Zoltar is my NFL football prediction computer program. It uses custom reinforcement learning and a neural network. Here are Zoltar’s predictions for week #16 of the 2021 season. It usually takes Zoltar about four weeks to hit his stride and takes humans about eight weeks to get up to speed, so weeks six through nine are usually Zoltar’s sweet spot. After week nine, injuries start having a big effect.

Zoltar:      titans  by    6  dog = fortyniners    Vegas:      titans  by    1
Zoltar:     packers  by    7  dog =      browns    Vegas:     packers  by    7
Zoltar:   cardinals  by    2  dog =       colts    Vegas:   cardinals  by    5
Zoltar:     falcons  by    6  dog =       lions    Vegas:     falcons  by  6.5
Zoltar:  buccaneers  by    4  dog =    panthers    Vegas:  buccaneers  by   11
Zoltar:      ravens  by    0  dog =     bengals    Vegas:     bengals  by  2.5
Zoltar:    chargers  by    2  dog =      texans    Vegas:    chargers  by 10.5
Zoltar:        rams  by    0  dog =     vikings    Vegas:        rams  by  3.5
Zoltar:    patriots  by    4  dog =       bills    Vegas:    patriots  by    2
Zoltar:        jets  by    5  dog =     jaguars    Vegas:        jets  by  1.5
Zoltar:      eagles  by    6  dog =      giants    Vegas:      eagles  by    9
Zoltar:    seahawks  by    6  dog =       bears    Vegas:    seahawks  by    7
Zoltar:      chiefs  by    6  dog =    steelers    Vegas:      chiefs  by  9.5
Zoltar:     raiders  by    5  dog =     broncos    Vegas:     raiders  by    1
Zoltar:     cowboys  by    6  dog =    redskins    Vegas:     cowboys  by  9.5
Zoltar:      saints  by    4  dog =    dolphins    Vegas:      saints  by  3.5

Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. In mid-season I usually use 3.0 points difference but for the first few weeks and last few weeks of the season I go a bit more conservative and use 4.0 points difference as the advice threshold criterion. In middle weeks I sometimes go ultra-aggressive and use a 1.0-point threshold.

Note: Because of Zoltar’s initialization (all teams regress to an average power rating) and other algorithms, Zoltar is much too strongly biased towards Vegas underdogs. I need to fix this.

For week #16:

1. Zoltar likes Vegas underdog Panthers against the Buccaneers
2. Zoltar likes Vegas underdog Texans against the Chargers
3. Zoltar likes Vegas underdog Vikings against the Rams
4. Zoltar likes Vegas underdog Steelers against the Chiefs
5. Zoltar likes Vegas underdog Redskins against the Cowboys
6. Zoltar likes Vegas favorite Titans over the 49ers
7. Zoltar likes Vegas favorite Jets over the Jaguars
8. Zoltar likes Vegas favorite Raiders over the Broncos

For example, a bet on the underdog Panthers against the Buccaneers will pay off if the Panthers win by any score, or if the favored Buccaneers win but by less than the point spread of 11.0 points (in other words, by 10 points or less). If the Buccaneers win by exactly 11 points, the bet is a push.

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 #15, against the Vegas point spread, Zoltar went 3-2 (using the standard 3.0 points as the advice threshold. Overall, for the season, Zoltar is 51-44 against the spread (~53%).

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 #15, just predicting the winning team, Zoltar went 10-6 which is about average.

In week #15, just predicting the winning team, Vegas — “the wisdom of the crowd” also went 10-6, the same as Zoltar.

Zoltar sometimes predicts a 0-point margin of victory, which means the two teams are evenly matched. There are two such games in week #16. 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 prediction system is named after the Zoltar fortune teller machine you can find in arcades.

Right: I’ve experimented with adapting Zoltar’s underlying prediction algorithms to college basketball. That system is named Zoltara.