Zoltar is my NFL prediction computer program. It uses a deep neural network and reinforcement learning. The results for the first half of the season were completely crazy. The covid-19 pandemic had a massive effect on the NFL. I switched Zoltar into experimental mode starting in week #8. For example, I changed the definition of a rout-win from winning by 20+ points to winning by 22+ points. The changes made Zoltar more conservative with regards to his betting strategy.
To me, the most interesting parameter is the home field advantage. For decades, both humans and Zoltar have given 3 points to the home team. But with covid-19, you can argue that the home field advantage is dramatically greater (say 5 or 6 points or more), or dramatically smaller (say 0 or 1 point), or even negative. I won’t know until after the 2020 season is completed and I have enough data to analyze.
Using the experimental parameters, here are Zoltar’s predictions for week #18 (wild card playoff season games) of the 2020 NFL season:
Zoltar: bills by 5 dog = colts Vegas: bills by 7 Zoltar: seahawks by 5 dog = rams Vegas: seahawks by 4.5 Zoltar: buccaneers by 1 dog = redskins Vegas: buccaneers by 8 Zoltar: ravens by 1 dog = titans Vegas: ravens by 3.5 Zoltar: saints by 7 dog = bears Vegas: saints by 9.5 Zoltar: steelers by 5 dog = browns Vegas: steelers by 4
Old Zoltar theoretically used to suggest betting when the Vegas line is more than 3.0 points different from Zoltar’s prediction. At this point, it’s beginning to look like an improved strategy will be something like bet on Vegas underdogs only if Zoltar and Vegas differ by more than 6 points, or bet on Vegas favorites if Zoltar and Vegas differ by more than 2 points. I’ll have to do an analysis after this season is over.
When you bet on an underdog, your bet pays off if the underdog wins by any score, or if the game is a tie, or if the favorite team wins but by more than the Vegas point spread. If the favorite team wins by exactly the point spread, the bet is a push. You lose your bet if the favorite wins by more than the Vegas point spread.
Theoretically, if you must bet $110 to win $100 (typical in Vegas) then you’ll make money if you predict at 53% accuracy or better. But realistically, you need to predict at 60% accuracy or better.
Because experimental Zoltar is very conservative, he mostly agrees with the Vegas point spreads in week #18. However, Zoltar seems to like Washington (Redskins) against the Tampa Bay Buccaneers. Vegas says the Buccaneers will win by 8.0 points but Zoltar thinks the Buccaneers are only 1.0 point better. So, a bet on Washington will pay off if Washington wins by any score, or if the Buccaneers win but by less than 8 points. If the Buccaneers win by exactly 8 points, all bets are a push.
Against the Vegas point spread last week, experimental Zoltar was a very good 4-1. Note that in two games (Chiefs-Chargers, Eagles-Redskins) one team had nothing to gain (Chiefs, Eagles) and so held their starters out of the game and therefore Zoltar didn’t have an opinion on those two games.
For the regular 2020 season, Zoltar finished 43-28 (60.5% accuracy) against the Las Vegas point spread. This is a bit worse than most seasons.
For fun, I track how well Zoltar and the Vegas point spreads do when just predicting which team will win (not by how much). In week #17, just predicting the winning team, Zoltar was a decent 12-4. The Las Vegas point spread was better at 13-3 just predicting winners.
For the season, just predicting which team would win, Zoltar finished 175-80 (68.6% accuracy) and Las Vegas finished nearly the same at 173-82 (67.8% accuracy).
My system is named after the Zoltar fortune teller machine you can find in arcades. Fortune telling machines were some of the very first coin-operated machines. Left: Mystic Mirror from the 1920s had a mutoscope screen (an early motion picture device) to display fortunes. Amazing. Center: A Roovers Brothers company Puss in Boots machine from about 1899. There are many reproductions but originals are extremely valuable to collectors. Right: A Clawson dice rolling fortune teller from about 1895. You play a nickel, and six dice in the two cups are rolled, which determine your fortune. It was really a gambling machine placed in bars, where the bartender would manually pay out winning results (4, 5 or 6 of a kind).