Who is the best?
That’s the question that drives Gridironrank.com along with NFL fans, analysts, members of the front office, and that annoying coworker you have that always brags about his fantasy team.
In professional football this question can be, and often is, answered through qualitative analysis; watching film of players and teams and comparing their results to everything else we’ve seen in the sport. With such a complex sport this process is extremely important and oftentimes spot-on, but at Gridironrank.com we seek to add a quantitative explanation that can either support or give us the evidence to challenge what we see on film.
In fact, we accomplish our goal using exactly the same methodology, but from a number’s perspective; a team’s performance is calculated as the difference between team’s results and the average result in similar situations.
The key to calculating any statistic is coming up with a means of comparison and calculating it for similar situations. Up until now we have had to settle for the stats of points and yards allowed or gained, which have been around since, well, about 1869 when Princeton lost to Rutgers in the very first football game ever played. Since then we’ve had some time to think about a better way of doing things and we believe that our concept of Expected Points provides the smartest way of accomplishing this goal.
Expected points (EP) are really simple: they’re the average number of points scored next given any game situation (think down, yards to go, field position, etc.)
So imagine this scenario: It’s 1st & 10 at your own 20 my expected points model says that you should score 0.36 points on average.
Compare this with starting 1st & 10 at your own 1, my model predicts that you will score -0.63 points on average (This just means that the defense is more likely to score the next points, and on average they will score 0.63 points).
With this information in hand, a team’s performance is simply calculated as their expected points after a play minus their expected points before that play. Based on our example above, if a team had 1st & 10 at their own 1 and then picked up 19 yards, they would finish the play at 1st & 10 from their own 20. The team added 0.36 – (-0.63) = 0.99 expected points of value in that play.
To get a measure of team performance we simply add these values up over a set of plays. We call this one comprehensive measure TEAM PS, or Team Production Stat. This gives us one number that assigns value on every single play based on the context of each play, instead of simply adding up yards or points. That little draw play on 3rd & 20 that picks up 15 yards is still short of a 1st down, so it’s considered a negative TEAM PS play. Picking up only 3 yards on 4th & 2 still keeps a drive alive and so is rewarded with a positive TEAM PS.
This simple concept of expected points allows us to analyze exactly how effective a team was in all facets of the game, and to explain what made them effective from the perspective of the numbers. Add this data to all the info you get from breaking down film, and our stats will allow you to enjoy the game on a whole new level, or at least help show that coworker how much smarter you are than them.
Sound good? Check out where your team ranked in 2013
Still confused about that expected points stuff, read a detailed explanation here