One of the most difficult tasks in head-to-head fantasy leagues is trying to determine who the “hot” and “cold” players will be in any given week. You can’t use historical or seasonal averages because, well … by their very definition, averages do not tell you how the average was achieved.
Years ago, I implored Mighty Max, my fantasy football supercomputer, to implement a statistical methodology to quantify player consistency. Mind you, I wasn’t looking for “Andrew Luck averages seven 300-yard games a season” or “Marshawn Lynch consistently finishes the season strongly” statements, because they are worthless when filling out weekly lineups — and are usually the province of hack fantasy writers who have bought authoritative-sounding domain names.
Max came up with what is now a critical component of our Best Damn Draft Method 2015 the industry innovation known as the Sports Grumblings’ Consistency Rankings (CR). To help illustrate the value of the CR, consider the following hypothetical situation of two quarterbacks over a four-week period:
Who’s the quarterback you’d want on your fantasy team? Both Chucker and Thrower have amassed 1,000 total passing yards; both average 250 yards a game. But here’s where the CR becomes important: Thrower’s CR weighs in at 129.09; Chucker’s would be 21.98. The CR would accurately tell us that Chucker is the steadier performer (the lower the CR, the steadier the player).
So what is the CR, and how is it calculated?
The CR is the degree to which a set of data points varies. For those who are mathematically inclined, the CR is a “coefficient of variance”:
For those of you who just want to get to the bottom line, trust me when I say that the math works. I am, after all, a descendant of Archimedes and Aristotle …
The larger the CR is, the more inconsistent the player; the smaller the CR the more consistent the player.
On the surface, both players seem to have had remarkably identical passing performances. But for fantasy owners who had to fill out a roster every week, there is no indication of how these stats were accumulated; did Brees get the bulk of his stats in the first half of the season? Did Roethlisberger toss 12 touchdowns in two games and 20 in the other 14? In other words, who was the more consistent player from week to week?
The results might surprise you. Mighty Max took the weekly fantasy scores for both players; he produced the following CR for each player:
To a fantasy owner playing in a total points league, this information is no big deal; it’s the players’ total performance which matters most. But in a head-to-head league, Brees provided his owners with a more dependable scoring source than did Roethlisberger. In fact, Brees was roughly 11 percent more consistent than Roethlisberger!
Listed below are the five most consistent players at each offensive position, according to their 2014 weekly performances (I took out players with insignificant scoring):
Of course, the CR does not make any claims to how good a player is, it just tells you how consistent he is; a really bad player may have a really good CR, i.e., he is consistently bad.
Note: Last year, I offered a spreadsheet of Consistency Rankings for every offensive and defensive player to anyone who asked for it. So many of you asked for the data and I’m offering it to you once again this year. If you want a free copy of the 2015 Consistency Rankings, just send me a note at firstname.lastname@example.org and I’ll send you a copy. If you find it useful, you can do me a solid in return by visiting my weekly NSFW talk show at The Riot.