A lot of people ask me for help in their fantasy football drafts and I’m happy to help. The really odd thing is that I’m significantly better at playing fantasy football than I am at fantasy baseball. When I was at ESPN and we had a “Fantasy Show” league, who won? Me. Not Howie Schwab, me.
What bothers me is when people expect me (or any other expert - I see Jay Clemons and Chris Harris getting a lot of similar questions, and believe me, those are two guys I’d go to for help!) what to pick. I’m not there. Here in Indy, Peyton Manning always goes too high. I see the same biases in leagues around the country. You know who has those biases or likes. There’s always a guy who has to have Maurice Jones-Drew or someone. You can tell who read Football Outsiders because they’re the ones drafting Rashard Mendenhall in the first. Those are things I can’t know.
I think the problem is that people want to pick players, not points. You HAVE to pick points. There’s two levels of this. On the one hand, you have to pick the most points, or actually, the player that you feel will get you the most points. This is pretty easy at the top - do you think Adrian Peterson or Chris Johnson will get the most? In almost every league, that’s your 1-2. If you simply took a projection list and went down it from most to last, you’d be ok. That’s essentially what an autodraft team ends up as and you know, those orphan teams often end up mid-pack. (They’d probably be better than mid-pack if they were actively run.)
The second thing you have to keep in mind is the point differential for positions. In most leagues, you’ll be drafting 2 QBs and 4 RBs. With 30 NFL teams, that’s 30 starters at each, more or less. (We’ll ignore injuries, timeshares, etc, for simplicity.) The spread between QB #1 and QB#10 is seldom as much as the difference between RB # 1 and RB #20, a principle I first heard from Matthew Berry and that I’m sure he still expounds on. Let’s check this:
According to MockDraftCentral, the #1 RB is Chris Johnson. The #20 is Jahvid Best. The #1 QB is Aaron Rodgers and the #10 QB is Eli Manning. (He’s actually #11, but Brett Favre skews things, so I’ll use Manning.) Ignore whether or not you agree with these rankings. It’s as non-biased a ranking method as I could think of. Now, let’s look at the points expected from these, using Football Outsiders Almanac for the points:
Johnson (259)/Best (120) = spread of 139
Rodgers (295)/Manning (250) = spread of 45
There’s no question that Rodgers (or any of the top tier QBs) will put up more raw points than the top tier of RBs. It’s the relative dropoff that you have to take into account. You can wait and get the #11 QB and still do just fine; you’re losing less than 3 points a week. With the RBs, much, much more. This is why tiering your picks is so key. You have to know where those cliffs are and to adjust.
Which brings us to the theories of value based drafting and it’s more technical cousin, dynamic value based drafting. Trust me, if you’re not using this heading into your draft, you’re going to lose to someone who is. Go read up on it or better, find yourself a good program that does it for you. You’ll still be in control and you can always ignore the advice, but give yourself the best chance of winning on draft day.
And of course, read my football columns at SI so you won’t draft too much risk.
willcarroll posted this