Zoltar is my NFL football prediction computer program. It uses a neural network and a type of specialized reinforcement learning. Here are Zoltar’s predictions for week #16 of the 2025 season.

Zoltar:    seahawks  by    4  opp =        rams    | Vegas:        rams  by  1.5
Zoltar:     packers  by    0  opp =       bears    | Vegas:     packers  by  1.5
Zoltar:      eagles  by    4  opp =  commanders    | Vegas:      eagles  by    6
Zoltar:    chargers  by    2  opp =     cowboys    | Vegas:     cowboys  by    2
Zoltar:      saints  by    6  opp =        jets    | Vegas:      saints  by    3
Zoltar:      chiefs  by    9  opp =      titans    | Vegas:      chiefs  by    4
Zoltar:  buccaneers  by    0  opp =    panthers    | Vegas:  buccaneers  by    3
Zoltar:       bills  by   12  opp =      browns    | Vegas:       bills  by    9
Zoltar:     vikings  by    5  opp =      giants    | Vegas:     vikings  by    3
Zoltar:    dolphins  by    6  opp =     bengals    | Vegas:     bengals  by  2.5
Zoltar:     falcons  by    0  opp =   cardinals    | Vegas:     falcons  by  2.5
Zoltar:     broncos  by    8  opp =     jaguars    | Vegas:     broncos  by    3
Zoltar:      texans  by   18  opp =     raiders    | Vegas:     texans   by    9
Zoltar:       lions  by    6  opp =    steelers    | Vegas:       lions  by    6
Zoltar:      ravens  by    4  opp =    patriots    | Vegas:      ravens  by    3
Zoltar:       colts  by    2  opp = fortyniners    | Vegas: fortyniners  by  6.5

Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. In the beginning and end parts of the season, I use a conservative 4 points difference as the threshold. In the middle part of the season I use a more aggressive 3 points difference as the threshold.

For week #16 Zoltar likes Vegas favorites Chiefs and Broncos, and Vegas underdogs Seahawks, Dolphins, Texans, Colts:

rams         at     seahawks: Bet on Vegas underdog seahawks
chiefs       at       titans: Bet on Vegas favorite  chiefs
bengals      at     dolphins: Bet on Vegas underdog dolphins
jaguars      at      broncos: Bet on Vegas favorite broncos
raiders      at       texans: Bet on Vegas favorite texans
fortyniners  at        colts: Bet on Vegas underdog colts

For example, a bet on the underdog Seahawks against the Rams will pay off if the Seahawks win by any score, or if the favored Rams win but by less than 1.5 points (i.e., 1 point). If a favored team wins by exactly the point spread, the wager is a push. This is why point spreads often have a 0.5 added — called “the hook” — to eliminate pushes.

I use the early Vegas point spreads, usually posted on late Monday night, right after the Monday Night Football game. By the time you read this, the point spreads will certainly have changed. I’ve noticed that, compared to last year, point spreads are changing wildly by late Tuesday. A swing of 7 points is not uncommon.

I speculate that the wild swings are due to a huge increase in betting. When a lot of money is bet on one team in a matchup, the bookmakers must make a huge change in the point spread to encourage betting on the other team. Bookmakers only make money when the betting amounts are close to equal on both teams.

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 take into account overhead and things like data entry error.

In week #15, against the Vegas point spread, Zoltar went 3-0 using 4.0 points as the advice threshold. Zoltar correctly liked Vegas favorite Eagles (they destroyed the Raiders), and Vegas underdogs Colts (they only lost by 3 to the Seahawks) and Broncos (they won outright against the Packers).

For the season, against the spread, Zoltar is 42-24 (~63% 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 #15, just predicting the winning team, Zoltar went 9-2 with 5 games too close for Zoltar to express an opinion.

Vegas was 10-6 at just predicting the winning team. Not good but not terrible either.

My system is named after the Zoltar fortune teller machine you can find in arcades. Many years ago (1971-73), the Kellogg’s breakfast products company put rules for an NFL simulation into their Pop Tarts snacks (introduced in 1964). The simulation used a deck of cards as the random number generator. For example, the player who has the ball decides to run or pass and then picks a card from the shuffled deck. Suppose the player chooses run and picks a 3 of clubs. The result is a 12-yard gain. A 3 of diamonds, hearts, or spades results in a 5-yard loss.