Zoltar is my NFL prediction computer program. It uses a deep neural network and reinforcement learning. The results for the first half of the season were completely crazy. The covid-19 pandemic has had a massive effect on the NFL.

I switched Zoltar into experimental mode starting in week #8. I fiddled with five of Zoltar’s parameters:

Base rating pts for a win: 16 to 14
Rout definition: win by 20+ to win by 22+
Rout rating pts: 4 to 2
Underdog away win bonus pts: 4 to 2
Home field advantage: 3 pts to 2 pts

These changes made Zoltar more conservative with regards to his betting strategy.

The most interesting parameter is the home field advantage. For decades, both humans and Zoltar have given 3 points to the home team. But with covid-19, you can argue that the home field advantage is dramatically greater (say 5 or 6 points or more), or dramatically smaller (say 0 or 1 point), or even negative. I won’t know until after the 2020 season is completed and I have enough data.

Using the experimental parameters, here are Zoltar’s predictions for week #10 of the 2020 NFL season:

Zoltar:      titans  by    3  dog =       colts    Vegas:      titans  by    2
Zoltar:      browns  by    4  dog =      texans    Vegas:      browns  by    3
Zoltar:       lions  by    5  dog =    redskins    Vegas:       lions  by    1
Zoltar:  buccaneers  by    1  dog =    panthers    Vegas:  buccaneers  by    6
Zoltar:      eagles  by    1  dog =      giants    Vegas:      eagles  by  3.5
Zoltar:     packers  by   10  dog =     jaguars    Vegas:     packers  by   14
Zoltar:       bills  by    1  dog =   cardinals    Vegas:   cardinals  by    2
Zoltar:    dolphins  by    5  dog =    chargers    Vegas:    dolphins  by    3
Zoltar:     raiders  by    5  dog =     broncos    Vegas:     raiders  by    5
Zoltar:    seahawks  by    1  dog =        rams    Vegas:        rams  by  5.5
Zoltar:    steelers  by   10  dog =     bengals    Vegas:    steelers  by  7.5
Zoltar:      saints  by    5  dog = fortyniners    Vegas:      saints  by  9.5
Zoltar:      ravens  by    1  dog =    patriots    Vegas:      ravens  by    3
Zoltar:       bears  by    1  dog =     vikings    Vegas:     vikings  by    3

Old Zoltar theoretically used to suggest betting when the Vegas line is more than 3.0 points different from Zoltar’s prediction. It’s not clear how that advice parameter will change.

When you bet on an underdog your, bet pays off if the underdog wins by any score, or if the game is a tie, or if the favorite team wins but by more than the Vegas point spread. If the favorite team wins by exactly the point spread, the bet is a push. You lose your bet if the favorite wins by more than the Vegas point spread.

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.

Experimental Zoltar likes three Vegas underdogs in week #10:

1. Underdog Panthers vs. Buccaneers
2. Underdog Seahawks vs. Rams
3. Underdog 49ers vs. Saints

Last week, experimental Zoltar’s predictions were good and he went 5-2 against the spreads, correctly liking underdogs Jaguars, Bills, Cowboys, Saints, Jets; but incorrectly liking underdogs Broncos and Bears. I used a criterion to place a hypothetical bet if Zoltar and the Vegas point spread differed by more than 4 points.

Just predicting the winning team, Zoltar was a pretty good 11-3 which was a bit better than the Las Vegas point spread 9-5 just picking the winner.

Just like everyone else Zoltar will be very glad when covid goes away.

My system is named after the Zoltar fortune teller machine you can find in arcades. Zoltar used to be obscure and not known to many people until about two years ago when Zoltar started appearing in many TV commercials for an insurance company. Zoltar is made by a company called “Characters Unlimited, Inc.” from Boulder City, Nevada. The company makes custom machines too, like Confucius and Old Pirate here. It would be cool to have a Zoltar machine made that looks like you. Maybe.