Zoltar is my NFL prediction computer program. It uses a deep neural network and reinforcement learning. Here are Zoltar’s predictions for week #5 of the chaotic covid-19 2020 NFL season:

Zoltar:       bears  by    4  dog =  buccaneers    Vegas:  buccaneers  by  5.5
Zoltar:       bills  by    0  dog =      titans    Vegas:       bills  by    8
Zoltar:        jets  by    1  dog =   cardinals    Vegas:   cardinals  by  6.5
Zoltar:    patriots  by    9  dog =     broncos    Vegas:    patriots  by  8.5
Zoltar:    steelers  by    6  dog =      eagles    Vegas:    steelers  by  7.5
Zoltar:      chiefs  by   10  dog =     raiders    Vegas:      chiefs  by   12
Zoltar:     falcons  by    2  dog =    panthers    Vegas:     falcons  by  3.5
Zoltar:      ravens  by   18  dog =     bengals    Vegas:      ravens  by 13.5
Zoltar:      texans  by    6  dog =     jaguars    Vegas:      texans  by  6.5
Zoltar:        rams  by    4  dog =    redskins    Vegas:        rams  by    9
Zoltar: fortyniners  by   10  dog =    dolphins    Vegas: fortyniners  by  3.5
Zoltar:       colts  by    0  dog =      browns    Vegas:       colts  by  2.5
Zoltar:     cowboys  by    6  dog =      giants    Vegas:     cowboys  by  9.5
Zoltar:    seahawks  by    6  dog =     vikings    Vegas:    seahawks  by    7
Zoltar:      saints  by   11  dog =    chargers    Vegas:      saints  by  8.5

Zoltar theoretically suggests betting when the Vegas line is more than 3.0 points different from Zoltar’s prediction. For week #5 Zoltar has seven hypothetical suggestions.

The teams that Zoltar likes in week #5 are:

1. Underdog Bears against the Buccaneers
2. Underdog Titans against the Bills
3. Underdog Jets against the Cardinals
4. Favorite Ravens over the Bengals
5. Underdog Redskins against the Rams
6. Favorite 49ers over the Dolphins
7. Underdog Giants against the Cowboys

Note: Late line change on Saints-Chargers game.

Note: Several of these hypothetical suggestions don’t make any sense to me as a human. When I get some time, I need to tear Zoltar apart and see what’s going on. I suspect there is a fundamental change in the NFL this year: teams are scoring huge amounts of points relative to other years, which means that underdogs don’t have the mathematical advantage they have had in past seasons. And Zoltar has been programmed to love underdogs. Additionally, the Vegas point spreads are huge, and change wildly, and covid is causing havoc with rosters. Unfortunately, I don’t have the week or so it would take me to do a good analysis.

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.

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.

Zoltar did poorly in week #4. Against the Vegas point spread, Zoltar was 2-3 for the second week in a row. For the season, Zoltar is 14-11 (56.0%) against the spread. Just predicting winners, Zoltar was a terrible 8-7, one of his worst weeks ever. (There was one postponed game, Titans vs. the Steelers). Just picking winners, the Vegas line went 11-4 which is pretty good.

My system is named after the Zoltar fortune teller machine you can find in arcades. A couple of years ago I created a Little Zoltar machine (about 12 inches tall) that had speech recognition and could talk back to you. I was most proud of the little crystal ball I made that would flash red for “no” or green for “yes”. Sadly, Little Zoltar was lost during one of my many recent office moves at the huge tech company I work for.