Zoltar is my NFL football prediction computer program. It uses a neural network and a type of reinforcement learning. Here are Zoltar’s predictions for week #9 of the 2024 season.

Zoltar:      texans  by    4  dog =        jets    Vegas:        jets  by    2
Zoltar:     cowboys  by    0  dog =     falcons    Vegas:     falcons  by    2
Zoltar:       bills  by    8  dog =    dolphins    Vegas:       bills  by    6
Zoltar:      saints  by    4  dog =    panthers    Vegas:      saints  by    4
Zoltar:     bengals  by    6  dog =     raiders    Vegas:     bengals  by  7.5
Zoltar:      browns  by    5  dog =    chargers    Vegas:      browns  by    1
Zoltar:  commanders  by    0  dog =      giants    Vegas:  commanders  by    2
Zoltar:      titans  by    4  dog =    patriots    Vegas:      titans  by    3
Zoltar:      ravens  by    6  dog =     broncos    Vegas:      ravens  by  9.5
Zoltar:       bears  by    0  dog =   cardinals    Vegas:       bears  by    1
Zoltar:      eagles  by    8  dog =     jaguars    Vegas:      eagles  by    6
Zoltar:       lions  by    0  dog =     packers    Vegas:       lions  by    2
Zoltar:    seahawks  by    2  dog =        rams    Vegas:        rams  by    1
Zoltar:     vikings  by    4  dog =       colts    Vegas:     vikings  by    6
Zoltar:      chiefs  by    8  dog =  buccaneers    Vegas:      chiefs  by    8

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 #9, using my typical 3.0 points threshold, Zoltar likes two Vegas underdogs and one favorite:

texans       at         jets : bet underdog texans
chargers     at       browns : bet favorite browns
broncos      at       ravens : bet underdog broncos

But none of these three suggestions are very strong.

A bet on the Vegas underdog Texans against the Jets will pay off if the Texans win by any score, or if the favored Jets win but by less than 2.0 points (i.e., exactly 1 point). If the favored Jets wins by exactly the point spread of 2 points, the wager is a push. This is why point spreads often have a 0.5 added — called “the hook” — to eliminate pushes.

Note: I noticed a very large difference this week between the opening Vegas point spreads and the current (Tuesday) spreads. For example, in the Browns vs. Chargers game, the opening line had the Browns favored by 4.0 points and then a few hours later, the Chargers were favored by 2.0 points, and then a few hours later the Browns were favored by 1.0 points. An analysis of opening point spreads versus later point spreads is something I’d like to do if I ever get free time.

Week #9 often marks the beginning of the part of the season where injuries start to become a big factor.

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 #8, against the Vegas point spread, Zoltar went a good 6-2 (using 3.0 points as the advice threshold).

For the season, against the spread, Zoltar is 27-17 (~61% 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 #8, just predicting the winning team, Zoltar went 10-6 which is so-so. Vegas also went 10-6, at just predicting the winning team.

My Zoltar system is named after the Zoltar fortune teller machine you can find in arcades. Halloween is this week. Here are three clever Zoltar-inspired do-it-yourself Halloween costumes.