Consistency – The Key to Winning Fantasy Football, Part 2

The Jamaal Charles / Frank Gore example in “Consistency – The Key to Winning Fantasy Football, Part 1,” was a simple one in that both player’s weekly fantasy point production had virtually the same mean or average. Comparing their standard deviations, then, was an easy way to determine the more consistent fantasy running back. However, running backs have various means and standard deviations, so how does one identify a consistent running back from one who is less so? One way would be to graph a player’s fantasy point production by week, as was done at the beginning of Part 1, and visually make a determination – but that is a lot of work and eyeballing a chart is not always the best way to make an accurate assessment of consistency. So what is a better way?

Consistency and the Coefficient of Variation

An easier method would be to calculate a player’s coefficient of variation or CV. The CV is found by dividing the standard deviation by the mean as demonstrated in the table below.

Coefficient of
Variation (CV)
Jamaal CharlesKC12.89.20.72
Frank GoreSF12.75.10.40

For Jamaal Charles, the coefficient of variation is 0.72, which says that the standard deviation is 72% of the mean. A large CV such as this indicates large variations in the data points from the mean – a condition we have previously observed in Part 1 of this article.

For Frank Gore, the CV is 0.40. This is a much smaller percentage compared to the CV of Jamaal Charles. Based on these two CV values, it can be said that Frank Gore’s fantasy point production has more consistency than that of Jamaal Charles.

So how can you tell if a particular CV value is good or bad? It depends on the the CV values of those being compared. For the top 20 running backs in 2012, the maximum value observed was 0.77 (standard deviation 8.2 / mean 10.7) for Reggie Bush of the Miami Dolphins. The lowest observed value was 0.34 (standard deviation 5.6 / mean 16.4) for Arian Foster of the Houston Texans. A coefficient of variation alone, though, cannot identify the more valuable running back if their means are dissimilar. However, based on the table below, a few significant conclusions can be drawn.

Running BackTeamGames
Coefficient of
Variation (CV)
Arian FosterHOU1616.45.60.34
Jamaal CharlesKC1612.89.20.72
Frank GoreSF1612.75.10.40
Reggie BushMIA1610.78.20.77

Arian Foster and Frank Gore show relative consistency in their fantasy point production as they both have low CV values (0.34 and 0.40). Between the two, most team owners would prefer the player with the higher mean – Arian Foster. However, if Arian Foster habitually missed, let’s say, four games per season, would he be valued as highly even if his mean, standard deviation and CV were the same? The answer is – of course not. He may still be preferred over Frank Gore, but not as highly. Even a highly talented running back such as Arian Foster is less valuable if he is not available for a significant portion of the season. It is a judgment call every team owner has to make and both consistency in production and availability are major factors in that decision.

Also from the table above, one could state that both Jamaal Charles and Reggie Bush lack consistency in their fantasy point production as they both have relatively high CV values. Again, between the two, most team owners would likely prefer the player with the higher mean – Jamaal Charles.

Consistency plus value!

Let’s assume these are the only four players in the draft. How should they be ranked? Based on the mean and CV, one could argue the ranking should be:

  1. Arian Foster
  2. Frank Gore
  3. Jamaal Charles
  4. Reggie Bush

But let’s not forget about the average draft position for each player. Even though Frank Gore is ranked 2nd in our preference above, Jamaal Charles will still be drafted before Frank Gore. (Read my article, “Online Draft Strategy and the Curious Case of Quarterback Matt Ryan,” for a further discussion as to why Jamaal Charles would likely be drafted before Frank Gore. Hint – it has to do with the default player ranking list.). This is where your competitive advantage comes in. The shrewd team owner, knowing that Jamaal Charles will likely be ranked higher than Frank Gore by most team owners, will wait until after Jamaal Charles has been drafted before selecting Frank Gore. There is no need to spend a high pick on a player you can get later, right? Always look for value in your draft picks. Know the average draft position of the players before participating in a draft.

Although large spikes in fantasy production, like those of Jamaal Charles, may be useful when the rest of a fantasy team is having a bad week, success is more easily achieved over the course of a season when a fantasy roster is built for consistency. Granted, consistency is difficult to achieve. Many factors, such as injuries, game-plan decisions, weather, schedule and opponents have a significant impact. However, taking the time to identify the players who perform with consistency can be the difference in having a winning fantasy football team or a season of unfulfilled potential and disappointment.

Fantasy Points Production Analysis

Think it takes too long to calculate the mean, standard deviation and coefficient of variation (CV) for every scoring player in the NFL? Well, Fantasy Football Impact has done the hard work for you. Get one of the most useful research tools created for fantasy football team owners – our Fantasy Points Production Analysis. It comes as  part of our seven-workbook Fantasy Football Metrics Package for 2017.

Related articles at Fantasy Football Impact:

Return to the Drafting Strategies overview page

Consistency - The Key to Winning Fantasy Football, Part 2

Disclaimer: Fantasy football is a game of speculation; therefore, Fantasy Football Impact disclaims all warranties, expressed or implied, regarding player performance or results associated with its opinions or advice.