Zoltar is my NFL football prediction computer program. It uses a neural network and a form of reinforcement learning. Here are Zoltar’s predictions for week #15 of the 2025 season.

Zoltar:  buccaneers  by    6  opp =     falcons    | Vegas:  buccaneers  by    4
Zoltar:       bears  by   11  opp =      browns    | Vegas:       bears  by  7.5
Zoltar:      texans  by   11  opp =   cardinals    | Vegas:      texans  by  7.5
Zoltar:      chiefs  by    4  opp =    chargers    | Vegas:      chiefs  by  5.5
Zoltar:      ravens  by    0  opp =     bengals    | Vegas:      ravens  by  2.5
Zoltar:     jaguars  by    9  opp =        jets    | Vegas:     jaguars  by    6
Zoltar:       bills  by    0  opp =    patriots    | Vegas:       bills  by    2
Zoltar:  commanders  by    0  opp =      giants    | Vegas:      giants  by  2.5
Zoltar:      eagles  by   18  opp =     raiders    | Vegas:      eagles  by 11.5
Zoltar:        rams  by    1  opp =       lions    | Vegas:        rams  by    4
Zoltar:    seahawks  by    6  opp =       colts    | Vegas:    seahawks  by 10.5
Zoltar: fortyniners  by   11  opp =      titans    | Vegas: fortyniners  by   10
Zoltar:     broncos  by    4  opp =     packers    | Vegas:     packers  by    1
Zoltar:    panthers  by    0  opp =      saints    | Vegas:    panthers  by  2.5
Zoltar:     vikings  by    0  opp =     cowboys    | Vegas:     cowboys  by    3
Zoltar:    steelers  by    6  opp =    dolphins    | Vegas:    steelers  by    3

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

For week #15, using 4.0 points as the threshold, Zoltar likes two Vegas underdogs and one Vegas favorite:

raiders      at       eagles: Bet on Vegas favorite eagles
colts        at     seahawks: Bet on Vegas underdog colts
packers      at      broncos: Bet on Vegas underdog broncos

I noticed that Zoltar thinks that the Eagles are an enormous 18-points better than the Raiders. As far as I can recall, that is the largest predicted margin of victory that Zoltar has ever computed.

As has been the case in previous weeks, the opening point spread on Monday evening varies wildly from the point spread on Tuesday morning.

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 more dramatically by late Tuesdays than they used to. A swing of 7 points is not uncommon.

I speculate that these 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 logistics and careless mistakes like data entry errors.

In week #14, against the Vegas point spread, Zoltar went 3-1 using 4.0 points as the advice threshold. Good results but if a few balls bounced another way, Zoltar could easily have gone 0-4 last week. Stay humble Zoltar.

For the season, against the spread, Zoltar is 39-24 (~61% accuracy). Interestingly, last year after week #14, Zoltar was also 39-24 — exactly the same.

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 #14, just predicting the winning team, Zoltar went 7-4 with 3 games too close for Zoltar to offer an opinion. Vegas was similar at just predicting the winning team, going 8-6.

My prediction system is named after the Zoltar fortune teller machine. I love prediction systems and I love old pinball machines. Here are two old NFL football theme pinballs.

Right: The “Monday Night Football” game by Data East was produced in 1989. It is solid-state electrical and cost about $3000 at the time. I give it a B grade.