon State, what happens at a place like Nebraska?” theory behind the hire, which was the entire selling point. We just haven’t seen it yet at Nebraska. Riley is -0.3 over just his two years in Lincoln, and we’ll find out if that initial theory was faulty or if given more time Riley will approach the rate he achieved at Oregon State.
If the latter happens, it makes this a pretty interesting test case for Nebraska’s program value. If we can assume that Riley will overachieve only slightly over a long enough span, what does that look like at Nebraska? Is it 10 wins? Is it eight wins? Can you change that number based on recruiting or, at a place like Nebraska (as opposed to Alabama for instance) is the best way to change that baseline number, whatever it is, to find the coach who consistently surpasses it (a.k.a. The Snyder Method)?
I’m eager to get answers — even if they’re partial, and they always are — to all of those questions.
The Grab Bag
- Al Pacino is going to play Joe Paterno in an HBO movie and it seems like nothing good can come from that. But props to HBO. High degree of difficulty there.
- Speaking of bold decisions, Ohio State parted ways with basketball coach Thad Matta yesterday. Myron Medcalf of ESPN writes that it was a messy breakup.
- Interesting read: How sports scientists are trying to change college football.
- Netflix CEO Reed Hastings says his company has no interest in bidding on the rights to live sports.
Today’s Song of Today