Zoltar is my NFL football prediction computer program. It uses reinforcement learning and a neural network. Here are Zoltar’s predictions for week #11 of the 2022 season.
Zoltar: packers by 4 dog = titans Vegas: packers by 3
Zoltar: falcons by 5 dog = bears Vegas: falcons by 3
Zoltar: bills by 6 dog = browns Vegas: bills by 9.5
Zoltar: colts by 4 dog = eagles Vegas: eagles by 8
Zoltar: commanders by 0 dog = texans Vegas: commanders by 2.5
Zoltar: rams by 0 dog = saints Vegas: saints by 3.5
Zoltar: patriots by 6 dog = jets Vegas: patriots by 3
Zoltar: giants by 4 dog = lions Vegas: giants by 3
Zoltar: ravens by 6 dog = panthers Vegas: ravens by 11.5
Zoltar: raiders by 0 dog = broncos Vegas: broncos by 2.5
Zoltar: cowboys by 0 dog = vikings Vegas: cowboys by 1
Zoltar: chiefs by 0 dog = chargers Vegas: chiefs by 7
Zoltar: steelers by 2 dog = bengals Vegas: bengals by 5
Zoltar: cardinals by 1 dog = fortyniners Vegas: fortyniners by 8
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 #11 Zoltar likes five Vegas underdogs:
1. Zoltar likes Vegas underdog Colts against the Eagles
2. Zoltar likes Vegas underdog Panthers against the Ravens
3. Zoltar likes Vegas underdog Chargers against the Chiefs
4. Zoltar likes Vegas underdog Steelers against the Bengals.
5. Zoltar likes Vegas underdog Cardinals against the 49ers.
For example, a bet on the underdog Colts against the Eagles will pay off if the Colts win by any score, or if the favored Eagles win but by less than 8 points (i.e., 7 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.
Zoltar’s prediction that the Colts will beat the Eagles is mystifying to me. The Eagles loss to the Commanders in week #10 must have hurt the Eagles a lot, at least in terms of Zoltar’s algorithms.
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 #10, against the Vegas point spread, Zoltar went 4-1 (using 4.0 points as the advice threshold). I was impressed that Zoltar picked the underdog Steelers and underdog Packers to win outright, which both teams did.
For the season, against the spread, Zoltar is 36-17 (~68% 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 #10, just predicting the winning team, Zoltar went only 8-6 which isn’t very good — just slightly better than a coin flip. Vegas was even worse at 6-8 — slightly worse than a coin flip at just predicting the winning team.
Zoltar sometimes predicts a 0-point margin of victory. There are five such games in week #11. In those situations, to pick a winner (only so I can track raw number of correct predictions) in the first few weeks of the season, Zoltar picks the home team to win. After that, Zoltar uses his algorithms to pick a winner.
My system is named after the Zoltar fortune teller machine you can find in arcades. Coin-operated fortune teller machines have been around for well over 100 years.
Left: The Blind Man fortune teller machine was made about 1920 by Walter Hart of Ramsgate, England. It accepted a penny. The blind man spins and his cane points to a message such as, “Go, get thee a wife. She’ll mind thy folly.”
Right: The Columbian Fortune Teller was manufactured about 1892 by the World’s Fair Slot Machine Company. It was a combination fortune teller and early slot machine. It accepted a nickel, which was quite a bit of money back then — roughly equivalent to about $5.00 today.


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