ESPN's Interesting FCS Change to its Football Power Index

It may not have seemed like much, a bit of news quietly released during the news dumping ground of Friday afternoon by ESPN Stats and Info.

"Upgrades sharpen ESPN's college FPI model," the release was titled, a piece which explained some changes that were being made to ESPN's "Football Power Index".

One of those changes involve how FCS wins and losses were represented in their model.

And it's actually a bigger deal than you think.


ESPN's FPI "is an estimate of team strength," they write, "not a ranking of who will have the highest win total (which is dependent on schedule) and not who is most likely to make the College Football Playoff. Preseason FPI is calculated using three major components for each FBS team: prior years’ offense, defense and special teams efficiencies; returning starters and head coach information; and recruiting rankings."

ESPN's Football Power Index is a complex formula incorporating a boatload of different data points that was missing one key ingredient - FCS vs. FBS games.

As I had complained about years ago,

Dr. Saturday makes the unfortunately common mistake of lumping all FCS schools together, incorrectly associating teams like 1-10 Idaho State and FCS National Champion North Dakota State together simply because they share a subdivision - kind of like lumping Ohio State and Middle Tennessee State together, simply because they're FBS schools. 

In fact, tests from FCS schools are depressingly common - if you know which FCS schools to look for. 

Until they started packing the schedule with cupcakes like partial-scholarship Austin Peay and transitioning FCS school South Dakota, Wisconsin was no stranger to FCS scrapes.  Wisconsin was given everything they could handle by The Citadel in 2007 before winning 45-31. In 2008, the Badgers were a missed field goal away from being stunned by Cal Poly, squeaking by 36-35.

It was this yawning gap between fantasy and reality, where an loss to FCS North Dakota State would be seen as much, much worse than one to FBS Georgia State, that gave me zero faith that any measure of "schedule strength" for the College Football Playoff, whether by computers or humans, would ever accurately be registered.

Much to my surprise this weekend, I read this.

Most college football fans would acknowledge that playing four-time reigning FCS champion North Dakota State is not nearly the same as facing a bottom-tier FCS team. The Bison have won five straight games against FBS opponents dating to 2010, including four straight against Power 5 teams. That’s as many Power 5 wins as Kansas has since 2010! 

So when North Dakota State headed to Iowa State last season, it was clear to the average fan that the Cyclones were not facing a typical FCS opponent. In the previous iteration of FPI, however, all FCS teams were regarded as being of the same caliber and rated below the weakest FBS team. Based on that logic, Iowa State entered the game with a 97 percent chance to win, and when they lost 34-14, the Cyclones fell 31 spots in FPI. 

In a small 12-game sample, misrepresenting the actual strength of even one opponent could have a significant impact on a team’s FPI, which has a trickle-down effect on a number of FPI’s outputs -- such as future game projections and strength of schedule. 

From the moment FPI was released in 2013, we knew this was a problem that had to be solved, but because of limitations with FCS data, we rolled out FPI with the intention to resolve this issue in the future.

After my spit-take, I realized that indeed they did, changing their model to take some basic data from the top FCS teams in order to attempt to figure out their relative  "power ranking" in the FCS world, and where that might end up in the FBS list.

This may not seem like a big deal, but it is on several levels.

We're Taking Your Guarantee And the Win, Cyclones (Grand Forks Herald)

First - this is a giant leap of respect as to how games vs. FCS games are considered.  It took, on average, an upset a year or thereabouts for nearly a decade for FBS fans to realize that the topmost levels of FCS teams are as good or better than a good portion of FBS teams.

App State over Michigan, for a while, was seen as a once-in-a-millennium fluke that could never be repeated.

But as North Dakota State kept embarrassing Power 5 programs, however, the latest being a 34-14 domination over Iowa State, even casual college football fans really couldn't really deny that the very best of FCS are pretty good teams.  Maybe not Ohio State or Oregon good, but very good teams.

ESPN's FPI wasn't altered to make FCS teams feel better - it was done to attempt to rectify this discrepancy in misrepresenting the relative strength of FBS teams.

You have to believe the folks who care deeply about these computer models are paying attention to their models, too, especially since ESPN trumpets that the FPI correctly predicted the outcome 76% of the time, "which was a higher percentage than almost any other system out there -- including the Vegas closing line -- and the majority of ESPN analysts," they said.

The second aspect involves FBS scheduling decisions.

Earlier in the summer, Big Ten commissioner Jim Delany splashily noted that he was attempting to eliminate scheduling of FCS teams in the next couple of years.

Why Yes, Bo Has Something To Say About That

In a recent Youngstown Vidicator article, they mention that the main reason for it is "perception".

While it’s not mandatory, Delany has been pushing the latter policy for the past few years in an effort to boost the league’s strength of schedule for the College Football Playoff. It’s been bad news for Youngstown State, which was originally going to play a lucrative game against Ohio State this fall (the Penguins will now play Pitt) and must now work harder to find FBS games.

...

Penguins coach Bo Pelini has been on both sides of the issue, having coached at Big Ten/Big 12 member Nebraska the past seven years before getting hired at YSU in December.

When asked about the Big Ten’s new policy, Pelini said: “If that’s what they feel they need to do, that’s their prerogative. Obviously, I’m sure it has to do with strength of schedule points or perception in terms of getting into playoff position. 

“Not every conference has adopted that [policy]. I think there’s a time to [play FCS schools] and a time not to. Our job when we get the chance to play FBS schools is to be ready to play and ready to get after them.”

But in ESPN's FPI, now a game scheduled against, say, Northern Iowa could carry a lot more "schedule strength" weight in their system than a low level MAC team.

Today, each FCS team has an FPI rating based on the final score of games dating back four years. As with FPI for FBS teams, the rating is a measure of team strength and represents the expected scoring margin against an average FBS opponent. 

For example, North Dakota State enters the season with an FPI rating of +0.9, meaning if the Bison were to play an average (or 64th-ranked) FBS team, they would be favored slightly on a neutral field. 

Wait, it gets better.

When you take a look at that +0.9 rating in the actual FPI index, you realize some more fun facts:

• North Dakota State's +0.9 is greater than the FPI rating of every Sun Belt school.  Appalachian State, at +0.8, is the highest of that conference.
• It is greater than every MAC school except one: Western Michigan, who only beats the Bison's rating by clocking in at +1.0.
• Including the WMU Broncos, there are only 8 schools in the "Group of Five" that are rated above the Bison.  Boise State (+10.9) is the highest-rated, perhaps unsurprisingly, but they're the only Mountain West team ranked above North Dakota State. 
• The gap between North Dakota State and the worst-ranked FBS team UNLV, was +0.9 to -20.1, or 21 points.  

Whether they intended to do so or not, ESPN's preseason Football Power Index inadvertently ranked North Dakota State as essentially the equivalent of the best teams of both the Sun Belt and MAC, and the upper half of Conference USA, the Mountain West and the AAC.

That's a massive amount respect - and from the release, they're not the only team afforded that much respect.  It turned out that 32 FCS teams were ranked higher in their system than the lowest FBS team, which was UNLV out of the Mountain West.

This newfound respect for the higher FCS teams could actually harm any conference that refuses to schedule them - like the Big 10.

Put in another way, the Big 10 is now saying it wants to replace games with an FCS school that could have a rating of +0.9, and instead replace them with FBS teams with negative ratings, some much more so.

Going through with their scheduling policy could actually harm their standing in the FPI.

Now, lest one go overboard, ESPN's Football Power Index certainly isn't the only thing that the College Football Playoff committee will be looking at when they enter that smoke-filled room and try to figure out which four teams will be competing in the plus-one-playoff.

But Delany's decision to stop scheduling FCS schools could be coming at the worst time possible - when college football fans, and computer indexes, are discovering that there are some pretty good FCS schools out there, and are changing their expectations and formulas accordingly.

It's possible that the Big 10's scheduling decisions may very well put them at a disadvantage in at least one computer index that will be finding themselves in the hands of the decision makers.

And  it could be a change that has some very positive ripple effects down the line for FCS schools - especially as other indexes and polls start to do the same thing.  As data reflects what fans have been seeing for years, bans on FCS games could be placed in the same bad idea box as New Coke, Webvan and the XFL.