Strength of Schedule: How Much It Should Actually Move a Player on Your Board

What a Preseason Schedule Tile Measures

A preseason strength-of-schedule graphic bolts two things together: a fixed calendar and a set of opponent ratings. The calendar is settled the day the league releases the slate. The ratings attached to it are, almost always, last season's defensive results.

Defenses turn over harder than skill-position roles do, and a secondary can lose two starters and its coordinator in one offseason and grade nowhere near its December level. A new coordinator can change a unit's coverage scheme outright, leaving last year's rating attached to a defense that no longer lines up that way. The tile still prices December.

The room reaches for that tile as a tiebreaker anyway, which is where the cost lands, because a tiebreaker is supposed to be the most reliable input left once two players are otherwise even. Preseason schedule sits near the bottom of that list, below role and price. Sorting ties with a year-old rating runs the job backwards.

What the Board Does With Schedule

The season model behind the draft board runs three documented steps: production and volume, efficiency, and games and age. None of the three looks at the defenses a player will face, and the team-strength number it computes from closing lines is held at neutral, so no offense or opponent moves a projection. The methodology page lays each step out in full. Every season median on the board is a schedule-free number.

The board earned its record without a schedule input. In the three held-out seasons with market data, the season model tracked actual finish more closely than ADP every time, by as little as 0.0068 Spearman in 2024 and as much as 0.107 in 2022. Both columns sit on the methodology page, next to the table where it lost 2022 to a repeat-last-year baseline by 0.0063, a wider margin than any of its wins against that baseline.

That hands the schedule question to you. Any adjustment on draft day is one you add by hand, on top of a backtested projection built with no calendar, using the least stable rating available. Size decides whether it matters.

Sizing a Schedule Edge on an Invented Board

Take an invented stretch of tight end board, with numbers made up to show the shape and tied to no current run. Five season medians read 171, 166, 163, 158, and 137, and the adjacent gaps run 5, 3, 5, and 21. That leaves one tier break sitting under four rows spaced three to five points apart.

Price a generous schedule edge against that spacing. Call the softest slate at the position worth four percent of a season median, a figure also invented for this illustration; on the 166 row that comes to 6.6 points. Those 6.6 points clear every gap inside the tier and cover less than a third of the 21-point break below it.

Taken at face value, the edge still stops at the tier's edge. The rating under it described a different defense, so an honest estimate shrinks the 6.6 toward zero before it ever reaches the break. A player carried across 21 points on that read has been moved by a stale rating. On a flatter receiver tier, where adjacent medians sit one or two points apart, the same four percent can pass several names without touching a break, and the tiebreak rule below still takes them one pair at a time.

Where Schedule Earns Weight

Matchups belong in the weekly numbers, and the weekly numbers already carry them. Each game's spread and total set an implied team total, and the weekly projections behind the start/sit comparer scale with that environment. The market setting those totals sees each defense as it stands that week, injuries and coordinator included. The weekly model reruns daily.

An implied team total is a price with money behind it, set in the days before kickoff and moved by injury news as it lands. A preseason defensive rating is a summary of games already played, frozen in spring. Use the start/sit comparer for the job the draft board never attempts: compare two starters after the week's lines post, and let the totals carry the schedule.

The weekly numbers carry a graded record of their own. Across four held-out seasons the weekly model beat both of its naive baselines in 10 to 12 of 17 weeks per season, and even at its best it lost 5 of 17 weeks. In 2025 its average miss ran 4.54 points per player-week, against 4.69 for a season-average-to-date baseline and 5.75 for last week's score, and the weekly ledger grades the first numbers posted before each kickoff.

Putting It in Your Draft Plan

Take the medians and gaps as the board states them, find the tier breaks, and let schedule settle ties inside a tier. The tie rule is house judgment: two players in the same tier within about three percent of each other, with role and price already failing to separate them.

Put schedule last in your prep, after tiers and prices, and give it whatever minutes are left over. ADP moves daily and absorbs whatever the room believes, popular schedule graphics included, so a bump you add by hand may already sit inside the price you are about to pay. Read the season backtest on the methodology page and the weekly ledger, then lean on the board, and on this piece, only as hard as that record earns.