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 #12 of the 2024 season.

Zoltar:       bills  by    0  opp =      texans    | Vegas:       bills  by  3.5
Zoltar:    steelers  by    0  opp =       bears    | Vegas:       bears  by  2.5
Zoltar:     packers  by    4  opp =     vikings    | Vegas:     packers  by  3.5
Zoltar:      chiefs  by    6  opp =       colts    | Vegas:      chiefs  by  9.5
Zoltar:    seahawks  by   10  opp =      titans    | Vegas:    seahawks  by  1.5
Zoltar:      ravens  by   11  opp =        jets    | Vegas:      ravens  by 11.5
Zoltar:    patriots  by    0  opp =     bengals    | Vegas:     bengals  by  4.5
Zoltar:       lions  by   18  opp =      giants    | Vegas:       lions  by  8.5
Zoltar:     raiders  by    3  opp =      browns    | Vegas:     raiders  by  3.5
Zoltar:     jaguars  by    0  opp =   cardinals    | Vegas:   cardinals  by    3
Zoltar:      eagles  by    5  opp =     cowboys    | Vegas:      eagles  by  4.5
Zoltar:     falcons  by    0  opp =      saints    | Vegas:     falcons  by  2.5
Zoltar:        rams  by    6  opp =  buccaneers    | Vegas:        rams  by  2.5
Zoltar: fortyniners  by    6  opp =    panthers    | Vegas: fortyniners  by    7

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

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. Compared to last year, point spreads are changing more dramatically by late Tuesdays than they used to. A wild swing of 7 points is not uncommon.

For example, this week, the Chiefs went from massive 9.5-point favorites against the Colts, to mild 3.5-point favorites. The Seahawks opened as weak 1.5-point favorites against the Titans, but by Tuesday morning, the Seahawks jumped to 12.5-point favorites — a gigantic 11-point change. And the Bengals went from 4.5-point favorites against the Patriots to 7.5-point underdogs! A wild 12-point swing in a few hours.

I speculate that these wild swings are due in part 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. And towards the end of the season, injuries become a big factor too.

For Week #12, using 3.0 points as the threshold, and the opening point spreads, Zoltar likes three Vegas underdogs and three Vegas favorites:

bills        at       texans: Bet on Vegas underdog texans
colts        at       chiefs: Bet on Vegas underdog colts
seahawks     at       titans: Bet on Vegas favorite seahawks
patriots     at      bengals: Bet on Vegas underdog patriots
giants       at        lions: Bet on Vegas favorite lions
buccaneers   at         rams: Bet on Vegas favorite rams

For example, a bet on the underdog Texans against the Bills will pay off if the Texans win by any score, or if the favored Bills win but by less than 3.5 points (i.e., 3 points or less). 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 am utterly baffled by the point spread for the Bills vs. Texans game. In week #11, the Bills looked unstoppable when they destroyed a very good Buccaneers team, while the Texans barely squeaked by the unimpressive Titans. And then Vegas favors the Bills by only 3.5 points. Zoltar likes the Texans to keep the game close, but my human eye says the Bills will destroy the Texans. We’ll see. Update: I just watched the game — Zoltar was correct — the underdog Texans beat the Bills by a score of 23-19. I should have had more confidence in Zoltar.

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 (if you’re actually betting money), you need to predict at 60% accuracy or better, to account for logistics and careless mistakes (data entry, etc.)

In week #11, against the Vegas point spread, Zoltar went 2-1 (using 3.0 points as the advice threshold). Zoltar would have been 3-0 if the horrible-this-year Vikings hadn’t choked away a win against the Bears on a field goal with no time left on the clock. Argh.

If Zoltar had been very aggressive in week #11 and used 2.0 points as his advice threshold, he would have gone 4-1 against the spread.

For the season, against the spread, Zoltar is 27-19 (~58% 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 #11, just predicting the winning team, Zoltar went 8-4 with 3 games too close to call. Vegas went 9-6 at just predicting the winning team, which is typical for this part of the season.

My system is named after the Zoltar fortune teller machine you can find in arcades. Arcade Zoltar uses a crystal ball to make his predictions. Crystal balls were a common plot device in movies of the 1930s, including the 44 Charlie Chan films produced from 1931 to 1949.

Right: In “Black Magic” (1944), Chan solves a murder committed at a seance where the bullet disappears and no gun is found. The weapon was a disguised cigar case with bullets made of frozen blood. Yikes!