Zoltar is my NFL prediction computer program. It uses a deep neural network and reinforcement learning. The results for the first seven weeks have been absolutely crazy. The covid pandemic has had a massive effect on the NFL.

I have switched Zoltar into experimental mode. For week #8 predictions I fiddled with five of Zoltar’s key parameters:

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

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, 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 disastrous 2020 NFL season ends and I have enough data.

Using the experimental parameters, here are Zoltar’s predictions for week #8 of the chaotic covid-19 2020 NFL season:

Zoltar:    panthers  by    4  dog =     falcons    Vegas:    panthers  by    3
Zoltar:        rams  by    1  dog =    dolphins    Vegas:        rams  by  3.5
Zoltar:       bills  by    5  dog =    patriots    Vegas:       bills  by  3.5
Zoltar:      chiefs  by   10  dog =        jets    Vegas:      chiefs  by 19.5
Zoltar:      browns  by    3  dog =     raiders    Vegas:      browns  by  2.5
Zoltar:     packers  by    6  dog =     vikings    Vegas:     packers  by  6.5
Zoltar:       colts  by    1  dog =       lions    Vegas:       colts  by    3
Zoltar:      titans  by    4  dog =     bengals    Vegas:      titans  by  5.5
Zoltar:      ravens  by    4  dog =    steelers    Vegas:      ravens  by  3.5
Zoltar:     broncos  by    5  dog =    chargers    Vegas:    chargers  by    3
Zoltar:    seahawks  by    4  dog = fortyniners    Vegas:    seahawks  by    3
Zoltar:      saints  by    1  dog =       bears    Vegas:      saints  by  2.5
Zoltar:      eagles  by    5  dog =     cowboys    Vegas:      eagles  by  7.5
Zoltar:  buccaneers  by    3  dog =      giants    Vegas:  buccaneers  by 10.5

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.

Left: My system is named after the Zoltar fortune teller machine you can find in arcades. Apparently there are fortune tellers from many countries, at least according to a search of stock photos for Halloween costumes. From left to right: Russian fortune teller, Chinese fortune teller, Egyptian fortune teller. The authenticity of these costumes is questionable.