Tuesday, January 27, 2009

Fantasy Football: Ignoring Touchdowns

In 2007, Braylon Edwards and Roddy White posted similar statistics in receptions and yards.

Edwards: 80 receptions, 1289 yards
White: 83 receptions, 1202 yards

Those numbers amount to a difference of 0.2 receptions and 5.5 yards per game, a negligible difference. Yet going into 2008, most fantasy footballers ranked Edwards far ahead of White (Fantasy Football Weekly, for example, ranked Edwards as the #3 WR, White as the #36 WR). Part of that difference was team context: Edwards would be catching passes from a returning Pro Bowl quarterback, while White would likely be catching passes from a rookie quarterback.

But a major difference was in touchdowns: Edwards had 16 touchdowns, White six. It seems quite reasonable to assess players by their touchdowns, since touchdowns play a major role in fantasy football. But Edwards' TD numbers should have raised crimson banners: 16 touchdowns is disproportionate to his other receiving numbers. One out of every five Edwards receptions went for a TD, and he had a touchdown for every 80.6 yards receiving. But it is simpler than that: 80-1,200 yard seasons are relatively common, 16 TD reception seasons are relatively rare.

That in 2008 Edwards was a major disappointment (55-873-3) and White continued to perform well (88-1382-7) of course cannot be attributed solely to Edwards' disproportionate TD numbers of 2007 (Edwards dropped a lot of balls, Cleveland QBs and offense in general struggled, Matt Ryan performed well). However, those touchdowns are critical to fantasy footballers--here's why.

Fantasy football is about guessing a player's actual value in contrast with his perceived value. This is true of auction or snake drafts: you don't just want to have ideas about what players will succeed or struggle, but you need to measure those ideas against the consensus of perceived value. Touchdowns, I think, disproportionately affect a player's perceived value--other statistics might be better indicators of a player's future performance. I'm not saying you should have known Edwards would struggle in 2008; I'm saying there were indicators he wouldn't live up to his perceived value.

And when preparing to draft, you need to find gaps between a player's perception and his performance. When a major fantasy football publication ranks 32 WRs between two WRs that posted nearly identical receptions and yardage, you know you've found a gap between perceived value and actual value. This can indicate players that will be drafted too early and too late, helping you to decide what high-ranked players to avoid, and what low-ranked players might be later-round steals.

There are a few lesson here. Beware of players whose TDs are disproportionate to their other numbers. Pay attention to different numbers to figure out which players have overrated perceived value. Pay attention to other numbers that indicate future success. You could even try do what I did specifically with Edwards and White here: try find a low-ranked player that is statistically even with a high-ranked player in other categories.

But to quote imminent theologian "The Misfit" of Flannery O'Connor's "A Good Man is Hard to Find,

"If He did what He said, then it's nothing for you to do but throw away everything and follow him, and if He didn't then it's nothing for you to do but enjoy the few minutes you got left the best way you can."

We don't need no stinking moderation: either devote yourself wholly, or forget about it and do whatever you want.

So I'm taking this theory further. I'm going to rank fantasy football players for 2009 with no regard for their 2008 touchdown numbers. The Hazelweird League is a touchdown heavy league, but in ranking players, I'm going to pretend touchdowns don't exist. I'll assess all the other possible individual statistics, examine those statistics for cumulative numbers, efficiency numbers, and game log numbers. I'll consider players on factors like team context (quality of teammates, quality of offensive system, quality of coaches, difficulty of schedule, etc.). I'll look at long-term performers, and I'll guess at players I think will break out.

My theory is that good fantasy players will score touchdowns, and I can look to other numbers to tell me who the good fantasy players are (or will be). But because touchdowns may distract from a good guess of a players' actual value for the following year, I'm going to pretend there are no touchdowns when I rank players.

Perhaps I will make two lists: the first factoring touchdowns, the second ignoring touchdowns. But I would guess that the first list will give me a closer idea of what order players will be drafted in, and the second list will give me a better idea of how players will actually perform. The second list will tell me what high-valued players to avoid early, what low-valued players to target late. It is the second list that will make some overlooked players jump out, some overrated players reveal themselves. It's the second list I'd go with.


  1. Anonymous9:10 PM

    Will you be sharing this list with your faithful readers?

    Thanks for bringing up Braylon Edwards. Just wonderful.

  2. Anonymous8:54 AM

    Interesting take on touchdowns. Look at AP, nowhere near the number of TD's that LT had a couple of years ago, but AP still had a great season.

  3. I've always considered football TDs to be sort of like baseball RBIs, when it comes to drafting for fantasy leagues. They're very dependent on what other people do and can only somewhat be ascribed to the individual player's skills.

  4. David Kaeser here, I'm currently in a touchdown only league since 1992.Case it point.Willie parker 16 td performance a few years back.The last 2 years have been supar for Willie Parker.You need to take the body of work to account for a fair value of a player.I stay away from 1 year wonders.
    My website is http://trashmouthfantasyfootball.weebly.com/

    thanks Dave