Newsletter

Weekly Consistency - Summary

Draft Recommendation: Do not pay attention to a player's consistency, a good player has more value than a consistent player.

The idea for this article came from FFToday, a site well-known for its Crank Scores. “Crank Scores rank the consistency of a player (Crank = Consistency-Rank). Crank Scores is a concept designed to help identify players that consistently give the best performance on a weekly basis in head-to-head leagues.”  I noticed they explained why they prefer to have consistent players on their team but without backing it up with any statistics.

I also noticed that quarterbacks and running backs are much more consistent than wide receivers so following that theory, a top QB or RB would have more value than a top WR because odds are that the top QB or RB won’t let you down as often as the top WR might. I decided to test this out to determine if the consistency of players at specific positions can affect your overall draft strategy.

The first step to this analysis was to find out which position is more consistent than others. I used the coefficient of variation to measure this on statistics from 2005 to 2009 and, as I had previously believed, quarterbacks are much more consistent than the other positions while wide receivers and tight ends are the most inconsistent. In theory this should increase the value of quarterbacks over wide receivers and tight ends since come playoff time, your top QB is less likely to let you down that your top WR is. However, I decided to test this and find out if this was also true for a starting lineup of nine players.

Using two fictional fantasy teams that averaged the same total of fantasy points, I saw that a team filled with very consistent players will defeat a team filled with very inconsistent players only 51.25% of the time. To put this in perspective, I created two more fictional teams that both have the same consistency and the team that scored only 1.8 points per week more than the other team won 52.36% of the time.

This does not even account for the fact that it is quite difficult to predict a player’s consistency from year to year. If we assume that: you can predict how consistent a player will be from week to week and you can draft the most consistent players on your team without neglecting talent, your advantage will be less than half a point per week for your entire team.

There is no statistical evidence that shows you should stay away from Terrell Owens or Randy Moss because they are inconsistent players. Although in theory a more consistent player should help your fantasy team win more games, with nine players on your team there are always some that will have good weeks and bad weeks so it neglects the effect of a single player’s inconsistency. Moreover, there are no NFL players that are consistent or inconsistent enough from week to week for this theory to have a real impact.

In conclusion, if another manager in your league says he is staying away from Ryan Grant because he has a low “Crank Score”, give him a fake smile and be very happy to draft him (if he’s the best player available of course).

See the complete analysis