[sirsean]

Maybe something like this could be altered to make predictions, but that wasn't really my goal.

In my league this year, there was a team that started 11-0. He thought that proved he was the best fantasy football player and had the best team. Everyone else pointed out that he simply had the fewest points against. This was a way to calculate what his record should have been if the other players had set their roster optimally. (It turns out he should have finished 11-2 instead of 12-1, and was like 4th in total optimum points for.)

Optimum points is better for recognizing which players you should have started, not for who you should pick up or draft. If anyone has any input on what it should do to be better at prediction, I'm all ears -- I just don't know if that could really work reliably.

[jaxn]

Sounds like you are "moving the goal posts" to me.

[sirsean]

Heh. It probably looks like that. But determining who has "the best team" doesn't change who won the league championship. It's just an interesting footnote. (Does anyone think the Phillies were actually the "best" team in the MLB, or that the Giants were actually the "best" team in the NFL? The "best" team doesn't always win the championship.)

And, honestly, the primary reason I wrote it was to determine how much I suck at fantasy football -- it always seemed to me that players would score really well when they were on my bench, and suck if I started them. Optimum points, and the difference between optimum and actual, are a way to measure that.

(It turns out my gap between optimum and actual wasn't as big as I thought it'd be, and other teams had a much larger gap. Just because you have a feeling about something doesn't make you right.)