Data scientist, physicist, and fantasy football champion

Week 7 QB Results

Another pretty normal week. Carr at the top was a little unexpected if you believe that he does worse in-division (which I don’t think is statistically significant https://www.thedatastream.org/data-exploration/2017/9/26/does-in-division-affect-fantasy-scoring) . Brady and Ryan both had OK weeks, but this week OK just wasn’t good enough to break the top-12. McCown made it up there again, though. Hundley’s fantasy day was saved by his rushing TD, but otherwise wasn’t nearly up to the level of Aaron Rodgers (a totally fair comparison, I think).

Let’s see how each model did last week:

.accuracy graph-1.png
Model estimate std.error
ModelA 0 NA
ModelB -14.3417142857143 6.81662394544009
ModelC -10.9588571428571 6.81662394544007
ModelD -16.1417142857143 6.81662394544008
ModelE -14.5702857142857 6.81662394544008
ModelF -12.9274285714286 6.81662394544008
ModelG -16.4837142857143 7.76406374622289
ModelH -22.1937142857143 7.76406374622288
ModelI -13.4533439153439 8.51412447134917
ModelJ -6.34001058201059 8.51412447134917
ModelYahoo -5.69599999999999 6.81662394544008


And it’s Model H again, but this time by a ton. Model H is the simple linear model that uses only 2017 data. Players that have done well this year have continued to do se each week. It’s not significantly better than the next model back (D or G) but it’s definitely the model I’m following whenever possible.

Week 7 K Results

Week 7 QB Predictions