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 2018 NFL season:

Zoltar:    patriots  by   11  dog =       colts    Vegas:    patriots  by   10
Zoltar:      titans  by    0  dog =       bills    Vegas:      titans  by    3
Zoltar:    panthers  by   11  dog =      giants    Vegas:    panthers  by  6.5
Zoltar:     bengals  by    5  dog =    dolphins    Vegas:     bengals  by    6
Zoltar:      ravens  by    6  dog =      browns    Vegas:      ravens  by    3
Zoltar:       lions  by    2  dog =     packers    Vegas:     packers  by  1.5
Zoltar:      chiefs  by    6  dog =     jaguars    Vegas:      chiefs  by  3.5
Zoltar:     broncos  by    0  dog =        jets    Vegas:        jets  by    2
Zoltar:    steelers  by    6  dog =     falcons    Vegas:    steelers  by  3.5
Zoltar:    chargers  by    6  dog =     raiders    Vegas:    chargers  by  5.5
Zoltar:      eagles  by    4  dog =     vikings    Vegas:      eagles  by    3
Zoltar:        rams  by    1  dog =    seahawks    Vegas:        rams  by    7
Zoltar: fortyniners  by    4  dog =   cardinals    Vegas: fortyniners  by    4
Zoltar:     cowboys  by    3  dog =      texans    Vegas:      texans  by  3.5
Zoltar:      saints  by    6  dog =    redskins    Vegas:      saints  by  6.5

(Note: I noticed I botched my title — Zoltar likes four Vegas underdogs.)

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

1. Zoltar likes the Vegas favorite Panthers against the Giants. Zoltar thinks the Panthers are 11 points better than the Giants but Vegas has the Panthers favored only by 6.5 points so Zoltar thinks the Panthers will cover the spread.

2. Zoltar likes the Vegas underdog Lions against the Packers. Zoltar thinks the Lions are 2 points better than the Packers but Vegas has the Packers as 1.5-point favorites. A bet on the Lions will pay off if the Lions win outright, or if the Packers win by less than 1.5 points (in other words, by exactly 1 point), or if the game is a tie.

3. Zoltar likes the Vegas underdog Seahawks against the Rams. Zoltar thinks the Rams are only 1 point better than the Seahawks but Vegas has the Rams favored by 7 points, therefore, Zoltar thinks the Rams will win but not cover the spread.

4. Zoltar likes the Vegas underdog Cowboys against the Texans. Zoltar thinks the Cowboys are 3 points better than the Texans but Vegas thinks the Texans are 3.5 points better than the Cowboys. This is a pretty big difference of opinion — usually means a player injury of some sort.

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 (not by how many points). This isn’t useful except for parlay betting.

Zoltar sometimes predicts a 0-point margin of victory. There two such games in week #5: Bills-Titans and Broncos-Jets. In the first four weeks of the season, Zoltar picks the home team to win. After week #4, Zoltar uses historical data for the current season (which usually, but not always, ends up in a prediction that the home team will win).

==

Zoltar did OK in week #4. Against the Vegas point spread, Zoltar was 3-2. For the season so far, against the Vegas spread Zoltar is 14-9 which is not quite 61% accuracy.

Just predicting winners, Zoltar was a very good 12-3. Vegas went 11-4 to ever-so slightly underperform Zoltar. For the season, Zoltar is 41-20 (67% accuracy) and Vegas is 40-20 (67% accuracy).

My system is named after the arcade fortune teller machine, which in turn is named after a machine in the 1988 fantasy movie “Big” starring Tom Hanks.