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 #13 of the 2025 season.

Zoltar:       lions  by    6  opp =     packers    | Vegas:       lions  by  2.5
Zoltar:      chiefs  by    4  opp =     cowboys    | Vegas:      chiefs  by  3.5
Zoltar:      ravens  by   10  opp =     bengals    | Vegas:      ravens  by  5.5
Zoltar:      eagles  by    7  opp =       bears    | Vegas:      eagles  by  6.5
Zoltar:     jaguars  by    4  opp =      titans    | Vegas:     jaguars  by  1.5
Zoltar:        rams  by    4  opp =    panthers    | Vegas:        rams  by  3.5
Zoltar: fortyniners  by    4  opp =      browns    | Vegas: fortyniners  by    6
Zoltar:       colts  by    4  opp =      texans    | Vegas:      texans  by  1.5
Zoltar:    dolphins  by    6  opp =      saints    | Vegas:    dolphins  by    6
Zoltar:     falcons  by    0  opp =        jets    | Vegas:     falcons  by  1.5
Zoltar:  buccaneers  by    7  opp =   cardinals    | Vegas:  buccaneers  by  3.5
Zoltar:    seahawks  by    6  opp =     vikings    | Vegas:    seahawks  by  1.5
Zoltar:    chargers  by   11  opp =     raiders    | Vegas:    chargers  by  6.5
Zoltar:       bills  by    0  opp =    steelers    | Vegas:       bills  by  4.5
Zoltar:     broncos  by    3  opp =  commanders    | Vegas:  commanders  by  2.5
Zoltar:    patriots  by   11  opp =      giants    | Vegas:    patriots  by    3

Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. During the early and later parts of the season, I typically use a threshold of 4 points, and in the middle of the season I use a threshold of 3 points.

For week #13, using a relatively aggressive 3.0 points as the advice threshold, Zoltar likes six Vegas favorites and three Vegas underdog:

packers      at        lions: Bet on Vegas favorite lions
bengals      at       ravens: Bet on Vegas favorite ravens
texans       at        colts: Bet on Vegas underdog colts
cardinals    at   buccaneers: Bet on Vegas favorite buccaneers
vikings      at     seahawks: Bet on Vegas favorite seahawks
raiders      at     chargers: Bet on Vegas favorite chargers
bills        at     steelers: Bet on Vegas underdog steelers
broncos      at   commanders: Bet on Vegas underdog broncos
giants       at     patriots: Bet on Vegas favorite patriots

Using a more conservative 4.0 points as the threshold, Zoltar drops his recommendations for Vegas favorites Lions and Buccaneers. Using a very conservative 5.0 points as the threshold, Zoltar recommends only Vegas underdogs Colts and Broncos, and favorite Patriots.

A bet on the underdog Colts against the Texans will pay off if the Colts win by any score, or if the favored Texans 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. Compared to previous years, point spreads are changing wildly by Tuesday. Swings of 10 points are not uncommon.

For example, the Commanders opened as 2.5 favorites against the Broncos, but just a few hours later the Commanders became huge 6.5 underdogs — a swing of 9 points. Other wild swings occurred in Jaguars-Titans, Rams-Panther, Texans-Colts, Vikings-Seahawks

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.

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 logistics and careless errors (data entry, etc.)

In week #12, against the Vegas point spread, Zoltar went 5-1 using 3.0 points as the advice threshold. This is a good result but is based on opening point spreads.

For the season, against the spread, Zoltar is 32-20 (~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 #12, just predicting the winning team, Zoltar went 7-2, with 5 games too close for Zoltar to have an opinion. Vegas was 9-5 at just predicting the winning team.

My system is named after the Zoltar fortune teller machine you can find in arcades. That machine is named after the Zoltar machine from the 1988 movie “Big”. And the 1988 Zoltar was named after the “Zoltan” arcade fortune teller from the 1960s.

Right: The “Zodi” machine (circa 1940) actually typed out a fortune using a pneumatic powered typewriter. Also amazing technology.