What changes have I made?
I’m including this blurb in each of these posts.
First, with a better chance to analyze which statistical factors are most consistent from the beginning of the season to the end, I made some slight tweaks in the weighting of each statistical factor (the short version: efficiency carries even more weight now). I also worked marginal efficiency and marginal explosiveness into the equation.
Then, I implemented the changes I made during 2018 for previous years. From each week’s rankings post:
- I changed the garbage time definition. S&P+ stops counting the major stats once the game has entered garbage time. Previously, that was when a game ceased to be within 27 points in the first quarter, 24 in the second, 21 in the third, and 16 in the fourth. Now I have expanded it: garbage time adjustments don’t begin until a game is outside of 43 points in the first quarter, 37 in the second, 27 in the third, and 21 in the fourth. That change came because of a piece I wrote about game states at Football Study Hall.
- Preseason projections will remain in the formulas all season. Fans hate this — it’s the biggest complaint I’ve heard regarding ESPN’s FPI formulas. Instinctively, I hate it, too. But here’s the thing: it makes projections more accurate. Our sample size for determining quality in a given season is tiny, and incorporating projection factors found in the preseason rankings decreases the overall error in projections. (For previous years, from before I actually released any sort of preseason projections, I found the most predictive success by keeping a layer of five-year history within the ratings. It’s a small percentage, but it’s in there.)
- To counteract this conservative change, I’m also making S&P+ more reactive to results, especially early in the season. If I’m admitting that S&P+ needs previous-year performances to make it better, I’m also going to admit that S&P+ doesn’t know everything it needs to early in a season, and it’s going to react a bit more to actual results. Basically, I’ve added a step to the the rankings process: after the rankings are determined, I go back and project previous games based on those ratings, and I adjust the ratings based on how much the ratings fit (or don’t fit) those results. The adjustment isn’t enormous, and it diminishes dramatically as the season unfolds.
One more recent change had the most impact, however: I made S&P+ more reactive to conferences as well. It’s similar to step 3: after the rankings are determined, I project previous games based on those ratings, and I track each conference’s average performance versus projection. For the top conference, I found that by the end of the season it was aiming low by two or three points per game per team. For the bottom conference, it was the reverse.
By shifting each team’s rating based on this conference average, and by increasing the weight of said adjustment as the season progresses, it adds basically improves against-the-spread improvement by about one percent per season and cuts the average absolute error by somewhere between 0.2 and 0.3 points per game. That doesn’t seem like much, but look at the Prediction Tracker results and note how much of a difference 1% and 0.3 points per game could make to your projective ranking there. It’s pretty big.
It does, however, mean a fundamental shift in how mid-major teams are judged. Not to spoil the suspense, but look at the difference this adjustment made in some 2018 rankings:
- Fresno State: originally ninth, now 16th
- UCF: originally eighth, now 18th
- Utah State: originally 19th, now 21st
- Appalachian State: originally 11th, now 29th
It’s a pretty harsh adjustment, though it both makes the numbers better and perhaps passes the eye test a bit more. So we’re going with it.
Wait, so you’re including previous years of history in each season’s ratings? How could that possibly be right?
Back when I mentioned at the beginning of 2018 that I’d be using priors in the rankings all year, I had an interesting conversation with some readers on Twitter about what happens at the end of the year. Would I be removing the priors for the year-end ratings so that a) we’d be evaluating teams’ performance based only on what happened in that year, and b) the “recent history” prior wouldn’t then carry influence right into the next year’s projections (since they would include the priors that were included in the previous year’s rankings)?
It was a legitimate question, one to which I don’t think there’s a right answer. To me, it simply comes down to this question: when you’re looking back at previous seasons’ ratings, what are you looking for?
To me, it’s usually to get a sense of who might beat whom, right? I understand the draw of a “this year only” evaluative look, but S&P+ is intended to be a predictive measure, and I’ve decided to include these priors because they make the predictions better. In that sense, removing those priors at the end of the season makes it less of a predictive measure (even though obviously any “predictions” made of a team from 2006 in 2019 is theoretical only).
(How will this impact the preseason projections moving forward? The weights of each of the primary projection factors — recent performance, recruiting, and returning production — are always based on updated correlations, meaning that it depends on how strongly each factor predicts the next season’s performance. I’m dumping the new S&P+ ratings into the old engine, and, well, we’ll find out. Maybe the recent performance numbers end up with a lower correlation. We’ll see.)
On with the rankings.
2006 S&P+ rankings
|Team||Rec||S&P+||Rk||Off. S&P+||Rk||Def. S&P+||Rk||ST S&P+||Rk|
|Team||Rec||S&P+||Rk||Off. S&P+||Rk||Def. S&P+||Rk||ST S&P+||Rk|
|San Jose State||9-4||-1.8||67||24.0||59||26.1||75||0.3||51|
|New Mexico State||4-8||-10.2||93||26.6||46||35.7||114||-1.2||98|
|San Diego State||3-9||-13.2||100||14.1||107||25.6||73||-1.7||110|
The 2005 rankings re-release righted a wrong of sorts: national champion Texas rose from second in the old rankings to first, passing the USC team it defeated in the national title game. This re-release, on the other hand, managed to muddy the waters in a pretty spectacular way. National champion Florida, originally No. 1 in S&P+, falls all the way to fifth, stuck in the middle of a mass of pretty dang equal teams.
Really, the chaos that unfolded in 2007 actually began the year before. While Texas’ and USC’s percentile ratings were both at 98.8 percent or higher in 2005, new No. 1 LSU peaked at 97.9 percent. There was no truly elite, No. 1-level team, but there were about five No. 3s. This year was maybe as messy as 2007, but Ohio State’s ability to skate through the regular season unbeaten prevented us from realizing that at the time.
Not as many significant shifts as 2005
- Teams that saw their rankings rise by 20 or more places from old method to new method: Wisconsin (up 27 spots to 15th), Kent State (up 25 spots to 81st), San Jose State (up 21 spots to 67th), NIU (up 20 to 69th)
- Teams that saw their rankings sink by 20 or more places: Cincinnati (down 27 spots to 44th), MTSU (down 21 to 101st), Air Force (down 21 to 89th)
While these ratings tweaks resulted in eight 2005 teams moving up at least 20 spots and eight more moving down at least 20, there were only four and three, respectively, here. But while the Big Ten wasn’t as strong in 2006, the efficiency-based ratings tweak did efficiency-based Wisconsin some favors.
Here are your average S&P+ rankings by conference:
- SEC (+13.3 adjusted points per game, up 5.7 from 2005)
- Big East (+11.8, up 8.1!)
- Big 12 (+10.0, down 2.1)
- Pac-10 (+8.5, down 2.6)
- Big Ten (+7.7, down 5.5)
- ACC (+6.8, down 0.9)
- Mountain West (+1.0, up 1.8)
- Conference USA (-6.1, up 0.1)
- WAC (-7.3, up 3.1)
- MAC (-11.2, down 1.2)
- Sun Belt (-19.4, up 1.3)
Man oh man, what a shift we saw from 2005 to 2006. The Big Ten was the No. 1 conference in ‘05 but fell to fifth in 2006, the SEC went from fifth to first, and the Big East (!!) went from a distant sixth to second. Two Big East teams ended up ranked higher than Michigan (which was three points from reaching the BCS title game unbeaten), and there were more MWC teams in the top 21 than ACC teams. After the chalkiness of 2018, I think that, sans Clemson and Bama fans, we could all use a strange lurch like this in 2019. Not counting on it, though.
Timing is everything
- In 2006, LSU and Hawaii ranked first and 33rd, respectively, in S&P+. The Rainbow Warriors were first in Off. S&P+ (despite the extra opponent adjustments I installed, which tamped mid-majors down quite a bit — that’s what 5,500 passing yards from Colt Brennan will get you), while the Bayou Bengals were fourth on offense and second on defense. In terms of post-game win expectancy, these teams produced stats that should have produced 23.6 wins in 27 games.
- In 2007, LSU and Hawaii ranked fourth and 35th, respectively. The Warriors were 13th on offense, and the Tigers were seventh on both offense and defense. In terms of post-game win expectancy, they produced stats that should have produced 22.1 wins in 27 games.
- The 2006 LSU and Hawaii teams went a combined 22-5, finishing third and unranked, respectively, in the final AP poll.
- The 2007 LSU and Hawaii teams went a combined 24-3, finishing first and 19th, respectively. LSU beat Ohio State for the national title, and Hawaii went unbeaten in the regular season and reached the Sugar Bowl.
Your best teams aren’t always the ones that break through. Even when you peak, you still need some breaks to peak in the win column. Just ask 1998 Tennessee. Or 1994 Nebraska. Or 2007 LSU and Hawaii.
Maybe the best defense of the 2000s
In 2006, Virginia Tech pitched four shutouts and allowed seven or fewer points three other times. Miami and North Carolina each scored 10 points on the Hokies; it was tied for the fifth-most Tech allowed in a game. Boston College scored 22 points but gained only 264 yards in the process. Georgia scored 31 but gained only 200.
Virginia Tech’s Def. S&P+ rating of 4.6 adjusted points per game is the lowest of what you might call the S&P+ era (2005-18). Only three other teams have been at even 6.5 or lower (Alabama in 2009, Alabama in 2011, Alabama in 2017). This unit erased you. That the Hokies lost three games with this defense — 22-3 to BC, 31-24 to UGA, 38-27 to Georgia Tech — was a miscarriage of justice.
After ranking 20th in Off. S&P+ the year before, the Hokies fell to 64th in 2006, falling into an offensive funk from which they wouldn’t emerge until 2009. (The renaissance only lasted 2-3 years.) But damn, was this defense a work of art.