Zoltar is my computer program that predicts the results of NFL football games. The 2024 season finished a few weeks ago when the slight-underdog Philadelphia Eagles (1.5 points) beat the slight-favorite Kansas City Chiefs by a score of 40-22 on Sunday, February 9, 2025 in Super Bowl LIX.

Last week I looked back at the 2024 season and analyzed the strategy of betting on the home team when that team is the Vegas favorite. This week, one morning before work, I figured I look at the other naive betting strategy: bet on the home team when that team is the Vegas underdog.

Suppose the Rams are playing at home against the Seahawks, and Vegas has the Rams as underdogs by 7.0 points (“Rams +7” in Vegas terminology — if you’re not familiar with betting it probably seems odd that “+” means an underdog). A bet on the home underdog Rams will win if the Rams win by any score, or if the favored Seahawks win but by less than the 7.0 point spread. If the favored Seahawks win by exactly the point spread of 7, for example, 27-20, then all wagers are a push (tie) and money is returned to bettors.

My analysis shows that, for the 2024 season, betting on the home underdog team would have won 43 times and lost 60 times, for an accuracy of 43/103 which is about 42%. This is not even close to break-even because of the house edge. Typically, you must bet $110 in order to win just $100, which is a house advantage of about 53%, but in practical terms you really must win with about 60% accuracy to be profitable.

Some sample output from the analysis:

Team 26 saints is a home underdog
Home/underdog score = 19
Visitor/favorite score =  20
Visitor/favorite pred margin victory = 3.0
Home dog lost as expected
But favorite didn't cover spread!

. . .

Team 22 raiders is a home underdog
Home/underdog score = 20
Visitor/favorite score =  34
Visitor/favorite pred margin victory = 5.5
Home dog lost as expected
And favorite covered the spread

. . .

Team 17 jets is a home underdog
Home/underdog score = 32
Visitor/favorite score =  20
Visitor/favorite pred margin victory = 1.0
Home dog won outright!

=====

Total times bet on Vegas home underdog wins against spread = 43
Total times bet on Vegas home underdog loses against spread = 60

The analysis of bet-home-underdog strategy also gives you the bet-visitor-favorite results: win 60 times, lose 43 times = 58% accuracy. This is mathematically better than break-even, but in practical terms not good enough to make a consistent profit. However, this 58% accuracy on the bet visitor-favorite is the best simple strategy accuracy I’ve seen in many seasons. It will be interesting to see how well the bet visitor-favorite strategy holds up next season.

The results did not completely surprise me. Using any of the four simple strategies — bet home favorite, bet home underdog, bet visitor favorite, bet visitor underdog — depends on the accuracy of the Vegas point spread. Because millions of people bet hundreds of millions of dollars on NFL football games, overall, the Vegas point spread is remarkably accurate.

If any simple strategy worked (meaning win more than 60% of the time), then it would be well known, and the Vegas point spread would instantly adapt and then the strategy wouldn’t work.

Anyway, good fun and a nice look back at the 2024 NFL football season.

The Super Bowl has become so huge, that TV advertisements now cost millions of dollars for just 30 seconds, and so ads are not nearly as creative as they used to be. During Super Bowl 54, five years ago, the Reese’s candy company had a hilarious ad for a new “Take 5” candy bar. The ad took place in a typical office and played off of common phrases for a naive person.

Super funny!