Understanding How Projections Work: How to Use Fantasy Football Projections Wisely
Why understanding projections matters: fantasy football projections — the estimates of how many points a player will score — are everywhere, driving rankings, start/sit tools, and trade calculators. But many managers either ignore them or treat them as gospel, and both are mistakes. Projections are built from real inputs (opportunity, efficiency, matchup) and are genuinely useful, but they are estimates with significant uncertainty, since football is unpredictable. Understanding how projections are made, why they carry uncertainty, and how to use them wisely turns them from a black box into a useful tool. The goal is to lean on projections for what they do well (aggregating information into an expected value) while respecting their limits (they cannot predict the unpredictable), so you use them to inform decisions rather than blindly follow them.
Understanding How Projections Work
How projections are built:
The inputs: (1) opportunity is the foundation — projections start from a player’s expected volume (carries, targets, touches), since opportunity is the most predictive input for production (the opportunity share and usage basis); (2) efficiency estimates convert opportunity to points — projected yards per touch, catch rate, and touchdown rate turn expected volume into expected points, though efficiency is noisier than volume; (3) matchup and game environment adjust the estimate — the opponent, game script, and implied team total shift a projection up or down (the game script and implied totals inputs); (4) the projection is the combination, aggregating these inputs into a single expected-points estimate, which is its value — it synthesizes a lot of information into one number.
Why projections are uncertain:
The limits: (1) football is unpredictable — touchdowns, big plays, and game flow vary wildly week to week, so even a well-built projection has a wide range of possible outcomes around it; (2) efficiency regresses and fluctuates — the efficiency inputs are noisy, so projections built on them carry that noise (the touchdown dependency and regression issues); (3) a projection is an average, not a guarantee — it estimates the expected value, but actual results scatter around it, so a player projected for 12 points might score 4 or 24; (4) the uncertainty is largest week to week, since single-game projections are noisier than season-long ones, which is why projections are more reliable in aggregate than for any one game.
The common mistakes and how to use projections wisely:
Using them well: (1) do not treat projections as gospel — blindly following a single projection ignores its uncertainty, so use projections as one input among several, not the final word; (2) do not ignore them either — projections aggregate real information into an expected value, so dismissing them entirely throws away a useful tool; (3) use consensus and ranges over single numbers — an average of multiple projections is more reliable than any one, and thinking in ranges (floor to ceiling) respects the uncertainty better than a point estimate (the floor vs ceiling lens); (4) use projections to inform, not replace, judgment — lean on them for close calls and to synthesize information, while applying your own read on opportunity, matchup, and risk, so projections are a tool that sharpens your decisions rather than a crutch that makes them for you.
| Projection input | What it estimates | Reliability |
|---|---|---|
| Opportunity | Carries, targets, routes, snaps | Stronger near-term signal |
| Efficiency | Yards, catch rate, TD rate | Noisier week to week |
| Game environment | Pace, total, script | Context-dependent |
| Matchup | Opponent strength | Best as tiebreaker |
| Range of outcomes | Floor and ceiling | Better than one number |
Worked Example: Same Projection, Different Risk
Two players project for the same point total. One has a stable role and a narrow range. The other depends on a deep touchdown. The projection number is equal, but the decision may not be. In a safe lineup spot, the stable player may fit better; in a tournament or underdog matchup, the ceiling player may be more useful.
Projections are starting points. Roster context decides how to use them.
Common mistake: treating a projection as a promise. A projection is an average estimate, not a statement of what will happen.
Using Projections Well
Use projections to identify close calls, then check the inputs. If a projection changed because opportunity changed, take it seriously. If it changed only because of a fragile touchdown assumption, be more cautious.
The best decisions combine projection, role, health, and lineup need.
Projection Inputs
Projection systems usually start with volume assumptions: snaps, routes, carries, targets, pass attempts, and team scoring environment. Then they apply efficiency assumptions, touchdown expectations, and matchup adjustments. When a projection moves, the useful question is which input changed.
If a projection rises because a player gained a starting role, that is more meaningful than a small move caused by a fragile touchdown assumption. If a projection falls because injury risk increased, the lineup decision may change even when the median number still looks acceptable.
Worked Example: Same Projection, Different Risk
Two flex players project similarly. One has a stable role and modest ceiling. The other depends on a questionable teammate sitting out. The projection number may be the same, but the uncertainty is not.
Use projections as a starting point, then check the role and risk underneath.
Common mistake: treating projections as truth instead of estimates. The number is only as good as the assumptions behind it.
For how projections fit a broader approach, see our advanced analytics guide and fantasy football tips hub. Start at the learn hub for the full fundamentals library.