Fantasy Footbal Data -- I Want to Know How Bad I Am At This

This week, Deangelo Williams scored four touchdowns. My fantasy team's opponent was starting him.

In my previous game (two weeks ago, due to an annoying bye week, which I got because I lost in the first round of the playoffs), Brian Westbrook went off. I lost largely because Westbrook went off. Also because most of my team didn't show up; except for my bench players. I'll get to this later.

For most of the season, I've been among the most-scored-upon teams in the league. I've also had to listen to debates between other people claiming their team is really good versus other guys who pointed out that they have by far the fewest points scored against them and aren't that great.

Frankly, I think ESPN doesn't give us enough data. You know, to really figure out who's good, who's bad, and more importantly, who's the unluckiest bastard in the league.

I want to figure out which team's opponents scored above their average points level the week they faced him.

I want to figure out the magnitude above average those teams were.

More granularly, I want to know which team's opponents' players scored the highest above their average points level the week they faced him. (So if you got rocked by someone starting Donnie Avery in one of the two weeks he played well, you got jobbed. Why was Avery starting? That kind of deal.)

Also, it'll be nice to know the opposite. You know that guy in your league who always seems to win because his opponent has a bad week every damn week? Well this would be a way to quantify that, as opposed to it just being a feeling. Because everyone knows feelings are worthless.

And another thing! I want to be able to examine bench points. I want to go through your scores for your starting lineup and your bench, and compute your "optimal" points for that week, if you'd started everyone properly. That way, it'd be possible to figure out which guy is the worst fantasy football manager in the league, by looking for the largest gap between optimal and actual scoring. I guess it's not actually the worst -- he has good players -- just the unluckiest. When he zigs, reality zags. I have a feeling I'd be high on this list (a few times this year, my bench has outscored my starting lineup), but I also have the feeling that most people have that feeling. It's because you're so close to your own team and not anyone else's. So I want to calculate it so I know just how awful I am at this.

For example, two weeks ago when I lost in the playoffs. I started Ted Ginn Jr. He got me 1 point. Meanwhile, Vincent Jackson got 20 points for my bench. If I had played "optimally," I would have 19 more points just from that one move. And for all those weeks I started Donnie Avery in the hope that he'd score big (and didn't), I was playing sub-optimally. I want to know how much these mistakes cost per week, and over the course of the season.

Also, I want to be able to calculate the optimal scoring for each week, and for each of your opponents, and calculate an "optimal" record. IE, what your record would have been had YOU played optimally each week AND your opponents also played optimally. I think the optimal record would give the truest indication of which team is actually the best. Much better than actual record. Also, the difference between optimal record and actual record is another indication of luck ... and "who to start" skill.

I don't know if any fantasy football sites already offer stats like this. I just use ESPN, and they keep it to the minimum (though they make the minimum pretty smooth and easy to use). I don't even need them to offer these stats in a nice little package for me. I just want access to the data.

I want to be able to write a query, or a program, to calculate these things for my own league. Why can't I do that? It doesn't even have to be live -- they can make me download a giant XML file of the data for each week, I don't care. But why don't they make this data public? Is the NFL afraid that if someone has access to how many points Tarvaris Jackson scored one week, that they're going to start a separate league competitive with the NFL? Or that if you care this much about the stats from the games, that you won't watch any more? Does their paranoia make sense?

Just give me access to the data, you jerks. I want to know just how terrible I am at fantasy football.

Also, I hate Deangelo Williams right now. He is not a friend of the Doomsday Church of Oblivion.

Posted by Sean Schulte at 2008-12-22 10:05:22

blog comments powered by Disqus