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 #18 (the Wild Card games) of the 2017 NFL season:
Zoltar: chiefs by 6 dog = titans Vegas: chiefs by 8 Zoltar: falcons by 0 dog = rams Vegas: rams by 6.5 Zoltar: bills by 0 dog = jaguars Vegas: jaguars by 8 Zoltar: saints by 5 dog = panthers Vegas: saints by 6.5
Zoltar theoretically suggests betting when the Vegas line is more than 3.0 points different from Zoltar’s prediction. There are four games and Zoltar has two hypothetical suggestions.
1. Zoltar likes the Vegas underdog Atlanta Falcons against the LA Rams. Vegas has the Rams as a significant favorite by 6.5 points, but Zoltar believes the Rams and the Falcons are evenly matched. A bet on the Falcons would pay off if the Falcons win outright, or if the Rams win but by less than 6.5 points (in other words, 6 points or less).
2. Zoltar likes the Vegas underdog Buffalo Bills against the Jacksonville Jaguars. Vegas has the Jaguars as 8.0 points better than the Bills, but Zoltar thinks the two teams are equal. A bet on the Bills will pay if the Bills win, or if the Jaguars win by 7 points or fewer.
Zoltar agrees closely with Vegas on the Kansas City Chiefs vs. the Tennessee Titans game (picking the Chiefs) and the New Orleans Saints vs. the Carolina Panthers game (picking the Saints to win).
(Way to go Zoltar! The one team I always love to see lose is Carolina because of their thoroughly obnoxious quarterback!)
Zoltar finished the last week of the regular season with a minor thud. Against the Vegas spread, which is what Zoltar is designed to predict, Zoltar went only 2-3. Zoltar correctly liked the underdog Titans and the underdog Chiefs. But Zoltar missed on the underdog Eagles, the underdog Raiders, and the favorite Rams.
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.
Just for fun, I also track how well Zoltar does when only predicting which team will win. This isn’t really useful except for parlay betting. For week #17, Zoltar was a mediocre 10-6 just predicting which team would win
For comparison purposes, I also track how well Bing Predicts and the Vegas line do when just predicting which team will win. In week #17, Bing was 9-7, and Vegas was also 9-7 when just predicting winners.
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.