NFL Referee Over/Under Trends, Tested on 11 Seasons

NFL referee over/under trends do not carry from one season to the next. Across 156 back-to-back referee seasons from 2015 through 2025, the correlation between a crew chief's over rate in one season and the next was -0.09, which is statistically zero. Referees whose games went over 60 percent of the time or more saw 66.22 percent overs that season and 48.71 percent overs the following season. The Best Bet on Sports has built a verified $367,520+ in profit across all sportsbooks over more than twenty years, and referee trends are one of the most common things sold as an edge that we do not use.

Every week, some betting content lists the referee assignments and attaches a trend: this crew's games go over, that crew calls a lot of penalties, take the over. It sounds like inside information, because referees are human, crews have tendencies, and the assignments are public before kickoff.

So we tested whether a referee's over/under record actually predicts anything.

How Did We Test Referee Over/Under Trends?

We used every NFL regular-season game from 2015 through 2025 in the nflverse game file, which lists the crew chief for each game along with the closing total. That is 2,894 games worked by 33 referees.

For each referee in each season with at least ten games, we counted how many of their games went over and under the closing total, ignoring pushes. A typical crew chief works about 15 or 16 regular-season games a year. That gave 156 pairs of consecutive seasons for the same referee.

The test is simple. If a referee's over rate reflects something real about how they call games, a crew that sees a lot of overs this year should see more overs than average next year. If it is mostly noise, next year's rate should drift back toward 50 percent regardless of this year's.

Do NFL Referee Over/Under Trends Carry Over?

No.

All three numbers are effectively zero. A crew chief's over rate this season told you nothing about next season. Neither did the average number of points their games scored relative to the total, and neither did how often the home team covered in their games.

The same held when we skipped seasons entirely. For the 17 referees with at least 80 games, we compared each one's over rate in odd-numbered seasons with their rate in even-numbered seasons. The correlation was -0.09. Even splitting each career in half, with several seasons on each side, produced no consistency.

What Happened to the "Over Referees"?

This is the version of the trend that actually gets sold: find the crews with the most overs and bet their games.

The pattern is the same one we found for hot ATS starts. The extreme group was extreme in the season that defined it and ordinary immediately afterward. The "over referees" went from two-thirds overs to slightly under half. The "under referees" went to almost exactly 50 percent.

A bettor who took the over in every game worked by the previous season's three most over-friendly referees would have gone 223-244. At -110 pricing that is a clear loss.

Why Do Some Referees Look Like They Have a Trend?

Because 16 games is a small sample, and 17 crew chiefs per season is a lot of chances for one of them to look extreme.

If every game were a fair coin flip, a referee working 16 games would still finish at 60 percent overs or more, or 40 percent or less, quite often. When we simulated that, about 51.6 percent of referee-seasons landed in one of those extreme ranges by chance alone. In the real data, 56.1 percent did. The spread of referee over rates was 13.21 percentage points, compared with 12.73 points expected from pure coin flips.

In plain terms, nearly all of the variation between referees in a given season is what randomness produces. There may be a sliver of real difference left over, but it is too small to find in one season and, as the year-to-year test shows, too unstable to bet on.

Are There Referees With Real Career Tendencies?

Possibly, but the evidence is weaker than it looks. Among the 15 referees who worked 100 or more games in our sample, career over rates ranged from 42.1 percent for Shawn Hochuli (51-70) to 59.0 percent for Jerome Boger (72-50). Most sat between 45 and 50 percent.

Fifteen referees with about 110 to 170 games each will naturally produce one or two who look about two standard deviations from even. That is roughly what this list shows. A career number like 59 percent is worth noting. It is not, on this evidence, a reason to bet a game, especially because careers span different crews, rule changes and scoring environments.

This season the samples are smaller still. Through the first four weeks of 2026, each crew chief had worked about four games. A 4-0 over record for a referee in October means nothing at all.

Why Doesn't Officiating Show Up in the Total?

There are three reasons, and they compound.

First, crews are not consistent. Crew chiefs lead teams of officials whose makeup changes, and individual calls depend on game situations that vary week to week.

Second, the assignments are public. Referee assignments are reported before kickoff, and anyone pricing totals can see the same lists. If a crew really did add three points to every game, it would be among the easiest edges in sports to find, and the market would absorb it quickly.

Third, the total is mostly about the teams. Offenses, defenses, pace, weather and injuries move a total by many points. Whatever a referee adds or subtracts is small next to those factors, and it disappears into the noise of a 45-point game.

This study does not include penalty counts, so it cannot say whether certain crews throw more flags. It can say that whatever they do, it has not shown up as a repeatable effect on whether games go over or under the closing total.

To be precise about that callout: this study uses closing pregame totals and final scores. It contains no in-game prices, so it cannot measure whether live totals are easier to beat. That is structural reasoning, not a finding from these 2,894 games.

What Does This Mean When You Pay for Sports Picks?

Referee trends are a useful test of any service, because they share three traits with most weak betting angles. They are easy to look up. They sound like insider knowledge. And they rest on small samples that look convincing in a table.

So when a service offering nfl expert picks leans on a referee trend, ask the same questions this study asked. Did the trend hold up the following season? Was it graded against prices you could actually bet? Is it something the market can also see? If the answer to the last question is yes, the price has usually caught up.

The same check applies in basketball. NBA referee assignments are also posted before games, and basketball referee trends are sold the same way. This study covers the NFL only, so we are not claiming a result for the NBA. With the season opening later this month, it is a reasonable question to put to anyone marketing basketball handicappers or NBA handicappers on the strength of a referee table.

A real sports handicapping service should be able to explain where its edge comes from without relying on public lists. Our pay for sports picks page lays out what you are actually buying, the football handicappers page explains our approach to the NFL, and every graded play is on our results page. For more on separating real records from noise, see how to read a sports betting track record.

Where This Analysis Stops

Four limits.

This is NFL regular season only, 2015 through 2025, graded against closing totals. Opening totals may not adjust to referee assignments as fully, which would be a timing question rather than evidence that the trend is real.

The nflverse file lists the crew chief only. It does not track the rest of the crew, and crews change from season to season.

We did not have penalty data. A referee could call more penalties without affecting the total, and this study would not detect it.

And the career numbers are small for individual referees. A few seasons of data per official is enough to show that year-to-year trends do not hold, and not enough to rule out a small, stable difference for any one person.

Get Tonight's Live Picks

Want tonight's live in-game NFL picks and totals delivered during the game, priced on how it is actually being played and officiated?

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 read what arrives each day from daily sports betting picks.

Frequently Asked Questions

Do NFL referee over/under trends predict anything?

Not from one season to the next. Across 156 consecutive-season pairs for NFL crew chiefs from 2015 through 2025, the correlation between a referee's over rate in one season and the next was -0.09, which is effectively zero.

Should you bet the over with a high-scoring referee?

The evidence says no. Referees whose games went over the closing total 60 percent of the time or more saw 66.22 percent overs that season and 48.71 percent overs the next season. Betting the over with the previous season's top three over referees each year went 223-244.

Why do some NFL referees have extreme over/under records?

Mostly chance. A crew chief works about 16 regular-season games a year, and with 17 crew chiefs, some will land far from 50 percent by luck alone. A coin-flip simulation produced extreme records almost as often as the real data did.

Which NFL referees have the highest over rates?

Among referees with at least 100 games from 2015 through 2025, Jerome Boger's games went over 59.0 percent of the time and Shawn Hochuli's went over 42.1 percent. With 15 referees in that group, a spread like that is close to what chance alone would produce.

Do NFL referees affect which team covers the spread?

Not in a repeatable way. The correlation between a referee's home-team cover rate in one season and the next was 0.01. There was no sign that certain crews consistently favored home teams against the spread.

Do sportsbooks adjust totals for referee assignments?

Referee assignments are public before kickoff, so anyone pricing a total can see them. Whether a specific book adjusts is not something this study can measure, but the closing totals showed no referee effect that persisted from season to season.

How should I evaluate a picks service that uses referee trends?

Ask whether the trend held up the following season, whether it was graded at prices you could actually bet, and whether the information is public. In this study, referee over/under trends failed the first test, and they are public by definition.