FAQ
Tier records program status — Elite through Rebuilding, based on the preseason projection. Grade records the season narrative — A through F on Δ (delta), realized minus projection. A team can be Elite tier and receive a D. LSU 2025 is that case: Elite projection (78.7, #11), D grade (Δ −14.0).
They measure different things, so "better" depends on the question. Think of tier as a stock and grade as a flow. Tier is the standing balance the program brings into the season; grade is what the season added or subtracted from it. In a single year, a Contender A can absolutely outfinish an Elite C in realized SP+ — Indiana 2025 did exactly that (Solid tier proj 67.9, finished 96.3, grade A). Ohio State the same year sat at Elite proj 87.3, finished 93.0, raw grade C — nearly identical finish, very different label. If you want the pure "who was best in 2025" answer, use the realized-SP+ sort on the grade board. If you want the "who’s the stronger program" answer, use the tier board. Neither is trying to be the other.
Alabama's 2026 preseason projection is 78.3 — just under the Elite line, which bands off at 80.0. Alabama is #14 on the tier board. Tier is fixed for the season at whatever the August projection bought; in-season, a strong 2026 will show up as a positive Δ and a better grade, not as a tier promotion.
That's the live 2026 grade — each team's current in-season SP+ (Bill Connelly, via CFBD) measured against its frozen August projection, banded on the same A–F Δ scale as the 2025 grades. It recomputes weekly as SP+ updates; the projection and tier never move in-season. Alabama right now: projected 78.3, playing like 78.6, Δ +0.3, grade C. Note one difference from 2025: the live grade tests the full published composite projection, not raw preseason SP+ — documented on the methodology page.
Because Connelly's in-season SP+ opens the season anchored to its preseason priors and only gradually hands weight to actual results. In September most teams' current SP+ sits within a few points of where August had them, so most Δs are small and most live grades open at C — right now the distribution is A=0, B=16, C=92, D=30, F=0 across 138 teams. The spread widens every week as results accumulate; by December the live grade converges on the final retrospective grade. A September C is not a verdict — it's a season that hasn't diverged from its projection yet.
The Grade board sorts by Δ (realized − projection). Indiana projected 67.9 for 2025 and finished at 96.3 — a Δ of +28.4, one of the largest overshoots in 2025 (North Texas led the field at +31.0). Alabama projected 86.4, finished at 71.1 — Δ = −15.3, grade D. They came in high and undershot. Overshoot is what the Grade board measures — not raw quality.
This is exactly what the Ceiling / Floor grade is for. Ohio State 2025 projected 87.3, realized 93.0, Δ = +5.7 — raw Δ grade C. But their headroom was only 12.7 points; the +5.7 used 45% of it, which earns a Ceiling B. That's the two-lens view: raw Δ doesn't reward ceiling teams, the Ceiling grade does.
Same reason Ohio State's Ceiling grade is a letter above their raw — the second lens scales Δ against the room a team had to move. Georgia State 2025 projected 29.7, realized 15.0, Δ = −14.7 — raw Δ grade D (Δ sits in the −18 to −6 D band). But their downside room was only 29.7 points to begin with; the −14.7 used 50% of it, which lands in the Floor F band. A projection-79 team (LSU) losing 14 points is inside a Floor D; a projection-30 team losing 14 has almost no room left — that's the Floor F. Same nominal Δ, different downside room, different label.
Preseason SP+ (Bill Connelly) as the base, plus four residual layers: returning production (ESPN), 2-year recruiting composite (247/On3), a symmetric ±1 tiered coach signal, and a standardized portal-net-flow residual. Full formula and weights on the Methodology page.
The 2025 projection is Bill Connelly's August 13, 2025 preseason SP+ rating — the last preseason release before Week 1 — mapped to the 0–100 scale. The realized score is each team's final 2025 SP+ rating on the same scale. Δ = realized − projection. No smoothing, no regression, no retrofit.
The initial 2026 tier board publishes in August. Once the season starts, the model refreshes weekly — the pull runs Sunday morning and again Monday, since Connelly's in-season SP+ update typically lands Sunday night into Monday. Realized SP+ updates, and every live Δ and grade recomputes against the new week's data. Tier is preseason-fixed and does not move in-season.
CollegeFootballData for SP+, recruiting, returning production, coaches. Preseason 2026 SP+ cross-checked with Bill Connelly's public preseason release.
No — 138 FBS programs only (136 with a 2025 grade; North Dakota State and Sacramento State join FBS in 2026 and don't have a 2025 grade).
No. There are no spreads, totals, or moneylines here. Don't use these numbers to bet on games. They measure trajectory vs. projection — not game outcomes.
No. Every board and every team report card is free and open. No sign-in, no paywall, no email required.