Home / Learn / What Is an Interceptions Prop (QB)?

What Is an Interceptions Prop (QB)?

An interceptions prop is a bet on how many passes a quarterback throws to the other team in one game, posted as an over/under, almost always at 0.5. League wide in 2025, teams averaged 0.7 interceptions per game on a 2.2 percent interception rate, which puts the standard line close to a coin flip.

How the market is posted

The bet is a player total on a very low count stat. A book posts a number of interceptions, you take the over or the under, and the prices carry the margin. The default line is 0.5, so the over is really a yes or no question: does this quarterback throw at least one pick? You will occasionally see 1.5 posted for a turnover prone passer in a bad spot, which needs two or more, but 0.5 covers the large majority of the market.

Because the line sits at a half there is no push, and because the whole bet usually hinges on a single event, this is the closest thing the quarterback menu has to a coin flip. That has a specific consequence for pricing, covered below, and it is the thing most bettors skip past.

What a normal number actually looks like

Pro Football Reference's 2025 league totals show 380 interceptions on 17,439 pass attempts, an interception rate of 2.2 percent. Their per team game table puts that at 0.7 interceptions per game on 32.1 attempts.

The rate is remarkably stable season to season, which is unusual for a football stat:

2025: 0.7 interceptions per team game, 2.2 percent
2024: 0.7 per team game, 2.2 percent
2023: 0.8 per team game, 2.3 percent
2022: 0.8 per team game, 2.3 percent
2021: 0.8 per team game, 2.4 percent

Interceptions have drifted slightly down alongside falling pass volume, but the underlying rate has barely moved in five years. Unlike the completions market, where the league baseline has shifted enough to make old priors wrong, your instinct about how often quarterbacks throw picks is probably still about right.

The 0.5 line is genuinely close to a true coin flip

Take the 2025 league rate of roughly 0.70 interceptions per team game and treat picks as arriving at random around that mean. A Poisson approximation gives the shape of a typical quarterback's game:

Zero interceptions: about 50 percent of games
Exactly one: about 35 percent
Two or more: about 16 percent

So the over on 0.5 sits near 50 percent for a league average quarterback, and fair odds on either side would be close to even money. That is the single most useful fact about this market. When the true probability is that close to a flip, the book's margin is a much larger share of your expected return than it is in a market where you can be 8 or 10 points off the line. You need a real, defensible reason to think a specific quarterback is well away from 50 percent, not a hunch, before the price is even worth looking at.

One caveat on the model: interceptions clump more than a Poisson assumption allows. Multi pick games tend to come from the same causes, a pressure problem, a deficit that forces throws, weather, so real distributions carry slightly more weight in the zero and the two plus buckets than the arithmetic above suggests.

Rate separates quarterbacks far more than volume does

This is where an interceptions prop is the mirror image of a completions prop. In the completions market, attempts do the separating and accuracy barely matters. Here it is the reverse.

Among full season starters in 2025, attempts ranged from about 490 to a league leading 612 (Bo Nix), roughly a 1.25 to 1 spread. Interception rate ranged from 1.1 percent for Michael Penix Jr. (3 picks on 276 attempts) to 4.9 percent for J.J. McCarthy (12 on 243), a spread of more than four to one. Apply those rates to a league average 32 attempts and the per game expectation runs from about 0.35 to about 1.2 interceptions. The volume column cannot produce anything like that separation.

Raw interception totals mislead badly for the same reason. Trevor Lawrence and J.J. McCarthy both finished 2025 with 12 interceptions, but Lawrence threw 560 passes for a 2.1 percent rate and McCarthy threw 243 for 4.9 percent. Those are not similar quarterbacks, and a leaderboard sorted by total says they are.

The top of the 2025 interception board: Geno Smith 17 on 448 attempts (3.8 percent), Tua Tagovailoa 15 on 384 (3.9 percent), Sam Darnold 14 on 477 (2.9 percent), Justin Herbert 13 on 512 (2.5 percent), then Lawrence and McCarthy at 12. At the safe end, Penix at 1.1 percent, Caleb Williams at 1.2 (7 on 568), and Matthew Stafford, Cam Ward and Jalen Hurts all at 1.3.

What actually moves an interception projection

Start from the quarterback's rate, then apply attempts. Rate first, volume second. Doing it the other way around makes every high attempt passer look dangerous, and 2025 says several of them are among the safest.
Projected game script. A team likely to trail throws more and throws later, into coverage that knows it. Big underdogs are structurally the better side of an interceptions over, for the same reason they are a better completions over.
Pass rush and pressure rate. Pressure is the most reliable path to a bad throw. A quarterback facing a top pass rush behind a struggling line is a different projection than the same quarterback in a clean pocket.
Wind, specifically. Wind moves the ball off target on exactly the intermediate throws that get picked. It also reduces attempts, so the two effects partly cancel, but the accuracy hit is the larger one.
Average depth of target. Downfield offenses give up more interceptions per attempt than quick game offenses. A checkdown heavy plan lowers the risk even when attempts are high.
Opponent takeaway profile, weighed lightly. Defensive interception totals are driven mostly by how many passes they see and by tipped ball luck. It is real information, just weak information.
Form the number before you look at the price. With a true probability near 50 percent, anchoring on the posted line is even more damaging here than usual.

Where the market misleads

Interception rate is one of the least stable quarterback stats. Whether a pass gets picked or dropped by a defender is partly out of the passer's hands, so single season rates carry a lot of noise and regress hard. A quarterback coming off a four pick month is usually not that quarterback going forward.

Recency reads badly in both directions. Two clean games in a row gets priced as safety, and the base rate still says roughly one game in two produces a pick. The market overreacts to a small sample in a stat that is mostly small sample.

Volume is not risk. The league's attempts leader in 2025 threw 11 interceptions at a 1.8 percent rate, below the league mark. More throws is not the same as more danger per throw.

The juice is the whole game. Because the true number is so close to even, a heavily juiced side can be a losing bet even when you are right about the direction. This is one of the few markets where you can beat the projection and still not beat the price.

Interceptions do not move with the other quarterback markets. A passer can throw for 320 yards, three touchdowns and two picks in the same afternoon. If you are building around the quarterback, the passing yards prop and the passing touchdowns prop are separate questions with separate drivers.

How BetLogic helps

Our NFL QB passing sheet carries season interceptions next to pass attempts per game and the Vegas derived team implied total, which are the three inputs an interception projection is built from: interceptions over attempts gives you the quarterback's actual rate rather than a misleading raw total, attempts per game gives you the exposure to apply it to, and the implied total tells you whether the game script is likely to force more throws than usual. The sheet does not post a projected interception number, because at 0.7 per game the honest answer is usually "close to a coin flip," and we would rather show you the rate and the volume than dress up a guess. For the wider map of these markets, start with the NFL player props overview.

Frequently asked questions

What is the usual line for an interceptions prop?

Almost always over/under 0.5, meaning the over needs at least one interception. Books occasionally post 1.5 for a turnover prone quarterback in a difficult spot, which requires two or more. Because the line sits at a half point, there is no push.

How many interceptions does an average NFL quarterback throw per game?

About 0.7. Pro Football Reference's per team game averages show 0.7 interceptions on 32.1 attempts in 2025, a 2.2 percent interception rate, from 380 interceptions on 17,439 attempts league wide. The rate has sat between 2.2 and 2.4 percent every season since 2021.

Is the over or the under better on an interceptions prop?

Neither by default, which is the point. Treating interceptions as roughly Poisson around the 2025 league mean of 0.70 per game puts at least one pick at about 50 percent, so fair odds are near even money on both sides. That makes the posted price, not the direction, the thing that decides whether the bet is worth making.

Do pass attempts or interception rate matter more for this prop?

Rate, by a wide margin, which is the opposite of most quarterback markets. In 2025 full season starters ranged from roughly 490 to 612 attempts, about a 1.25 to 1 spread, while interception rate ranged from 1.1 percent (Michael Penix Jr.) to 4.9 percent (J.J. McCarthy), more than four to one.

Are interception totals a good way to compare quarterbacks?

Not on their own. Trevor Lawrence and J.J. McCarthy both threw 12 interceptions in 2025, but Lawrence did it on 560 attempts for a 2.1 percent rate and McCarthy on 243 for 4.9 percent. Divide interceptions by attempts before comparing anyone.

Is there an interceptions prop for defensive players?

Some books post a defensive interception market, usually a yes or no on whether a specific defensive back records a pick. It is a much thinner and more volatile market, since defensive interceptions depend heavily on how often the opposing quarterback throws in that player's direction and on tipped ball luck. The quarterback side is the one that is widely posted and widely priced.

BetLogic provides data-driven insights only and does not constitute financial or gambling advice. Please bet responsibly and only what you can afford to lose. You must be of legal betting age in your jurisdiction.