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 #6 of the 2023 season.

Zoltar:      chiefs  by   15  dog =     broncos    Vegas:      chiefs  by 10.5
Zoltar:      ravens  by    3  dog =      titans    Vegas:      ravens  by  3.5
Zoltar:     falcons  by    4  dog =  commanders    Vegas:     falcons  by  2.5
Zoltar:     vikings  by    4  dog =       bears    Vegas:     vikings  by  2.5
Zoltar:     bengals  by    4  dog =    seahawks    Vegas:     bengals  by    3
Zoltar: fortyniners  by    4  dog =      browns    Vegas: fortyniners  by    6
Zoltar:      saints  by    0  dog =      texans    Vegas:      saints  by    1
Zoltar:     jaguars  by    6  dog =       colts    Vegas:     jaguars  by  4.5
Zoltar:    dolphins  by   10  dog =    panthers    Vegas:    dolphins  by 13.5
Zoltar:       lions  by    0  dog =  buccaneers    Vegas:       lions  by    3
Zoltar:     raiders  by    2  dog =    patriots    Vegas:     raiders  by    3
Zoltar:      eagles  by    5  dog =        jets    Vegas:      eagles  by    7
Zoltar:        rams  by    6  dog =   cardinals    Vegas:        rams  by  6.5
Zoltar:       bills  by   10  dog =      giants    Vegas:       bills  by   14
Zoltar:     cowboys  by    0  dog =    chargers    Vegas:     cowboys  by    2

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 but 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 #6 Zoltar likes two Vegas underdogs and one favorite:

broncos      at       chiefs : bet favorite chiefs
panthers     at     dolphins : bet underdog panthers
giants       at        bills : bet underdog giants

For example, a bet on the underdog Panthers against the Dolphins will pay off if the Panthers win by any score, or if the favored Dolphins win but by less than 13.5 points (in other words, by 13 points or fewer).

This is an interesting part of the season for Zoltar. Zoltar has enough data to start getting more accurate, but humans have a lot of information too, including injuries, team chemistry, and so on.

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 #5, against the Vegas point spread, Zoltar went a poor 1-2 (using a conservative 4.0 points as the advice threshold).

For the season, against the spread, Zoltar is 19-14 (~57% 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 #5, just predicting the winning team, Zoltar went only 8-6 which is not very good. Vegas was slightly worse at 7-7 at just predicting the winning team, which is just a flip of the coin.

My system is named after the Zoltar fortune teller machine you can find in arcades. Zoltar is manufactured by Characters Unlimited, Inc., in Boulder City, Nevada, about 30 miles south of Las Vegas. The company makes several variations, including Confucius and Old Pirate. You can even get a custom machine made with your own face. Zoltar machines have many options. A more or less standard Zoltar machine costs about $9000.