Zoltar is my NFL football prediction computer program. It uses a neural network and a type of reinforcement learning. Here are Zoltar’s predictions for week #19 (wildcard playoff games) of the 2024 season.

Zoltar:        rams  by    4  opp =    panthers    | Vegas:        rams  by   10
Zoltar:       bears  by    2  opp =     packers    | Vegas:       bears  by  1.5
Zoltar:       bills  by    0  opp =     jaguars    | Vegas:       bills  by  1.5
Zoltar:      eagles  by    6  opp = fortyniners    | Vegas:      eagles  by  3.5
Zoltar:    patriots  by    2  opp =    chargers    | Vegas:    patriots  by  3.5
Zoltar:      texans  by    0  opp =    steelers    | Vegas:      texans  by  3.5

Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. For the beginning and end of the season, I use a relatively conservative threshold of 4.0 points difference. For the middle of the season, I use a relatively aggressive threshold of 3.0 points difference.

For week #19, five of Zoltar’s six predictions are less than 4.0 points difference from the Vegas point spreads and so Zoltar has no wagering recommendations on those games. But there’s a big discrepancy between Zoltar and Las Vegas for the Rams at Panthers game. Las Vegas picks the Rams to win by 10 points, but Zoltar thinks the Rams will win, but by only 4 points — a 6-point difference. Therefore, Zoltar suggests a wager on the underdog Panthers.

A hypothetical wager on the Panthers would win if the Panthers win by any sore, or if the favored Rams win but by less than 10.0 points. If the Rams win by exactly 10 points, the bet is a push

The annoying possibility of pushes are why point spreads often have a “point five”, such as 3.5, because then pushes aren’t possible.

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 to account for overhead and things like data entry errors.

In week #18, the final regular season week, against the Vegas point spread, Zoltar went 2-0 using 4.0 points as the advice threshold.

For the regular season, against the spread, Zoltar finished with a record of 46-30 (~60% accuracy). Not bad, but not great.

Just for fun, I track how well Zoltar does when just trying to predict just which team will win a game. This isn’t useful except for parlay betting. In week #18, just predicting the winning team, Zoltar went 10-4, with two games to close for Zoltar to make a prediction. In week #18, Vegas was 10-6 at just predicting the winning team.

For the season, just predicting the winning team, Zoltar finished 135-71 (with 66 games too close to call) for just over 65% accuracy. Vegas finished 164-107 (~60% accuracy)

My prediction system is named after the Zoltar fortune teller machine you can find in arcades. Fortune telling with a crystal ball has been around for over 2,000 years. It was described by ancient Roman Pliny the Elder as “crystallum orbis”. Almost all cultures have some form of fortune teller. Here are three results of Internet images searches, but I doubt that the costumes are historically accurate Left: “Russian fortune teller”. Center: “Chinese fortune teller”. Right: “Egyptian fortune teller”.