Zoltar is my NFL prediction computer program. It uses a deep neural network and reinforcement learning. Here are Zoltar’s predictions for week #18 (the Wild Card round) of the 2019 NFL season:

Zoltar:      texans  by    6  dog =       bills    Vegas:      texans  by    3
Zoltar:    patriots  by    6  dog =      titans    Vegas:    patriots  by  5.5
Zoltar:      saints  by    9  dog =     vikings    Vegas:      saints  by    8
Zoltar:    seahawks  by    0  dog =      eagles    Vegas:    seahawks  by  1.5

Zoltar theoretically suggests betting when the Vegas line is more than 3.0 points different from Zoltar’s prediction. For week #18 Zoltar has no hypothetical suggestions because his predictions agree closely with the Las Vegas point spread data.

The largest difference of opinion is on the Houston Texans vs. Buffalo Bills game. Las Vegas has the Texans favored by 3.0 points but Zoltar thinks the Texans are 6.0 points better than the Bills.

If Zoltar was forced to make suggestions, he’d advise bets on Las Vegas favorites Texans (over the Bills), Patriots (over the Titans), Saints (over the Vikings), and Vegas underdog Eagles (against the Seahawks).

Update: The Vegas point spread for the Texans went from -3 to -2.5 over the Bills so Zoltar (barely) recommends a bet on the Vegas favorite Texans.

===

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.

Zoltar was OK in week #17, going 2-1 against the Vegas point spread. Zoltar correctly liked Vegas favorite Chiefs over the Chargers and Vegas dog Ravens against the Steelers, but missed by picking the underdog Seahawks against the 49ers. The Seahawks choked by incurring a delay of game penalty with 22 seconds left and holding the ball on the 49ers 1-yard line. Utter incompetence. This must have enraged many thousands of bettors.

For the 2019 season, through week #17 (the end of the regular season), Zoltar finished an OK 53-33 (61.6% accuracy) against the Vegas spread.

Just for fun, I track how well Zoltar and Las Vegas do when just trying to predict only which team will win (but not by how much). This isn’t useful except for parlay betting.

Just predicting winners, Zoltar was a so-so 10-4. Las Vegas was also 10-4 last week. Two games were NA because teams held out key players to avoid injury.

For the regular season Zoltar finished a pretty decent 171-82 (67%) just picking winners and Vegas finished 163-87 (65%).

Note: Vegas has fewer games than Zoltar because Vegas had three pick’em games. Also there was one tie game in week #1 (Lions at Cardinals).

My system is named after the Zoltar fortune teller machine you can find in arcades. Traditionally, fortune tellers have been women, as shown by these old photographs I found on the Internet. The photos all look authentic to me but it’s possible to make realistic recreations.