High-Total Under Parlays: 11 Years of Real Results

Parlays built from unders on high NFL totals have not made money over the last eleven seasons. From 2015 through 2025, unders on closing totals of 48 points or more won 51.44 percent of 764 decided regular-season games and returned -0.13 percent per dollar. That is close to a coin flip after the vig, not an edge. Pairing every two of those unders played in the same week produced 1,575 tickets that hit 24.63 percent of the time at a median payout near +278, for a return of -6.85 percent. The Best Bet on Sports has built a verified $367,520+ in profit across all sportsbooks over more than twenty years, and we test a betting rule like this one before we call it an edge.

"Fade the big total" is one of the most repeated lines in football betting. The story goes that the public loves points, books shade totals upward to meet that demand, and a patient bettor collects by taking the under whenever the number climbs into the high 40s or 50s. Stack two of those unders together and the ticket pays close to 3-to-1.

It is a tidy theory. This post checks it against every high total the NFL has posted since 2015.

How Was the Study Built?

We used nflverse game data covering every regular-season NFL game from 2015 through 2025, 2,895 games in all with a closing total and a final score. Each under was graded against its closing total at its listed closing price. Where no price was listed, we used -110. Pushes were removed, leaving 2,869 decided games.

A "high total" here means a closing number of 48 or more. That threshold captures roughly the top quarter of totals in the sample and lines up with the point where bettors start calling a number inflated.

For the parlay test, we paired every high-total under with every other high-total under from the same season and week. That gives 1,575 two-leg tickets. Each was priced by multiplying the two decimal odds, which is how a sportsbook prices a standard parlay. We then resampled whole weeks 3,000 times to get a confidence range, because games from the same Sunday can move together.

Do Unders Win More Often When the Total Is High?

Barely, and not enough to beat the price. Here is how the under performed across total bands over the full eleven seasons.

Two things stand out. First, the totals of 50 or more did not produce a meaningfully better under than the overall market. They returned half a cent on the dollar, which is well inside normal noise. Second, the best band for the under was not the high end at all. It was the 41 to 43.5 range, the kind of number nobody calls inflated.

The worst band was the bottom. Unders on totals of 40.5 or lower lost 8.74 percent. Low totals are where bettors instinctively reach for the under, and that crowd has paid for it.

On average, games with a high total finished 0.36 points above the closing number. Games with a total of 43.5 or lower finished 0.53 points above. If books were shading high totals upward to catch over money, high-total games would finish well below the line. They did not.

What Happened When You Parlayed Two High-Total Unders?

The parlay turned a flat bet into a losing one.

At a median payout near +278, a two-leg ticket needs to hit about 26.5 percent of the time to break even. High-total pairs hit 24.63 percent. The week-level resampling put the true return somewhere between -21.07 percent and +8.95 percent, so a small positive result cannot be ruled out. But the center of the estimate is clearly negative, and it sits below every other way of pairing unders.

A parlay does not create an edge. It multiplies whatever edge each leg already has. Two legs priced near break-even become a ticket that pays the book twice. We cover that math in more depth in why most parlays lose, and the same logic sits behind our parlay picks approach.

Did Any Season Make the Strategy Look Good?

Plenty did, which is exactly why the idea survives.

Single high-total unders finished above 50 percent in 8 of the 11 seasons. The 2023 season was the standout at 24-12, a 66.7 percent win rate and +27.7 percent return on just 36 games. A bettor who started then would swear the rule works.

The bad years were just as loud. The under went 31-43 on high totals in 2016, a 41.9 percent rate. In 2020, the season played mostly in empty stadiums, sportsbooks posted 130 high totals, and the under won only 45.4 percent of them. The 2025 season was also poor at 28-35.

The parlays swung harder. Two-leg pairs were profitable in 6 of 11 seasons, but the losing years were deep: -34.5 percent in 2016, -23.9 percent across 456 tickets in 2020 and -25.5 percent in 2025. One good season of under parlays looks like proof. Eleven seasons look like a tax.

Is There Any High-Total Under Spot Worth Watching?

One cut stood out, and we are flagging it as noise until it proves otherwise.

When the closing total was 48 or more and the point spread was 7 or larger, the under went 123-84. That is a 59.42 percent win rate and a +15.41 percent return over 207 games, with a resampled range of 52.66 to 66.18 percent. The story is easy to tell: a big favorite gets ahead and runs the clock, while the underdog's offense stalls.

Before treating it as a system, look at how it was found. We tested seven different slices of the high-total sample, including early versus late season, small versus medium versus large spreads, and divisional versus non-divisional games. Running seven cuts and keeping the best one is the textbook way to manufacture a trend. The middle-spread slice, 3.5 to 6.5 points, went the other way at 45.68 percent and -11.51 percent, and that result has no better story behind it.

The big-spread under also cooled recently. It went 3-6 in 2023 and 9-10 in 2025. It is worth tracking, not worth staking.

Here is how the other cuts landed:

What Has 2026 Shown So Far?

The 2026 season has been rough on anyone fading big totals. Through Monday night of Week 6, unders on closing totals of 48 or more are 7-15. Across all totals, unders and overs are nearly even at 46-47, so the damage has been concentrated at the top of the board.

Week 6 was the exception. Arizona at the Rams closed at 51.5 and finished at 34. Dallas at Green Bay closed at 53.5 and finished at 36. Washington at San Francisco closed at 51.5 and landed at 40.

Week 7 carries five totals of 48.5 or higher in the current file: New England at Chicago on Thursday at 50.5, San Francisco at Atlanta at 50.5, Green Bay at Detroit at 53.5, and both Cincinnati at Baltimore and the Rams at Las Vegas at 48.5. None of those games has a spread of 7 or more, so none falls into the one slice that tested well. That is a description of the slate, not a pick on any of them.

How Should You Bet NFL Totals Instead?

The data points to a few practical rules.

Don't bet the number because it looks big. Totals of 50 or more have been priced about as accurately as anything else. "Too high" is a feeling, not a measurement.

Price-shop every total. At a median of -107 on these unders, the difference between -105 and -115 is a large share of any edge you might have. Two books posting 47.5 and 48.5 on the same game is a full point of value.

Keep totals parlays small, or skip them. If a single leg is break-even, a two-leg version costs you roughly 6 to 7 percent. If you want a totals ticket, our guide to all-overs parlay correlation math explains why same-direction legs across different games usually behave like independent coin flips.

Use the live number when the game tells you something. A high pregame total combined with two offenses that look stuck is a real piece of information. A high pregame total on its own is not.

Our football betting picks and NFL picks follow the same discipline. If a rule cannot survive eleven seasons of closing prices, it does not get sent to members.

What Are the Limits of This Study?

We used closing totals and closing prices. Bettors who got an earlier number, either higher or lower, had different results. Opening-line data was not part of this set.

We graded standard two-leg parlays only. Sportsbooks sometimes run boosts or insurance on totals parlays, which change the math on individual tickets.

The big-spread result is the product of slicing the data seven ways and should be read that way. And every number here describes what happened, not what is promised to repeat.

Get Tonight's Live Picks

Want tonight's live in-game picks, including totals priced on the game that is actually being played, delivered to your phone during the game?

The Best Bet on Sports is the only live betting handicapping service limited on all six major U.S. sportsbooks (FanDuel, DraftKings, Caesars, BetMGM, Fanatics, ESPN BET) for winning too much during in-game action. Verified profit: $367,520+. Picks delivered via Email and Discord during games.

→ Get tonight's live picks: $199 first month — 1-Unit package, full live betting access → Try a live pick first — reserve your spot for tonight's pick

Compare every package on buy sports picks, or see what a sports handicapping service should actually deliver before you pay for one.

Frequently Asked Questions

Do unders hit more often on high NFL totals?

Only slightly. From 2015 through 2025, unders on closing totals of 48 or more won 51.44 percent of 764 decided regular-season games and returned -0.13 percent per dollar at closing prices. That is almost exactly break-even after the vig. Totals of 50 or more returned +0.53 percent, which is within normal noise.

Are NFL under parlays profitable?

Not on high totals. Every possible two-leg pairing of same-week unders on totals of 48 or more, 1,575 tickets, hit 24.63 percent of the time and returned -6.85 percent per dollar. Pairing unders from any total band did better at -1.43 percent, but no version of the two-leg under parlay made money over the eleven seasons.

What hit rate does a two-leg under parlay need to break even?

At a median payout near +278, a two-leg under parlay needs to hit about 26.5 percent of the time. Standard -110 legs produce a +264 parlay that needs roughly 27.4 percent. High-total under pairs hit 24.63 percent from 2015 through 2025, which is short of either break-even point.

Which NFL totals have been best for the under?

The 41 to 43.5 band, not the highest totals. Unders in that range won 53.15 percent of 698 decided games and returned +3.05 percent from 2015 through 2025. The worst band was totals of 40.5 or lower, where unders lost 8.74 percent. Low totals attract under bettors, and that has cost them.

Do sportsbooks inflate high NFL totals?

The results do not support it. Games with a closing total of 48 or more finished an average of 0.36 points above the number from 2015 through 2025, while games with a total of 43.5 or lower finished 0.53 points above. If books were inflating high totals to catch over money, those games would finish well under the line.

Is the under on a high total with a big spread a good bet?

It tested well but should be treated as unproven. With a total of 48 or more and a spread of 7 or larger, the under went 123-84, a 59.42 percent rate, from 2015 through 2025. It was the best of seven cuts we tested, it went 3-6 in 2023 and 9-10 in 2025, and choosing the best slice after the fact tends to overstate an edge.

How have high-total unders done in 2026?

Poorly. Through Monday night of Week 6, unders on closing totals of 48 or more are 7-15, while unders across all totals are 46-47. Week 6 broke the pattern with three high totals going well under, including Dallas at Green Bay, which closed at 53.5 and finished at 36.