NFL 2017 Week 20 (Conference Championships) Predictions – Zoltar Agrees with Las Vegas

Zoltar is my NFL football machine learning prediction system. It’s a hybrid system that uses a custom reinforcement learning algorithm plus a neural network. Here are Zoltar’s predictions for week #20 (NFC and AFC conference championship games) of the 2017 NFL season:

Zoltar:    patriots  by   11  dog =     jaguars    Vegas:    patriots  by    9
Zoltar:     vikings  by    0  dog =      eagles    Vegas:     vikings  by    3

Zoltar theoretically suggests betting when the Vegas line is more than 4.0 points different from Zoltar’s prediction (3.0 points difference for regular season). Zoltar agrees with the Vegas point spread to within 3.0 points, so Zoltar has no recommendations.

But if forced to give advice, Zoltar would say:

1. Take the favorite Patriots against the Jaguars.
2. Take the underdog Eagles against the Vikings.


Zoltar went 0-0 against the Vegas point spread last week — Zoltar agreed with Vegas to within 4.0 points.

Just predicting winners last week, Zoltar went 3-1. He correctly predicted the wins of Eagles over Falcons, Patriots over Titans, Viking over Saints. But incorrectly predicted that the Steelers would beat the Jaguars. The Vegas point spread was 2-2 predicting winners.

For the 2017 regular season, against the Vegas point spread, Zoltar finished a pretty good 51-32 (61% accuracy). If you must bet $110 to win $100 (typical in Vegas) then you must theoretically predict with 53% or better accuracy to make money, but realistically you must predict at 60% or better accuracy.

I tracked how well Zoltar, Bing Predicts, and the Vegas line do when just predicting which team will win. For the 2017 regular season, just predicting the winning team, Zoltar finished 178-78 (70% accuracy), Bing finished 168-88 (66% accuracy), and Vegas finished 162-86 (65% accuracy). The best humans are typically about 67% accurate predicting winners, so Zoltar is slightly better than the best human experts. Bing did OK too, beating Vegas by a percentage point.

Note: Some of my numbers could be off a bit because of some weirdness with games played outside the U.S. (London and Mexico) earlier this season.

My system is named after the Zoltar arcade fortune teller machine

This entry was posted in Machine Learning, Zoltar. Bookmark the permalink.