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 #14 of the 2025 season.

Zoltar:       lions  by    8  opp =     cowboys    | Vegas:       lions  by    6
Zoltar:     vikings  by    6  opp =  commanders    | Vegas:  commanders  by  1.5
Zoltar:    seahawks  by    4  opp =     falcons    | Vegas:    seahawks  by    7
Zoltar:      ravens  by    6  opp =    steelers    | Vegas:      ravens  by  8.5
Zoltar:      browns  by    6  opp =      titans    | Vegas:      browns  by  1.5
Zoltar:       colts  by    0  opp =     jaguars    | Vegas:     jaguars  by    3
Zoltar:    dolphins  by    0  opp =        jets    | Vegas:    dolphins  by  1.5
Zoltar:  buccaneers  by   11  opp =      saints    | Vegas:  buccaneers  by  9.5
Zoltar:       bills  by   10  opp =     bengals    | Vegas:       bills  by  5.5
Zoltar:     broncos  by   12  opp =     raiders    | Vegas:     broncos  by  3.5
Zoltar:     packers  by    6  opp =       bears    | Vegas:     packers  by  3.5
Zoltar:        rams  by    5  opp =   cardinals    | Vegas:        rams  by  1.5
Zoltar:      chiefs  by    6  opp =      texans    | Vegas:      chiefs  by  6.5
Zoltar:      eagles  by    0  opp =    chargers    | Vegas:      eagles  by  2.5

Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. I typically use a conservative threshold of 4.0 points during the early and last parts of the season, and a more aggressive 3.0 points threshold during the middle of the season. Week #14 is on the borderline between middle and late.

For week #14 Zoltar likes three Vegas favorites and one Vegas underdog (using 4.0 points as the threshold):

commanders   at      vikings: Bet on Vegas underdog vikings
titans       at       browns: Bet on Vegas favorite browns
bengals      at        bills: Bet on Vegas favorite bills
broncos      at      raiders: Bet on Vegas favorite broncos

A bet on the Vegas underdog Vikings against the Commanders will pay off if the Vikings win by any score, or if the favored Commanders 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. This year point spreads are often changing wildly.

A swing of 7 points is not uncommon. I speculate that this is 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.

This week, the Seahawks opened as 1.5-point underdogs to the Falcons, but moved to 7.5-point favorites — a swing of 9 points. The Colts opened as 3-point underdogs to the Jaguars but moved to 2.5-point favorites. There were also big swings in Broncos-Raiders, Texans-Chiefs, Bears-Packers, and Rams-Cardinals.

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 like data entry errors.

In week #13, against the Vegas point spread, using 4.0 points as the advice threshold, Zoltar went 4-3 which isn’t very good.

For the season, against the spread, Zoltar is 36-23 (~61% 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 #13, just predicting the winning team, Zoltar went 8-6, with 2 games too close to judge. Not very good. Vegas was about the same, going 9-7 at just predicting the winner.

My system is named after the Zoltar fortune teller machine you can find in arcades. A few years ago I created a Little Zoltar machine (about 12 inches tall) that had speech recognition and could give you predictions if you asked something like, “Will the Rams beat the Chiefs?” I was most proud of the little crystal ball I made that would flash red for “no” or green for “yes”. Sadly, Little Zoltar was lost during one of my many office moves at the huge tech company I worked for.