Seasonal Projections vs. Weekly Projections: When to Use Each

Seasonal and weekly projections answer different questions — and mixing them up is one of the most common ways fantasy managers leave points on the board. Seasonal projections map a player's expected output across a full schedule, making them the backbone of draft decisions and long-horizon trades. Weekly projections zoom in on a single game slate, reacting to injury reports, weather, and matchup data that simply didn't exist three months earlier. Knowing which lens to reach for, and when, is a core skill that separates disciplined projection use from noise-chasing.

Definition and scope

A seasonal projection estimates a player's cumulative statistical output over an entire season — typically modeled as a single number per scoring category (rushing yards, touchdowns, receptions) or an aggregate fantasy-point total. These projections draw heavily on historical baselines, role expectations, team context, and positional aging curves. A running back projected for 1,200 rushing yards and 8 touchdowns in a full NFL regular season is operating on a seasonal model.

A weekly projection narrows the window to a single game or, in the case of MLB, a smaller multi-game window. The inputs shift substantially: opponent defensive rankings for the specific week, injury designations filed that Wednesday, Vegas implied team totals, and recent usage trends measured over the last 3–4 games. A quarterback going up against a secondary ranked 30th in pass yards allowed gets a weekly adjustment that his seasonal model almost certainly didn't anticipate.

The scope difference matters because both types appear on the same dashboards and are often presented with similar formatting. Treating them as interchangeable is a category error — the way reading a weather forecast for the full year as a substitute for today's radar report would be. For a broader look at how projection systems categorize their outputs, the projection models explained reference breaks down the architecture behind both types.

How it works

Seasonal projections are typically built once before a season opens and then revised at major inflection points — training camp depth charts, preseason injuries, significant trades. The modeling framework leans on large multi-year sample sizes. A wide receiver's seasonal target share projection, for example, might integrate 3 years of route participation data, team offensive coordinator tendencies, and opponent strength-of-schedule estimates averaged across all 17 games. The statistical inputs for fantasy projections page details what feeds these baselines.

Weekly projections operate on a shorter feedback loop. They start with the seasonal baseline and apply multipliers or adjustments driven by current-state variables:

1. Matchup adjustment — A tight end facing a defense that allowed the 5th-most fantasy points to the position in the prior 4 weeks gets a positive multiplier.
2. Usage-rate signal — Snap count or target share trends from the immediately preceding games. Snap count and target share data feeds directly into these short-window adjustments.
3. Vegas context — Implied team totals derived from the betting market function as an independent signal for projected scoring volume. The relationship between Vegas lines and fantasy output is detailed at Vegas lines and fantasy projections.
4. Injury context — A running back moving from "questionable" to "full participant" on Thursday changes the carry-distribution model for that week.
5. Weather — For outdoor games, precipitation and wind speed above approximately 15 mph have measurable effects on passing volume. Weather impact on fantasy projections covers the thresholds that matter.

Common scenarios

Draft day is the canonical home of seasonal projections. Every pick made in rounds 1–15 is a bet on cumulative production, not a single week's matchup. Using a weekly projection from Week 3 of the prior season as a draft signal is a form of recency bias with no structural justification.

Start/sit decisions belong almost entirely to weekly projections. A player with a solid seasonal baseline but a brutal week-specific matchup deserves a downward adjustment that the annual model won't provide on its own.

Trade evaluation sits at the intersection of both. If trading for a receiver who has one spectacular recent game, the weekly projection for that performance matters less than the seasonal rest-of-season outlook. Rest of season projections and trade value and projection data together address this dual-frame evaluation.

Waiver wire pickups lean weekly but aren't purely so. A handcuff running back worth picking up after a starter's injury needs both a weekly assessment (is the game this Sunday favorable?) and a seasonal one (how many carries is this backfield likely to produce going forward?). Using projections for waiver wire decisions walks through that layered logic.

Daily fantasy contests weight weekly — sometimes single-game — projections almost exclusively, since roster construction resets every slate. Daily fantasy sports projections covers the specific adjustments that apply in that format.

Decision boundaries

The cleaner the decision boundary, the better the projection type matches the decision type:

One structural principle holds across all of them: weekly projections are more volatile and more perishable. A seasonal projection built in August might retain 80% of its informational value in November. A weekly projection filed Tuesday is materially stale by Saturday if injury reports change. The projection update schedule describes how often each projection type is refreshed and why the timing windows are set where they are.

For those building a full projection workflow from the ground up, the main resource index at Fantasy Projection Lab maps the complete library of tools and reference pages organized by decision type.