Zoltar is my NFL football prediction computer program. It uses reinforcement learning and a neural network. Here are Zoltar’s predictions for week #18 (last regular season games) of the 2022 season.
Zoltar: chiefs by 5 dog = raiders Vegas: chiefs by 7.5
Zoltar: titans by 0 dog = jaguars Vegas: jaguars by 6.5
Zoltar: buccaneers by 2 dog = falcons Vegas: falcons by 4
Zoltar: bills by 6 dog = patriots Vegas: bills by 7.5
Zoltar: vikings by 7 dog = bears Vegas: vikings by 4
Zoltar: bengals by 7 dog = ravens Vegas: bengals by 6
Zoltar: colts by 6 dog = texans Vegas: colts by 2.5
Zoltar: chargers by 4 dog = broncos Vegas: chargers by 3
Zoltar: packers by 6 dog = lions Vegas: packers by 4.5
Zoltar: dolphins by 6 dog = jets Vegas: dolphins by 2.5
Zoltar: saints by 6 dog = panthers Vegas: saints by 3.5
Zoltar: eagles by 8 dog = giants Vegas: eagles by 14
Zoltar: steelers by 6 dog = browns Vegas: steelers by 3
Zoltar: seahawks by 4 dog = rams Vegas: seahawks by 6.5
Zoltar: fortyniners by 11 dog = cardinals Vegas: fortyniners by 14
Zoltar: cowboys by 4 dog = commanders Vegas: cowboys by 5.5
Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. For this season I’ve been using a threshold of 4 points difference but in some previous seasons I used 3 points.
At the beginning of the season, because of Zoltar’s initialization (all teams regress to an average power rating) and other algorithms, Zoltar is very strongly biased towards Vegas underdogs. I probably need to fix this. For week #18 Zoltar likes three Vegas underdogs:
1. Zoltar likes Vegas underdog Titans against the Jaguars.
2. Zoltar likes Vegas underdog Buccaneers against the Falcons.
3. Zoltar likes Vegas underdog Giants against the Eagles.
For example, a bet on the underdog Titans against the Jaguars will pay off if the Titans win by any score, or if the favored Jaguars win but by less than 6.5 points (i.e., 6 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.
These predictions use the score of the Bills (3) vs. Bengals (7) game which was suspended because of an injury to a player. Update: The game was ultimately canceled. I revised the result data but this didn’t change Zoltar’s advice.
At the end of the season, predictions get very dicey. Injuries have accumulated and teams that have nothing to gain often sit key players to avoid injuries.
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 #17, against the Vegas point spread, Zoltar went 2-1 using 4.0 points as the advice threshold.
For the season, against the spread, Zoltar is 56-33 (~63% 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 #17, just predicting the winning team, Zoltar went 12-4 which is quite good. Vegas was 10-6 at just predicting the winning team.

My football prediction system is named after the Zoltar fortune teller machine you can find in arcades. Fortune teller machines have been around for over 100 years. Center: One of the most famous machines among collectors is “The Gypsy”, made in about 1908 by Mills Novelty. The machine would speak a fortune by playing a wax cylinder style record player. Only one is known to exist. Right: “Princess Doraldina” was made about 1928 by the Doraldina Fortune Telling Machine Company, of Rochester, New York. The machine dispensed a card with a fortune.

.NET Test Automation Recipes
Software Testing
SciPy Programming Succinctly
Keras Succinctly
R Programming
2026 Visual Studio Live
2025 Summer MLADS Conference
2026 DevIntersection Conference
2025 Machine Learning Week
2025 Ai4 Conference
2026 G2E Conference
2026 iSC West Conference
You must be logged in to post a comment.