Zoltar is my NFL football prediction computer program. It uses a neural network and a form of quasi-reinforcement learning. Here are Zoltar’s predictions for week #6 of the 2025 season.

Zoltar:      eagles  by   12  opp =      giants    | Vegas:      eagles  by    7
Zoltar:     broncos  by    8  opp =        jets    | Vegas:     broncos  by  6.5
Zoltar:      ravens  by    2  opp =        rams    | Vegas:      ravens  by  5.5
Zoltar:     cowboys  by    0  opp =    panthers    | Vegas:     cowboys  by  1.5
Zoltar:       colts  by    6  opp =   cardinals    | Vegas:   cardinals  by  1.5
Zoltar:    seahawks  by    0  opp =     jaguars    | Vegas:     jaguars  by  1.5
Zoltar:    chargers  by    1  opp =    dolphins    | Vegas:    chargers  by  1.5
Zoltar:    steelers  by   11  opp =      browns    | Vegas:    steelers  by  6.5
Zoltar:    patriots  by    0  opp =      saints    | Vegas:    patriots  by    3
Zoltar:     raiders  by    4  opp =      titans    | Vegas:     raiders  by    3
Zoltar:  buccaneers  by    6  opp = fortyniners    | Vegas:  buccaneers  by  1.5
Zoltar:     packers  by    6  opp =     bengals    | Vegas:     packers  by  2.5
Zoltar:       lions  by    0  opp =      chiefs    | Vegas:      chiefs  by    3
Zoltar:       bills  by    1  opp =     falcons    | Vegas:       bills  by  5.5
Zoltar:  commanders  by    9  opp =       bears    | Vegas:  commanders  by  4.5

Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. In mid-season I typically use 3.0 points difference but for the first few weeks of the season I am a bit more conservative and use 4.0 points difference as the advice threshold criterion.

For week #6 Zoltar, using a mildly aggressive 3.0 points as the advice threshold, likes three Vegas underdogs and five favorites:

eagles       at       giants: Bet on Vegas favorite eagles
rams         at       ravens: Bet on Vegas underdog rams
cardinals    at        colts: Bet on Vegas underdog colts
browns       at     steelers: Bet on Vegas favorite steelers
fortyniners  at   buccaneers: Bet on Vegas favorite buccaneers
bengals      at      packers: Bet on Vegas favorite packers
bills        at      falcons: Bet on Vegas underdog falcons
bears        at   commanders: Bet on Vegas favorite commanders

For example, a bet on the Vegas underdog Rams against the Ravens will pay off if the Rams win by any score, or if the favored Ravens win but by less than 5.5 points (in other words, by 5 points or fewer).

This is an interesting part of the season for Zoltar. Zoltar has enough data to start getting more accurate, but humans have a lot of information too, including injuries, team chemistry, and so on. With 8 recommendations, this will be a make-or-break week for 2025 Zoltar.

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 mistakes).

In week #5, against the Vegas point spread, Zoltar had a poor 2-4 record (using 3.0 points as the advice threshold). Zoltar would have been 3-3 except for the incredible blunders by the Arizona Cardinals, where a running back deliberately dropped the ball inches before he crossed the goal line, and an interception-fumble-ball-kicked-into-end-zone play, and a snap into the QB’s face. Agh!

For the season, against the spread, Zoltar is 10-7 (58% accuracy).

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 #5, just predicting the winning team, Zoltar went only 4-8 (with 2 no-call games) which is horrible. Vegas also had an off week, going only 5-9 at just predicting the winning team.

Left: My system is named after the Zoltar fortune teller machine you can find in arcades.

Right: The mini fortune cards are reproductions of fortunes from old machines produced in the 1950s.