NFL 2018 Week 8 Predictions – Zoltar Looks to Rebound

Zoltar is my NFL prediction computer program. It uses a deep neural network and Reinforcement Learning. Here are Zoltar’s predictions for week #8 of the 2018 NFL season:

Zoltar:      texans  by    1  dog =    dolphins    Vegas:      texans  by    7
Zoltar:      eagles  by    3  dog =     jaguars    Vegas:      eagles  by    3
Zoltar:    panthers  by    6  dog =      ravens    Vegas:      ravens  by    1
Zoltar:       bears  by    4  dog =        jets    Vegas:       bears  by  6.5
Zoltar:     bengals  by    6  dog =  buccaneers    Vegas:     bengals  by    6
Zoltar:       lions  by    4  dog =    seahawks    Vegas:       lions  by  2.5
Zoltar:      chiefs  by   11  dog =     broncos    Vegas:      chiefs  by   10
Zoltar:    redskins  by    4  dog =      giants    Vegas:    redskins  by    1
Zoltar:    steelers  by   16  dog =      browns    Vegas:    steelers  by  7.5
Zoltar:     raiders  by    4  dog =       colts    Vegas:       colts  by    1
Zoltar:   cardinals  by    6  dog = fortyniners    Vegas: fortyniners  by    1
Zoltar:        rams  by   10  dog =     packers    Vegas:        rams  by    9
Zoltar:     vikings  by    4  dog =      saints    Vegas:     vikings  by  1.5
Zoltar:    patriots  by    4  dog =       bills    Vegas:    patriots  by   13

Zoltar theoretically suggests betting when the Vegas line is more than 3.0 points different from Zoltar’s prediction. For week #8 Zoltar has six hypothetical suggestions.

1. Zoltar likes the Vegas underdog Dolphins against the Texans. Zoltar thinks the Texans are just 1 point better than the Dolphins, but Vegas has the Texans as favorites by 7.0 points. Therefore, Zoltar thinks the Texans will win but not cover the spread.

2. Zoltar likes the Vegas underdog Panthers against the Ravens. Zoltar thinks the Panthers are 6 points better than the Ravens but Vegas has the Ravens as 1-point favorites.

3. Zoltar likes the Vegas favorite Steelers against the Browns. Zoltar thinks the Steelers are a whopping 16 points better than the Browns but Vegas thinks the Steelers are only 7.5 points better than the Browns.

4. Zoltar likes the Vegas underdog Raiders against the Colts. Zoltar thinks the Raiders are 4 points better than the Colts but Vegas thinks the Colts are 1.0 point better than the Raiders.

5. Zoltar likes the Vegas underdog Cardinals against the 49ers. Zoltar thinks the Cardinals are 6 points better than the 49ers but Vegas thinks the 49ers are 1.0 point better than the Cardinals.

6. Zoltar inexplicably likes the Vegas underdog Bills against the Patriots. Zoltar thinks the Patriots are just 4 points better than the Bills but Vegas has the Patriots favored by 13.0 points. (Update: I had a bad data file. Zoltar has the Patriots better than the Bills by 10 points, so Zoltar does not recommend a bet on this game).

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 are no such games in week #8. 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 didn’t do very well in week #7. Against the Vegas point spread, Zoltar was just 1-2. For the season so far, against the Vegas spread Zoltar is 23-13 which is not quite 64% accuracy.

Just predicting winners, Zoltar was a so-so 10-4. Vegas went 11-3 just predicting which team would win. For the season, Zoltar is 75-30 (71% accuracy) and Vegas is 74-30 (71% accuracy).

Left: My system is named after the Zoltar fortune teller machine you can find in arcades. With Halloween coming up, I always like to see Zoltar costumes.

This entry was posted in Machine Learning, Zoltar. Bookmark the permalink.

4 Responses to NFL 2018 Week 8 Predictions – Zoltar Looks to Rebound

  1. mainjb says:

    Hi there,

    You said zoltar went 10-4 in week 7 but i see that ypu went 9-5. Was this a typo?

    • It depends on how you score one of the games . . . I don’t remember which one (I think it involved the Bills) but in week #7, the quarterback was out in one game and Zoltar adjusted his prediction in mid-week by 6 points which flipped the predicted winner in that game. I’ll double-check when I get the chance.

  2. mirskontsa says:

    Did you make any money from betting with these programs? : )

    • I almost never bet, except maybe $20 or so on the annual football Super Bowl, or a few dollars on the annual college basketball championship. I enjoy the intellectual challenge of prediction.

      And I enjoy watching games on TV, but if I bet on the games, I’d be too nervous to enjoy the game I think.

      When I’m in Vegas, I will sometimes play Blackjack or Texas Holdem Ultimate — two games that have state and can be defeated by using mathematics.

Comments are closed.