Which college football coaches consistently produce overachievers or underachievers?

In each 2016 team preview, you'll see something called "win expectancy" in each team's "2015 Schedule & Results" section. I talked about that a bit in today's Northwestern preview.

Last offseason, I tinkered with a measure called second-order wins. It is basically my version of the Pythagorean Wins concept, where you look at a certain component (usually points or runs scored and allowed) and determine what a team's record probably should be as opposed to what it actually is. If you're losing a ton of close games but winning a bunch of blowouts, that's probably a sign that, on average, you would be faring better than you are.

My second-order wins concept looks at the single-game win expectancy figures you see in the 2015 Schedule & Results chart below. The idea behind win expectancy is simple: It takes the key stats from a given game (success rates, explosiveness, field position factors, and other factors that end up going into the S&P+ ratings), mashes them together, and says, "With these stats, you probably could have expected to win this game X percent of the time." Add those figures up over the course of a season, and you get a glimpse of what a given team probably could have expected its record to be.

For preview purposes, I note when teams strayed pretty far from their win expectations, one way or the other, and in 2015 three teams strayed really far from expectations: Boston College was 2.9 wins below where it should have been, while Houston and Northwestern were both 2.8 wins above.

From year to year, this is a sign of randomness. But over time, certain coaches fare better year-to-year than others. And as I noted last offseason, the coaches at the top and bottom of this list tend to be the ones fans say are particularly good or bad coaches.

If a random college football fan were asked to create a list of what we'll simply call "bad coaches," he or she would probably fairly or unfairly mention Ty Willingham, John L. Smith, Dave Wannstedt, Ted Roof, Houston Nutt. Washington State fans might mention Bill Doba. North Texas fans would definitely list Todd Dodge. They're all on one particular end of the chart.

Meanwhile, asking someone to list "good coaches" might produce a list with men like Bill Snyder, Gus Malzahn, etc. They can be found anchoring the other end. Again, I'm not sure this actually means anything, but it certainly might help to confirm what some might already think about given coaches.

You can potentially distill coaching into two things: building a team that produces great stats and figuring out how to maneuver in tight games when neither team has a statistical advantage. The former matters above all (Nick Saban and Steve Kragthorpe, after all, are nearly even on this list ... and on no other list in existence), but if nothing else, this list might help us to understand which teams/coaches are good or bad at the latter.

Since that post, I've made a few tweaks to how win expectancy is calculated (I'm always, always, always tweaking), and we've added another year of results. The list hasn't changed that much, but I thought I should share the list once more.

Below is a look at everyone who has been a head coach at least four years since 2005 and what their average difference is per year -- actual wins vs. win expectation.

Navy's Ken Niumatalolo remains the master of squeezing out one more year than expected each year, and fellow spread option master Paul Johnson remains in the 90th percentile. And this still takes shape like a conventional-wisdom look at "good coaches" (Bill Snyder, Pat Fitzgerald, Urban Meyer, etc.) and "bad coaches" (Todd Dodge, Ty Willingham, Dave Wannstedt, Houston Nutt, etc.).

Since I'm previewing Big Ten teams right now, I found the standing of Fitzgerald (plus-0.8 wins per year, seventh overall) and Indiana's Kevin Wilson (minus-1.0 per year, second-worst overall) particularly interesting. Fitzgerald is now 34-21 in one-possession finishes in his career, and while a lot of those wins were of the "letting an inferior team hang around for a while, then eking out a W" variety, Fitz remains a source of fascination. He was a ridiculous 26-13 in such finishes in his first seven years at NU, then skewed the other way: 3-8 in 2013-14. But 2015 was a master class in keeping games close and making one extra play. Despite a sub-50 S&P+ ranking, NU went 10-3 because of a 5-0 record in one-possession finishes.

Finishing as far above win expectation as Northwestern did in 2015 was almost certainly a fluke. But Fitz manages to pull off a surprisingly high result more often than not.

Wilson, meanwhile, is 7-18 in such games. His Hoosiers were 2-11 in his first three years and have evened out a hair, going 5-7 the last two years.

As we see from this list, this isn't purely a tempo thing, but it would be really interesting to compare game management styles and tactics from guys like Fitz and Bill Snyder and compare them to Wilson and Steve Addazio -- conservatism vs. aggression and whatnot.

Even with years of data, we're still suffering a bit from a small-sample issue, but the coaches on this list have coached at least 48-50 games. That's a decent sample. And it certainly seems to drop some heavy hints regarding those who are particularly good or bad at pulling out close games. Don't mess with Pat Fitzgerald, and really don't mess with Ken Niumatalolo.