NFL: Interpreting Best Ball Data

This article is part of our NFL Observations series.

We recently got some data from the Fanball/NFFC on their 12-team, 20-player best ball drafts over the last two years, and I took a quick look. 

Here are the players from 2018 and 2019 who on average caused teams to finish at least one standard deviation above the expected 6.5 (12-team league):

2018 League Winners

Before we get into the conclusions, let me explain the table above. I looked at each player's average finish, i.e., the place where teams who drafted him finished on average in their standings, and subtracted it from 6.5, the expected average if a player merely earned his draft slot, to get Average Value Over Replacement (Avg VORP.) 

I also tracked the number of league winners and money finishers (spots 1-3) by percentage, and you can see it tracks the average finish pretty closely. I think the average finish is a better barometer than the percent of cashes because the sample is larger, but it's good that it aligns well with the ultimate goal which is finishing in the money. 

I also noted the No. of leagues in which a player was drafted. Tre'Quan Smith teams did unexpectedly well in 2018, but that was a sample of only 1,603 leagues and should probably be thrown out since he scored only 88.7 points all year, i.e., it was basically dumb luck. (I also eliminated anyone drafted in fewer than 500 leagues from consideration at all.)

Also, I took the standard deviation of the average finish (0.7 in 2018) and divided the Avg. VORP by it to get the total number of standard deviations a player was above or below average in a given year (STD VORP.) This measures how much of an outlier he was, i.e., how much of an impact he made. 

I listed each player's ADP too, but I didn't use it because it's obviously priced in -- average finish for a player in the 20th round includes a full team of other players drafted in Rounds 1-19. Average finish for a player drafted in the first round only includes a team of other players drafted in Round 2-20. 

For 2018 (and it's the same in 2019 below), Christian McCaffrey was by far the most valuable player. After that, the descent in value is more gradual. Even Patrick Mahomes' monster 50-TD season didn't crack the top two and was nowhere near McCaffrey's, despite Mahomes having a much cheaper ADP. 

Of the 36 difference makers, two were QBs, 15 were RBs, 12 were receivers including the aforementioned Smith, five were TEs and one was the Bears defense. Among players drafted in the top two rounds (ADP 24 or lower), there were six players, four of which were running backs and two of which were receivers. If we extend the "difference-maker" category 32 more players deep to half a standard deviation better than average (.35), then we pick up three more running backs and one receiver. In short, among the top 67 most impactful players, more than twice as many backs from Rounds 1-2 were difference makers than receivers. (Overall there were 13 RB, 10 WR and one TE taken in the first-two rounds in 2018.)

The sweet spot for difference-maker receivers in 2018 was in Rounds 3-5 (picks 25-60.) Of the 12 who were a full standard deviation above expectation, five were taken in those rounds. For backs, there was not a single one. If we extend our range to half a standard deviation, we add two more receivers from those rounds and not a single extra running back. 

Two top tight ends (Travis Kelce and Zach Ertz) were difference makers, while three others came from the late-middle rounds. 

Now let's look at who destroyed you:

2018 League Killers

Le'Veon Bell, who sat out the whole year, just murdered teams, given his first-round cost. After him it was more gradual, but as you can see the running backs giveth, but they also taketh away. Of the 39 qualifying team-destroyers, 14 were backs, eight were tight ends, four were quarterbacks and 13 were receivers. But seven of the eight worst players were running backs, the lone exception being Dez Bryant whose ADP was 106 and who got you zero points. 

In sum, early round running backs in 2018 were most likely to win you the league, but they were also most likely to make you finish last. Keep in mind, though that fourth place pays the same as 12th place, so as long as you don't have any major side bets, running backs were still your best bet in the early rounds in 2018. 

Let's take a look at last year's numbers:

2019 League Winners

Once again McCaffrey was the overwhelming difference maker, even more so, despite his high acquisition cost. And again a quarterback, Lamar Jackson in this case, despite having a season for the ages at ADP 132, didn't even finish second. Quarterbacks lack anything close to the upside running backs have in this format. 

Interestingly, while the top two impact players were running backs last year, many more receivers disproportionately helped your chances to win. Among the 41 players above the one standard-deviation threshold, a whopping 21 of them were receivers, six tight ends, two defenses, four quarterbacks and only eight running backs. 

Among players taken in the top-two rounds, three running backs and two receivers hit the standard deviation threshold. One other top two-round back and no receiver hit the half-SD threshold. In total, four top-two round backs and two receivers significantly helped your cause in 2019. (There were 12 RB, 10 WR and 3 TE total in the first two rounds last year with two WR tying for ADP 24.)

In Rounds 3-5, picks 25-60, three running backs and six wide receivers cracked the 1 SD mark. Two running backs and three wideouts in that range reached the 0.5 SD threshold. Overall there were 18 wideouts and 16 backs taken in that range. 

Bottom line, in 2019, the top two difference makers were again running backs, but they comprised only three of the top 10, which also included three wideouts, two tight ends, two quarterbacks and the Patriots defense.

2019 League Killers

A.J. Green, a player universally drafted, but who didn't play at all, was last year's Le'Veon Bell. Andrew Luck, who fits the same description, was the second worst choice. Otherwise, the busts were more evenly distributed with six of the top-10 being wideouts.

One interesting bust was Mahomes, who cost you more than a full standard deviation despite scoring 315 fantasy points. Quarterbacks with top-40 ADPs have little room for error in this format. 

If I had to offer advice from this sample (admittedly only two seasons), it would be to draft at least one running back in the first two rounds, at least two receivers in Rounds 3-5 and no QBs before Round 4. Also, don't be afraid to pay up for tight end. While last year the two biggest profit sources were Darren Waller and Mark Andrews, both of whom came cheaply. George Kittle and Zach Ertz at picks 24 and 25, respectively, also delivered a nice profits, while Ertz and Travis Kelce both delivered in 2018.