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 #7 of the 2023 season.

Zoltar:     jaguars  by    0  dog =      saints    Vegas:      saints  by    3
Zoltar:     raiders  by    0  dog =       bears    Vegas:     raiders  by    3
Zoltar:      browns  by    0  dog =       colts    Vegas:      browns  by    2
Zoltar:       bills  by    5  dog =    patriots    Vegas:       bills  by    8
Zoltar:  commanders  by    0  dog =      giants    Vegas:  commanders  by    2
Zoltar:      ravens  by    2  dog =       lions    Vegas:      ravens  by    3
Zoltar:  buccaneers  by    6  dog =     falcons    Vegas:  buccaneers  by  2.5
Zoltar:    steelers  by    0  dog =        rams    Vegas:        rams  by    3
Zoltar:    seahawks  by   10  dog =   cardinals    Vegas:    seahawks  by  7.5
Zoltar:     packers  by    0  dog =     broncos    Vegas:     packers  by    1
Zoltar:      chiefs  by    8  dog =    chargers    Vegas:      chiefs  by  5.5
Zoltar:      eagles  by    6  dog =    dolphins    Vegas:      eagles  by  1.5
Zoltar: fortyniners  by    0  dog =     vikings    Vegas: fortyniners  by    7 

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

For week #7, using 3.0 points as the threshold, Zoltar likes two Vegas favorites and one Vegas underdog:

falcons      at   buccaneers : bet favorite buccaneers
dolphins     at       eagles : bet favorite eagles
fortyniners  at      vikings : bet underdog Vikings

For example, a bet on the underdog Vikings against the 49ers will pay off if the Vikings win by any score, or if the favored 49ers win but by less than 7 points (in other words, by 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”.

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 #6, against the Vegas point spread, Zoltar went a so-so 2-1 (using 3.0 points as the advice threshold).

For the season, against the spread, Zoltar is 21-15 (~58% 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 #6, just predicting the winning team, Zoltar went only 11-4 which is OK but not great. Vegas also was 11-4 at just predicting the winning team.

Zoltar sometimes predicts a 0-point margin of victory. There are seven such games in week #7 – over half of the 13 games scheduled. This is unusual and suggests that the NFL is more balanced this year than usual.

Football has been popular in the U.S. for a long time. Here are two old pinball style games that have a football theme. Left: “Quarterback” by Genco (1957). Players launch a ball towards the back wall that has receptacles to score points. One of the receptacles scores no points but drops the ball down to the lower playing surface where there are scoring opportunities. This game wasn’t too popular and was only sold for a few years.

Right: “Army-Navy” by Rock-Ola (1934). This was an amazing, purely mechanical (no electricity) game. A ball is launched into the top playing surface where it drops into a hole that scores yards for Army or Navy. The yellow ball-marker on the lower playing field automatically moves the associated number of yards. When the ball-marker reaches an end zone, 7 points are scored for the appropriate team. Fascinating!