How the MaxPreps high school football rankings work
- 6 months ago
- 0 comment
The National Football Rankings are a projection based on which MaxPreps staff writers believe will finish the year as the number one ranked team in the country. Because there is no true national championship at the high school level and most of the highly ranked teams do not play each other, it is based on judgment and human considerations behind the rankings. Our staff looks at past results, game film, talent, strength of schedule and other factors when determining the rankings.
State Computer Ranking
Computer ranking is based on results that do not involve any prediction or projection.
MaxPreps does not poll coaches, sportswriters or fans. Nor do our staff make any judgments on the merits of any individual team. The MaxPreps computer rankings do not consider previous season history, school size and message board comments.
The system uses the results of a large number of games stored in the MaxPreps database. Generally, the more wins a team has, the higher the ranking. However, the system takes into account quality wins (against other highly ranked opponents) and strength of schedule.
For example, a team's ranking suffers more by losing to a lower-ranked team than to a higher-ranked team.
Other factors that will affect rankings are when MaxPreps has incomplete or incorrect information. We will correct errors if reported to us If we miss a score, report it to us on the team's MaxPreps page. If we miss a game on the schedule, send us a correction request.
Playoff wins carry more weight than regular season games.
Movement refers to movement up or down within the rankings compared to our previously published rankings.
More on state football rankings
Assume the following starting ratings:
- Team A's rating is 10
- Team B's rating is 0
- Team C's rating is -5
- Team D's rating is -8
- Team E's rating is -10
The program systematically sorts through all results for the season (season-to-date results for a progressing season). It takes each result and compares it to what it should have been according to the teams' ratings. It knows that if A plays C, A would handle them fairly easily. If A loses that game, or even squeaks out with a narrow win, his rating suffers, while C's is helped.
The system checks through all the results for each team. Sticking with team A, if it plays D and wins by 15 (which is what they should have done, there's no real effect on either team's rating), destroys team B by 22 (which definitely helps their rating), and D - beat 10 (Not doing as well as expected, another "ding" against their rating.)
When all is said and done, they take the sum of how well or worse they did in all games than expected, divide by the number of games played, and adjust their rating accordingly. For example, if they perform an average of two points worse than expected, their rating drops from 10 to 8. (Please note: this is definitely over-simplified; it's not straight-forward-mathematics. Points aren't everything by any means - winning or losing is always the most important thing, even when margin is used. There is a "diminishing returns" principle when playing. In order not to give a team full credit for blowing out a weak opponent. Past the cutoff point that the margin doesn't count, there is a "minimum win" and a maximum - a number below which no win is credited as... (Because, of course, a one-point win is far better than just a one-point loss.) All teams are adjusted the same way, and then we start over with new ratings – A is now an 8 and performance is expected accordingly, etc. This is done repeatedly until there is no more movement in their ratings and they settle where they "should".
Remember when we told you to keep thinking about how they started back there? They didn't actually start there. All teams start at 0. There is no bias here – neither last year's stats nor pre-season projections are used as a starting point (again, see the one exception above). Everyone starts at 0 and the rating continues until the movement stops.
When running without margin, the process is the same, but margin of victory is not considered. A win is a win, and all wins count at the same level. Therefore, examples of getting your rating "dinged" due to a closer-than-expected win do not apply. What matters is the win (and who you played).
Powered By: MaxPreps