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 #8 of the 2024 season.

Zoltar:     vikings  by    0  opp =    chargers    | Vegas:    chargers  by    3
Zoltar:      texans  by    4  opp = fortyniners    | Vegas: fortyniners  by  1.5
Zoltar:     falcons  by    6  opp =    dolphins    | Vegas:     falcons  by  1.5
Zoltar:      ravens  by    6  opp =       bears    | Vegas:      ravens  by    7
Zoltar:       bills  by    3  opp =    panthers    | Vegas:       bills  by    7
Zoltar:     bengals  by    9  opp =        jets    | Vegas:     bengals  by  7.5
Zoltar:    patriots  by    6  opp =      browns    | Vegas:    patriots  by  5.5
Zoltar:      eagles  by   14  opp =      giants    | Vegas:      eagles  by 10.5
Zoltar:  buccaneers  by    5  opp =      saints    | Vegas:  buccaneers  by  5.5
Zoltar:     broncos  by    6  opp =     cowboys    | Vegas:     broncos  by  4.5
Zoltar:       colts  by   12  opp =      titans    | Vegas:       colts  by  3.5
Zoltar:     packers  by    0  opp =    steelers    | Vegas:     packers  by  1.5
Zoltar:      chiefs  by    6  opp =  commanders    | Vegas:      chiefs  by  3.5

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 or a very aggressive 2.0 points difference. 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 #8 Zoltar, using a typical 3.0 points threshold, likes two Vegas underdogs and three Vegas favorites:

fortyniners  at       texans: Bet on Vegas underdog texans
dolphins     at      falcons: Bet on Vegas favorite falcons
bills        at     panthers: Bet on Vegas underdog panthers
giants       at       eagles: Bet on Vegas favorite eagles
titans       at        colts: Bet on Vegas favorite colts

For example, a bet on the underdog Texans against the 49ers will pay off if the Texans win by any score, or if the favored 49ers win but by less than 1.5 points (in other words, by 1 point). If the 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 likely have changed. I’ve noticed that, compared to last year, point spreads are changing more dramatically by late Tuesdays than they used to. 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.

For example, this week, the Colts opened as just 3.5 points favorites against the Titans, but by Tuesday morning, the line had shifted by 11 points to Colts -14.5 points!

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.

In week #7, against the Vegas point spread, Zoltar went a not-so-good 2-2 (using 3.0 points as the advice threshold). If Zoltar had used an aggressive 2.0 points as the advice threshold, he would have gone 5-3 against the spread. Darn. For the season, against the spread, Zoltar is 18-11 (~62% 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 #7, just predicting the winning team, Zoltar went only 8-2 (with 5 games too close to call) which is pretty good. Vegas went 11-4 at just predicting the winning team.

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

Right: Uncle Sam’s Philosophy machine from the late 1800s was produced by the Automatic Amusement Co. of London. The spinner pointed to one of 18 rhyming predictions such as, “To-morrow going down the street, an ancient lover you will meet.” The machine reads “Consultation Fee 1D” which confused me until I discovered that D, from “denarius” (Latin for coin) was the old UK abbreviation for penny.