NFL 2021 Week 18 Predictions – Zoltar Likes the Vegas Underdogs

Zoltar is my NFL football prediction computer program. It uses custom reinforcement learning and a neural network. Here are Zoltar’s predictions for week #18 of the 2021 season. It usually takes Zoltar about four weeks to hit his stride and takes humans about eight weeks to get up to speed, so weeks six through nine are usually Zoltar’s sweet spot. After week nine, injuries start having a big effect.

Zoltar:      chiefs  by    4  dog =     broncos    Vegas:      chiefs  by    3
Zoltar:     cowboys  by    0  dog =      eagles    Vegas:     cowboys  by  3.5
Zoltar:     bengals  by    0  dog =      browns    Vegas:      browns  by  1.5
Zoltar:     packers  by   11  dog =       lions    Vegas:     packers  by   11
Zoltar:      titans  by    5  dog =      texans    Vegas:      titans  by   10
Zoltar:       colts  by    6  dog =     jaguars    Vegas:       colts  by  8.5
Zoltar:     vikings  by    4  dog =       bears    Vegas:     vikings  by    6
Zoltar:    redskins  by    0  dog =      giants    Vegas:    redskins  by    4
Zoltar:      ravens  by    1  dog =    steelers    Vegas:      ravens  by  6.5
Zoltar:  buccaneers  by   11  dog =    panthers    Vegas:  buccaneers  by 16.5
Zoltar:    patriots  by    0  dog =    dolphins    Vegas:    patriots  by  2.5
Zoltar:      saints  by    0  dog =     falcons    Vegas:      saints  by  3.5
Zoltar:       bills  by   11  dog =        jets    Vegas:       bills  by   17
Zoltar:   cardinals  by    6  dog =    seahawks    Vegas:   cardinals  by    5
Zoltar:        rams  by    6  dog = fortyniners    Vegas:        rams  by  6.5
Zoltar:     raiders  by    4  dog =    chargers    Vegas:    chargers  by  2.5

Zoltar theoretically suggests betting when the Vegas line is “significantly” different from Zoltar’s prediction. In mid-season I usually use 3.0 points difference but for the first few weeks and last few weeks of the season I go a bit more conservative and use 4.0 points difference as the advice threshold criterion. In middle weeks I sometimes go ultra-aggressive and use a 1.0-point threshold.

Note: Because of Zoltar’s initialization (all teams regress to an average power rating) and other algorithms, Zoltar is much too strongly biased towards Vegas underdogs. I need to fix this.

For week #18:

1. Zoltar likes Vegas underdog Eagles against the Cowboys
2. Zoltar likes Vegas underdog Texans against the Titans
3. Zoltar likes Vegas underdog Giants against the Redskins
4. Zoltar likes Vegas underdog Steelers against the Ravens
5. Zoltar likes Vegas underdog Panthers against the Buccaneers
6. Zoltar likes Vegas underdog Falcons against the Saints
7. Zoltar likes Vegas underdog Jets against the Bills
8. Zoltar likes Vegas underdog Raiders against the Chargers

Update: Oops. I computed these advices based on the mid-season 3.0 threshold instead of the end-season 4.0 conservative threshold. The advice should have been:

2. Zoltar likes Vegas underdog Texans against the Titans
4. Zoltar likes Vegas underdog Steelers against the Ravens
5. Zoltar likes Vegas underdog Panthers against the Buccaneers
8. Zoltar likes Vegas underdog Raiders against the Chargers

For example, a bet on the underdog Eagles against the Cowboys will pay off if the Eagles win by any score, or if the favored Cowboys win but by less than the point spread of 3.5 points (in other words, by 3 points or less).

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.

In week #17, against the Vegas point spread, Zoltar went 8-6 (using the ultra-aggressive 1.0 points as the advice threshold). Overall, for the season, Zoltar is 62-50 against the spread (~55%).

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. In week #17, just predicting the winning team, Zoltar went 13-3 which is very good.

In week #17, just predicting the winning team, Vegas — “the wisdom of the crowd” went 13-3, the same as Zoltar.

Zoltar sometimes predicts a 0-point margin of victory, which means the two teams are evenly matched. There are five such games in week #18. In those situations, to pick a winner (only so I can track raw number of correct predictions) in the first few weeks of the season, Zoltar picks the home team to win. After that, Zoltar uses his algorithms to pick a winner.



My prediction system is math-based in the sense that it computes a numeric rating for each team and then uses ratings to compute the predicted margin of victory of the better team. An entirely different approach for predicting NFL football scores is to use a sophisticated simulation program and then simulate a game many thousands of times.

Before computers were easily available for football simulations, there were simulation games that used dice. Left: A very simple “Let’s Play Football” game. Center: “1st and Goal” is more complex and allows players to set plays. Right: This old “Half-Time Football” game is from 1979.


This entry was posted in Zoltar. Bookmark the permalink.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s