Season-Long Props Without the Pick: What Actually Drives an NFL Projection

Season-long player props look deceptively simple. A sportsbook posts a number for receiving yards, rushing yards, touchdowns, receptions, or passing yards, and the entire question appears to be whether that number feels too high or too low.

That is usually the wrong place to start.

A season-long line is not really one prediction. It is the end result of several smaller assumptions stacked on top of one another. The market is implicitly making a call on how many games a player will be available, how much volume his offense will create, how large his individual role will be, and how efficiently he will turn those opportunities into production.

That is where the useful research lives.

Instead of asking whether a player will finish above or below a particular number, it is more informative to ask what needs to happen for that number to make sense in the first place.

A projection is really a chain of assumptions

Take a wide receiver projected for 1,200 yards. That total may look reasonable compared with his previous season, but the raw number tells us very little about how the projection was built.

One model may expect him to play 17 games, dominate routes and targets, but produce at fairly ordinary efficiency. Another could reach the same 1,200-yard projection by assuming fewer targets but much stronger production per opportunity.

Those are not the same forecast.

A useful receiving projection has to account for expected games played, team passing volume, route participation, target share, depth of target, catch efficiency and production after the catch. Running-back projections depend on a similar chain involving games played, team rushing volume, backfield share and rushing efficiency. Quarterback projections depend on attempts, efficiency and, depending on the stat being studied, rushing production as well.

The final number matters, but understanding which variables are doing the most work matters more.

Availability can change the entire projection without changing the player

One of the easiest ways to distort a season-long projection is to treat availability like a yes-or-no question.

Imagine a player projected for 1,200 receiving yards. Over 17 games, he needs to average about 70.6 yards per game. If the same projection assumes only 15 games, he suddenly needs 80 yards per game.

The player did not become more talented. The projection simply placed more pressure on his per-game production.

That is why injury research should focus on what an injury changes rather than merely whether one exists. An issue may reduce expected games played, limit workload while the player is active, affect practice participation or increase the probability that his role needs to be managed.

Official NFL injury designations are useful for understanding weekly availability, but they are not season-long forecasts. A player being listed as limited in practice or questionable for one game does not automatically justify a large change to a full-season projection. The important step is translating health information into a realistic change in opportunity.

Opportunity usually deserves more attention than last year's fantasy-friendly total

Previous production is useful because it tells us what happened. Opportunity data helps explain why it happened.

The 2025 receiving season offers a good example.

Ja'Marr Chase finished with 185 targets and 1,412 receiving yards. Jaxon Smith-Njigba saw fewer targets, 163, yet produced 1,793 receiving yards. That works out to roughly 7.6 receiving yards per target for Chase compared with about 11.0 for Smith-Njigba.

The lesson is not that one player was automatically better. It is that target volume by itself could not explain the difference in yardage.

Targets vary in value.

A receiver consistently working deeper downfield has a different yardage profile from a receiver living closer to the line of scrimmage, even if their target totals are similar. NFL Next Gen Stats reflects this distinction through metrics such as Average Targeted Air Yards and Team Air Yards Share, which help show both how deep a player is being targeted and how much of his offense's downfield opportunity belongs to him.

That is why projecting 130 targets is only half the job. You also need some idea of what those targets are expected to look like.

Team volume and player share should never be treated as the same thing

A player's role can stay almost identical while his total opportunities change significantly.

Suppose a receiver maintains a 25 percent target share. If his offense throws 600 passes, that role produces roughly 150 targets. If the offense drops to 520 attempts, the same target share produces only around 130.

Nothing changed about his standing within the offense. The size of the opportunity pool changed.

This distinction matters across positions. For receivers, researchers should separate projected team passing volume from the player's expected share of routes and targets. For running backs, the same logic applies to total rushing attempts and backfield opportunity share. Quarterback projections depend heavily on whether the offense itself is expected to generate enough dropbacks and plays.

This is also the right way to evaluate offseason changes. A new offensive coordinator matters only if there is a credible reason to think the change affects play volume, pass rate, pace or role distribution. A coaching change is not automatically projection-changing simply because it happened.

The useful question is what part of the offense is actually expected to change.

Volume and efficiency can produce the same total through very different paths

Running backs make this especially clear.

James Cook led the NFL in 2025 with 1,621 rushing yards on 309 attempts. De'Von Achane finished with 1,350 rushing yards on only 238 carries. Cook averaged roughly 5.25 yards per carry, while Achane was around 5.67.

Both produced elite rushing totals, but they got there differently. Cook's season was supported by much heavier volume. Achane generated more yardage per carry on fewer attempts.

That distinction matters when building a forward projection.

A forecast leaning heavily on stable workload is different from one that requires a player to maintain unusually high efficiency. Efficiency can move with blocking, defensive attention, explosive-play rate, game environment and variance. Volume can change when backfield competition or team tendencies change.

Neither input is inherently superior. What matters is knowing which one the projection depends on.

If a projection assumes a player will both gain workload and maintain exceptional efficiency, it is relying on two favorable developments at once. That does not make the forecast wrong, but it does make the assumptions easier to identify and challenge.

Passing volume does not automatically equal passing production

Quarterback projections run into the same issue.

Bo Nix attempted 612 passes during the 2025 regular season and finished with 3,931 passing yards, averaging 6.4 yards per attempt. Matthew Stafford attempted slightly fewer passes, 597, but produced 4,707 yards at 7.9 yards per attempt.

That is a difference of nearly 800 passing yards despite Stafford having fewer attempts.

The example shows why passing-volume projections and passing-yard projections should not be treated interchangeably. Two quarterbacks can be projected for almost the same number of attempts while having very different yardage expectations because the model is making different assumptions about efficiency.

Quarterback efficiency is influenced by more than the quarterback. Receiver quality, offensive structure, protection, depth of target and yards after the catch all contribute to what each attempt becomes.

A projection of 600 attempts at 7.8 yards per attempt tells a completely different story from 600 attempts at 6.5.

If the attempt projection looks reasonable but the yardage projection seems aggressive, efficiency may be the assumption worth investigating.

Role changes matter when they change the opportunity math

Offseason player-prop analysis can get swallowed by depth-chart narratives.

A receiver leaves in free agency, so another player is expected to “absorb his targets.” A running back departs, so the next player is assumed to inherit the entire workload. A coach talks about expanding someone's role, and the quote immediately becomes part of the projection.

Real usage rarely transfers that neatly.

If a receiver leaves behind 100 targets, those 100 targets do not automatically belong to the next player on the depth chart. The offense may throw less. A tight end may see more work. A rookie may take part of the role. Another receiver may run more routes. The distribution itself can change.

Running-back opportunity is even more sensitive to the type of work being vacated. Twenty early-down carries are not interchangeable with twenty targets or twenty goal-line opportunities.

That is why “more work” is not specific enough for projection research.

The useful question is which opportunities are actually becoming available and who is realistically positioned to receive them.

Touchdowns need to be studied separately from yards

Touchdown projections deserve their own treatment because scoring opportunities behave differently from yardage volume.

In 2025, Jonathan Taylor rushed for 1,585 yards and 18 touchdowns. James Cook actually produced more rushing yards, finishing at 1,621, but scored 12 rushing touchdowns. Bijan Robinson ran for 1,478 yards and seven touchdowns.

The yardage leaderboard and the touchdown leaderboard were not the same.

That is normal.

Touchdowns depend heavily on where opportunities happen. Goal-line carries, red-zone usage, team scoring environment and quarterback rushing involvement can all affect touchdown production without materially changing a player's total rushing volume.

A back can therefore have a very stable yardage projection while carrying far more uncertainty in his touchdown projection.

That distinction is important because projecting touchdowns by simply carrying forward last year's scoring rate can create a false sense of precision.

Historical production should be the baseline, not the answer

The cleanest way to begin a season-long projection is usually with what the player has already shown.

The mistake comes when last year's conditions are assumed to return unchanged.

Historical usage needs context. If the quarterback changed, target distribution may change with him. If the offensive coordinator changed, team passing or rushing volume may move. If the backfield added meaningful competition, last year's workload might no longer be realistic. If a player is returning from injury, previous per-game numbers may not represent the same availability assumption.

None of that means every offseason change deserves a projection adjustment.

Most do not.

The goal is to identify the changes that alter one of the important inputs rather than constructing a narrative around every transaction or quote.

If the underlying environment remains mostly intact, historical usage deserves substantial weight. If several important assumptions have changed at once, simply extending last year's production across another 17 games becomes much harder to defend.

The market number should be a reference point, not the starting point

The biggest advantage of this research process is that it prevents the line itself from controlling the analysis.

Seeing a number first creates an anchor. Once a researcher knows the market has placed a receiver at 1,050 yards, it becomes surprisingly easy to build a story explaining why 1,050 feels reasonable.

A cleaner process works in the opposite direction.

Start by forming an independent expectation for availability, offensive volume, player role and efficiency. Consider meaningful injury or depth-chart changes and identify where uncertainty is highest. That should produce a reasonable range rather than a perfectly precise number.

Only then does the market line become useful.

Sometimes the independent work will land very close to the number already available. That is not failed research. It may simply mean the market is already reflecting the most obvious assumptions.

The purpose is not to disagree with every line. The purpose is to understand what the line is asking you to believe.

Find the assumption carrying the projection

A good season-long projection should be explainable.

Maybe a receiver's forecast depends on him commanding a larger target share. Maybe a running back's number assumes he keeps a much larger portion of the backfield than he had last season. Maybe a quarterback projection needs a significant efficiency jump even though expected passing volume barely changes.

Those are much more useful observations than saying a number simply looks aggressive or conservative.

Season-long NFL production is too uncertain to reduce to one confident prediction months in advance. Injuries happen, roles change, offenses evolve and efficiency moves.

Research becomes more useful when it acknowledges that uncertainty instead of pretending to eliminate it.

So when looking at a season-long player line, do not begin by asking whether the final number is right.

Ask the better question:

Which assumption is doing the most work, and what happens to the projection if that assumption is wrong?