Carolina beat Arizona 33-30 in Canton on Thursday night, on a five-yard scramble by an undrafted rookie quarterback with the clock showing zero. The Hall of Fame Game opened the 2026 preseason with 63 combined points. The total opened at 35.5.
This site previewed the game on Monday, arguing that the preseason trend lists circulating every August rest on samples too small to carry weight. That piece closed by saying the result would land in somebody’s database and turn into next year’s trend. It will. The figures are worth writing down now, before anyone repackages them.
Where the Money Went Before Kickoff
DraftKings and BetMGM both opened Carolina as a 1.5-point favorite with the total at 35.5. VegasInsider tracked the early action and found the Under drawing 66 percent of bets and 74 percent of the money at BetMGM as of Tuesday afternoon. DraftKings shortened its number to 34.5, Over -115. Those are early figures rather than closing ones, which is worth holding onto, though the direction was never ambiguous. Books moved the total down and the money kept arriving on the Under.
Then the two teams scored 63 points. That lands 27.5 clear of the 35.5 opener, and 28.5 clear of the number DraftKings settled on.
A Trend With Sixteen Games Behind It
The Under was popular for a documented reason. Previewing last year’s game, CBS Sports reported that the Under had hit in 12 of the previous 16 Hall of Fame Games. Nobody invented that. Twelve of sixteen is 75 percent, and a figure like that reads as settled.
Sixteen games is also the whole problem, because a sample that size produces confident percentages it cannot support. Consider what has happened since the number was published. The 2025 game finished 34-7, which totals 41, clearing the 32.5 FanDuel had posted. Thursday’s finished on 63. Add both to the sixteen and the Under sits at 12 of 18, with the two most recent results going the other way.
The 2024 game is in that sample as well. Chicago beat Houston 21-17 before severe weather ended the thing permanently in the third quarter. A totals record absorbs a game like that without ever mentioning that a quarter went unplayed.
The Side Lost Too
Arizona was the popular team as much as the Under was the popular number. VegasInsider had 60 percent of spread bets and 93 percent of spread dollars on the Cardinals at BetMGM, with 73 percent of moneyline bets and 81 percent of moneyline cash pointing the same direction. Carolina moved from -1.5 to -1 at BetMGM by Tuesday afternoon.
Arizona nearly delivered. Kedon Slovis found Bryson Green in the corner of the end zone with 1:55 left for a 30-27 lead, on a play first ruled incomplete and then overturned on review. Haynes King, undrafted out of Georgia Tech, took Carolina back the other way and ran it in as time expired. So the 93 percent lost the spread and lost the game outright.
Preseason handle is thin and book limits sit well below regular season levels, which means 93 percent of the money describes a small pool of it. That caveat cuts both ways. It weakens the split as evidence about where informed money sits, and it makes August a poor place to go looking for public betting patterns anyone should act on.
What the Hall of Fame Game Result Is Worth
Not much, and consistency requires saying so out loud. A single game that contradicts a trend does not disprove it, in the same way the trend was never proof of itself. Both are small samples wearing percentage signs.
What Thursday does show is the mechanism underneath the number. A preseason total is a guess about rotations. NFL.com described a field of second and third stringers, and at kickoff nobody outside those two buildings knew how long Carson Beck or Kenny Pickett would play, or who would be taking snaps in the fourth quarter. Fading the public assumes the price carries an opinion worth fading. In August the price mostly carries a shrug.
Next summer someone will post that the Under has hit in 12 of the last 18 Hall of Fame Games. That will be accurate. It will also be eighteen games, one of which a storm ended in the third quarter, and one of which finished with a quarterback nobody drafted running into the end zone at zero.