Monday, September 12, 2022

Me vs. the 538 Model: NFL Forecasting, Week 1

Well, it's that time of year again ... FiveThirtyEight's NFL Forecasting Game ... and I almost forgot!

In fact I did forget until Sunday morning, which means I didn't have a pick in for the Bills/Rams game, and thus neither gained nor lost any points. I suppose I could say that the Bills would have been my (winning pick), but that would be kind of self-serving, wouldn't it?

I picked nine of 15 games correctly. Of those I didn't pick correctly, one I didn't pick at all and the other (Colts v. Texans) was a tie. So those two cost me no points.

The games I blew: I incorrectly picked the 49ers to beat the Bears, the Bengals to beat the Steelers, the Panthers to beat the Browns, and the Titans to beat the Giants. As it happens, every game I picked incorrectly, the 538 model failed on as well, but the model also blew the Buccaneers vs. the Cowboys and the Vikings vs. the Packers.

The final game of Week One is tonight, when the Broncos play the Seahawks. I've picked the Broncos and given them a 67% chance to win. I'll try to remember to update this post after the game.

My score so far: 56.8 points. The 538 model's score: -0.9 points. I am currently ranked in the 98th percentile, in 83rd place among 5,120 participants.

Update, Post-Monday-Night-Football: I gave the Denver Broncos a 67% chance of beating the Seattle Seahawks. FiveThirtyEight made the same pick, with only a 57% chance. The Seahawks won, knocking my score-for-season down by 19.9 points to 36.9, and the FiveThirtyEight model's score down to -8.4. I'm still in the 98th percentile on the leaderboard, but now rank 121st of 5,317 participants.

No comments: