Hello, and welcome to the new hosting site for my Tennessee High School Football Preseason Projections! Returning for my third season, The System™ produced excellent results in 2018 with a smaller, East Tennessee-only sample size in the regular season: a 279-54 (83.8%) likelihood of correctly calling the outcome of a game, well above the expected mark of ~77%. In the Tennessee state playoffs, which included all teams from Tennessee, not just those in my personal region, the ratings correctly predicted 177 of 221 games (80.1%), slightly above that same expected mark. On the whole, The System™ went 456-98 (82.3%) and correctly predicted the winner in 24 of 27 state semifinal/championship games. It was a good season; now comes the challenge of somehow topping that.
Even if that doesn’t happen and these projections regress to a mean of 77-78%, this still means they’re correctly projecting nearly four out of every five games played weekly in Tennessee. Considering Cal Preps/Max Preps’ ratings alone correctly projected 84% of winners over the last four regular season weeks + playoff action in 2018, the rate of correct projections could reasonably stay stable or even slightly rise. (Quick nerd note: anything above 75% correct is considered good; 80% or higher and you’re doing very well. 85%+ and I get hired by STATS LLC or something.)
Anyway, that’s too much precursory writing. If you clicked on this, you came for what the headline promises: statewide Tennessee high school football projections. For the first time since I’ve been doing this, all teams in the state of Tennessee are receiving full projections, not just East Tennessee. This covers 342 teams in region play in the state of Tennessee. While this does exclude a few non-region programs, this still gathers every program that can potentially play in the TSSAA playoffs this November/December.
In the below sheet, you’ll get the following to start with: a full sheet of projected records, win totals, and region records for all 342 teams, plus individualized team sheets. The team sheets feature schedule data and projections for each game. Throughout the season, these will be updated weekly and will still be available for viewing. Below are the official 2019 Preseason Win Projections for Tennessee High School Football, sorted by Class and Region:
Here’s the big, giant sheet, which is linked here as a finalized published version. At that link, you can access all 342 individual team sheets, with full game-by-game projections and season-long win totals. These will be updated weekly, even while I’m on a honeymoon in early October. (Don’t tell my fiance about that part.) While you can access season-long win projections on Massey Ratings (and, to some extent, Cal Preps), this is the first Tennessee-exclusive high school football analytics sheet that has existed, as far as I know.
After Week 1, all preseason projections will be frozen and available for viewing throughout the season on the 2019 Preseason Win Totals tab. They won’t change, but they’ll still be used on the 2019 Running Win Totals tab, which will be made available after Week 1. On that tab, you can keep track of how your team is overperforming or underperforming their preseason projections. It’s very nerdy, but also very useful to see (hypothetically) who’s coaching a team up well beyond their expectations.
As a reminder, here are the main features of the projection system:
- Team ratings from diverse statistical sources. This year’s sources are Cal Preps/Max Preps, Massey Ratings, and Sonny Moore’s power ratings. All three use different methods to achieve their final rating, and all three are worthy, useful tools. The current plan is that, after this season, we should be closer to achieving a Stats By Will rating as well. To start the season, the ratings are weighted towards the more reactive Cal Preps system, as Massey and Sonny Moore are more retrodictive and use last year’s numbers heavily to start a season.
- A small boost for home field advantage. Per a recent study, the average home team win rate in high school football is around 57%, which converts to almost exactly 2.5 points. Because there’s no objective way to measure home field advantage across high schools otherwise, this is the best we’ve got. So: if you’re at home, your projected margin is aided by 2.5 points.
- A normal distribution system for win likelihood, similar to Bill Connelly’s S&P+. The official equation, if you use Excel, is NORMDIST((relevant cell), 0, 17, True). 17 is an unusually conservative deviation number, but because of the elevated level of variance in high school football, it’s necessary. That’s why even the best teams in the metrics don’t have a 99% chance of going undefeated.
Added features for 2019:
- Project-A-Game. Nothing new to loyal followers, but this is my Single Game Projection Tool, now available for full season usage. You can project any game your heart desires at any location your heart desires.
- Game Score Adjustments. This is a tool that takes a game’s Power Score (basically, the combined ratings of the two teams) and adjusts it for how close or how not-close a game is expected to be. If a game is projected within ~2.5 points – the margin of home field advantage – it can get as large as a 30-point boost. If a game is projected to be a 30+ point destruction…well, it can receive a 30-point demerit. This is also a useful way of getting 1A, 2A, etc. games higher in the weekly rankings, as those teams won’t rank as highly in statewide rating systems.
This should cover everything, at least prior to the week-by-week preview posts. If you have any questions, comments, corrections, etc., please email statsbywill@gmail.com.
On the next pages, you can find the following:
Stats by Will 