Drafting Fantasy Football

Fantasy football season always drafts when the fantasy baseball season is coming down to the wire. This bodes poorly for my gridiron teams, which are often made up of auto-drafted players I have never heard of. So what do I do? I look at the stats, of course. And not just any stats…

To win the league, a successful draft is the first and most important step. Any fantasy player that understands a draft knows that the key is squeezing extra value out of every pick, especially in the later rounds. Almost everybody’s first and second round picks are easy no-brainers, but an unexpectedly awesome season from your 5th or 6th pick can make all the difference. In my mind, there are two things that make one position more important than another in the draft: the availability of other talent at the same position, and the predictability of fantasy points from players at that position. Here goes…

I have a lot of time on my hands, so I ask questions like, “should kickers be drafted earlier?” and, “what if lining up all your bye weeks is actually a good strategy?” Yeah, I’m cool. Kickers often garner only half the points quarterbacks and running backs do, but what’s more important is the spread within each position. In the last six seasons* the top kicker has outscored the 12th best kicker by an average of 33 fantasy points with a standard deviation of only 6 points. In other words, the best kicker is almost sure to get your fantasy team at least 25 more points than the 12th best kicker in the league, and probably more.

Comparatively, the top quarterbacks in my league’s scoring system beat out the 12th best quarterbacks of 2009 and 2010 by an average of 96 points. So don’t draft your kicker before your quarterback. Ok, dur, who didn’t know that? But what about your third wide receiver, your defense, or your tight end? Here’s a chart that shows the average difference between the first and 12th best at each position (2009 – 2010 averages).

Pos

Spread

RB1

149

QB

96

TE

86

WR1

61

RB2

50

DEF

50

WR2

38

K

33

WR3

22

While this is a small sample for most of the positions (just two seasons), it becomes somewhat obvious why quarterbacks and running backs cloud the top 20 picks in any fantasy format. This is where managers can make up the most fantasy points relative to the other options available at those positions. The running backs look far superior, but that spread is dragged up by two outliers known as Chris Johnson (2009) and Arian Foster (2010). The spread between the 2nd and 12th picks those years were more similar to the quarterbacks’ spread.

A couple other things to note: just looking at the spread, it’s not all that crucial that you draft wide receivers early, but drafting a top 2 or 3 tight end could pay off significantly more than I thought.  Also, kickers don’t make much of a difference (read: the thesis of this article just changed to “Positions that should be drafted before kicker = just about all of them”).

But knowing the spread is worthless without being able to predict who will finish 1st, 12th, and everything in between. If you know the spread is big in running backs, for instance, but it’s impossible to predict which backs will head the charts, then there’s no reason to use a high draft pick on a stab in the dark. In this section, I focus on the year-to-year correlation to show which positions are, on average, the most predictable. We’ll just look at the top 12 and 24, respectively, from 2009 and see how they fared in 2010.

Pos

R^2

MSE/Average

QB (1-24)

44%

25%

QB (1-12)

30%

24%

QB (13-24)

54%

25%

RB (1-24)

31%

44%

RB (1-12)

33%

39%

RB (13-24)

25%

53%

WR (1-24)

2%

38%

WR (1-12)

24%

41%

WR (13-24)

2%

32%

TE (1-24)

8%

38%

TE (1-12)

11%

40%

TE (13-24)

13%

35%

K (1-24)

3%

32%

K (1-12)

9%

24%

K (13-24)

1%

41%

DEF (1-24)

2%

26%

DEF (1-12)

33%

23%

DEF (13-24)

4%

25%

The R^2 column gives us an idea of how well 2009’s figures were able to predict 2010’s. 100% would be considered perfect, so none of these are very good. But that’s the nature of sports, it’s unpredictable. It’s what makes drafting hard. We can be pretty confident that running backs and quarterbacks are the most predictable of the bunch. The MSE/Average column gives us an idea of the how big the standard deviation of predicted points is compared to the average 2010 figures. The lower the better, here, because we’d like a low standard deviation. A lower standard deviation means smaller errors in our predictions.

I think everyone pretty much knows that QBs and RBs should be drafted first. What this tells me is that a second running back should be drafted before WRs, TE, and kickers. And it also suggests that WRs, TEs, DEFs and Kickers are pretty much crapshoots. Prediction is difficult for players in these positions, except maybe the top one or two players. So if we then look to the spreads, I might grab that top TE first before I think about taking a WR. A top 12 defense is more predictable than a 2nd WR, and provides a better point spread. This is surprising, but it could be that drafting a DEF right after your first WR is a good strategy.

This study could use more data, but I think it articulates a few valuable lessons. Sports are, by nature, very unpredictable, so don’t take it too hard if you don’t make the playoffs this season. Injuries and off-field issues, among other things, can make your seemingly smart 2nd round pick look like a joke. However, there is more predictability in QBs and RBs, and those positions also have the biggest spreads between the best and the worst (of those who will actually be on fantasy rosters in a 12-team league). So draft the best available RBs and QBs first, and don’t be shy about nabbing a top-tier TE in the 3rd or 4th round. Antonio Gates and Jason Witten are likely to be a hell of a lot more valuable than Dustin Keller and Zach Miller.

*Well, 2004-2008 and 2010. There was no breakdown of kicker distances in 2009 on Pro-Football Reference.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: