SF
@LAR
MIA
ARI
DEN
@SEA
WAS
@ATL
LV
@DAL
MIN
@NYG
@LAC
@KC
PHIBrock Purdy (QB, SF) — 2026 evaluation
Verdict — FADE (medium confidence)
Purdy is an elite-efficiency, low-rush pocket passer being drafted at QB12 (ADP 96.3, round 8) — the top edge of the QB7–12 "dead zone" (qb.md §9): starter cost without positional edge. Why the market is wrong: it isn't overrating his talent — QB4 PPG (21.0, 9 starts) priced at QB12 is already a discount — it's that the price band itself buys nothing in 1QB/4pt. His 2025 scoring rode a 7.0% pass-TD rate against a 5.3% two-year xTD rate, his rushing gives no floor (16.8 yds/gm, 1.9% designed-run rate, 2 inside-5 carries in 2025), and his TD-regressed xFP was 18.6–18.9 PPG in both seasons — the same tier as Stafford (100.9), Caleb Williams (106.3), and Nix (115.6) going 0.5–2 rounds later, while Hurts (93.1) and Daniels (86.6) sell true rushing floors at the same shelf. Fine player — arguably the best pure passer in this band — but wrong price: draft him only at a round-plus discount (pick ~110+), or re-verdict if the league turns out to be 6pt/superflex.
Bull case
- Best per-start passer in his price band, in a stable elite environment: EPA/dropback ≥ +0.18 and CPOE ≥ +2 in three straight seasons (+0.202 / +7.2 in 2025), sack rate down to 3.5%, same play-caller ten years running, 4/5 OL back, 10.5 win total — every start comes with a top-5-offense floor, and he just posted QB4 PPG (21.0) across 9 starts.
- Structural TD access is climbing: SF's inside-10 pass rate rose three straight years to 54.2%, pace flipped to 5th-fastest, and Purdy's 2025 passing xTD rate (6.0%) led this profile — if Evans/Kittle/Pearsall stay healthy, a 30+ TD season is live without any volume change (that's the 335 ceiling).
- The weapons got better, not worse: 2025's 7.55 YPA came with Kittle missing 6 games, Pearsall on a PCL, and Jennings/Bourne as the outside corps; 2026 replaces that with Evans + Kirk + R2 capital while CMC and a healthy Kittle return — YPA mean-reverts toward his 8.0+ career band.
Bear case
- No positional edge where the format pays it: 16.8 rush yds/gm, a 1.9% designed-run rate that is mostly QB sneaks, 2 inside-5 carries, and rushing xTD of 1.2 in 2025. In 1QB/4pt a rush yard is worth 2.5 pass yards — Purdy has to out-throw the entire QB12–18 blob on 31–32 attempts/gm behind a −0.9 to −2.2 PROE just to tread water with them.
- The price pays for a TD rate that regresses: 20 pass TD on 284 attempts (7.0%) vs a 5.3% two-year xTD rate, plus rush TDs over xTD in both seasons (5 vs 3.0; 3 vs 1.2). TD-regressed he was an 18.6–18.9 PPG player in both 2024 and 2025 — a QB10–13, not the QB4 the highlight PPG suggests.
- Games and turnovers dent the floor: 10 games missed across two seasons (turf toe re-aggravated once in 2025); IWT of 3.7–3.8% with a fair-to-favorable INT ledger means the 10-INT-in-9-starts pace was earned, not unlucky — multi-INT clunkers (3 vs CAR in Wk 12) are part of the profile, and there's no luck rebate coming.
Projection & comps
Bottom-up, league scoring (4pt pass TD, −1 INT, 0.1 rush yd, 6 rush TD, −2 fum lost). Games played varies by percentile — the games column is Purdy's main downside axis (8 games missed in 2025):
| Component | Floor (p20, 13 gm) | Median (p50, 16 gm) | Ceiling (p80, 17 gm) | Method |
|---|---|---|---|---|
| Pass attempts | 403 (31.0/gm) | 509 (31.8/gm) | 553 (32.5/gm) | Team profile: 64.5 plays × ~58% pass ≈ 37 db − sacks − scrambles ≈ 33.5 team att/gm; Purdy's own rate 30.4 (2024) / 31.9 (2025); PROE series −2.2/−2.2/−0.9 caps volume |
| Pass yards (YPA) | 3,023 (7.5) | 4,021 (7.9) | 4,479 (8.1) | Career-band YPA 8.47 (2024) / 7.55 (2025, injury-gutted corps); upgraded/healthy 2026 corps splits the difference |
| Pass TD | 19 (4.8%) | 27 (5.3%) | 32 (5.8%) | Anchored to computed passing xTD: 22.5 on 456 att 2024 (4.9%), 17.2 on 287 att 2025 (6.0%) → two-year xTD rate 5.34%; never to 2025's 7.0% actual |
| INT | 13 | 15 | 14 | From IWT: 3.7% (2024), 3.8% (2025, FTN) → INT rate ~2.9–3.0% (matches two-year actual 3.0%; no luck rebate owed) |
| Rushing | 195 yds, 1.5 TD | 264 yds, 2.5 TD | 306 yds, 3 TD | Projected separately: ~3.3 car/gm (2.2 scramble + 1.1 designed/sneak) × ~5.0 YPC; rush TD anchored to rush xTD (3.0 in 2024, 1.2 in 2025 — actuals of 5 and 3 were over) + live sneak role |
| Fumbles lost | −4 | −4 | −3 | 0 rushing fumbles lost 2024–25 (nflverse); sack-fumble allowance |
| Points | 210 | 290 | 335 | 16.2 / 18.2 / 19.9 PPG |
Median 290 ≈ QB9–11 by 2025 season totals, ≈ QB12 by PPG — i.e., exactly what pick 96 pays for. Ceiling requires 17 games plus the TD rate holding near 6%; his best full season (2023, historic efficiency) totaled ≈ 308 in 16 games under this scoring, which is why the ceiling stops at 335. Games risk: medium — turf toe cost 8 games in 2025 (re-aggravated once on return), 2 games missed in 2024; fully cleared and full OTA participant in June 2026, and rushing exposure is low, but back-to-back injury seasons cap it above "low."
Comps (this repo's computed league-scoring QB tables, nflverse 2024–25, calc'd 2026-07-07 — efficient run-lean pocket profiles, ex-sack-fumbles):
- Brock Purdy 2024 (self) — 285 (19.0/gm, QB9 PPG): the base case, almost exactly the median
- Jared Goff 2025 — ~311 (18.3/gm, QB12 PPG): efficient pocket passer on a run-lean contender
- Dak Prescott 2025 — ~320 (18.8/gm, QB8 PPG): ceiling-adjacent volume case
- Justin Herbert 2024 — ~284 (16.7/gm, QB14 PPG): the "efficiency but capped attempts" floor shape
- Trevor Lawrence 2025 — ~352 (20.7/gm, QB6 PPG): what the ceiling looks like if TDs hold
External sanity check: data/projections/ does not exist (no provider projections cached). Web consensus (Fantasy Life, CBS — June–July 2026) frames Purdy as a "low-end QB1 in all 1QB leagues," i.e., agrees with the median; the verdict difference here is band-structure, not projection disagreement.
Usage profile — opportunity core (qb.md §2)
Computed 2026-07-07 from nflverse pbp (REG only, kneels excluded from carries) + local FTN charting join; season aggregates cross-checked against data/stats/<yr>/passing.csv / rushing.csv (pulled 2026-07-07). Official 2025 line: 197/284, 2,167 yds, 20 TD, 10 INT in 9 games; pbp play-filter counts 287 attempts (2-pt edge cases) — rates below use pbp denominators.
| Metric | 2024 (15 gm) | 2025 (9 gm) | Band | Read |
|---|---|---|---|---|
| Rush att/gm (no kneels) | 4.1 | 3.2 | Low-Good → Concern-adjacent | Trending down; scramble-dependent |
| Designed rush rate | 1.9% of team plays | 1.9% | Concern (<2%) | And most designed runs are sneaks (14 of 17 in 2024; 7 of 11 in 2025, FTN) — no real keeper package |
| Scramble rate | 8.4% of db (2.9/gm, 300 yds) | 5.7% (2.0/gm, 131 yds) | Elite → Good | The fragile kind of rushing (qb.md §3): coverage/OL-dependent, fell by a third in 2025 |
| Rush yds/gm | 21.9 | 16.8 | Good → below-Good | 2023 was 9.7 — the two-sided sample says ~15–20 is the real level |
| RZ rush share | 14/99 team RZ carries (14.1%) | 7 RZ carries (≈13% est. in his 9 games) | Good-low | Real but thin |
| Inside-5 carries | 5 | 2 | Concern (≤2) | Sneak role exists (7–14 sneaks/yr) but SF throws inside the 10 now (54.2% in 2025, 3-yr climb — team profile) |
| Rushing xTD | 3.0 (actual 5) | 1.2 (actual 3) | Concern | Over xTD both years — rush-TD regression, not upside |
| Dropbacks/gm | 35.1 | 35.1 | Good | Remarkably stable |
| Pass att/gm | 30.4 | 31.9 | Good-low | Sack rate improved 5.9% → 3.5%, converting dropbacks to attempts |
| Team PROE | −2.2% (2024), −0.9% (2025) | Neutral-to-run-lean | 10.5 win total (BetMGM, 2026-07-07) adds leading-script cap; pace now 5th-fastest partially offsets (team profile) | |
| xFP (TD-regressed, computed) | 18.9 PPG | 18.6 PPG | QB1/2 fringe | Actual PPG 19.0 / 21.1 — 2025 outscored usage by ~2.5 PPG, all of it TD-rate |
No late-season designed-run install exists to override the season rates: the 11 designed carries in 2025 were scattered and sneak-heavy. Archetype (qb.md §10): nearest to *pocket volume passer* but below the 36-att/gm bar — an "efficient game-managed starter," the profile the methodology explicitly refuses to pay starter price for in 1QB/4pt regardless of efficiency.
Efficiency & talent signal (qb.md §5)
| Metric | 2023 | 2024 | 2025 | Band | Read |
|---|---|---|---|---|---|
| EPA/dropback | +0.311 | +0.184 | +0.202 | Elite (≥0.18) 3 straight yrs | The real thing — QB-owned, travels |
| CPOE (nflfastR pbp) | +5.38 | +2.06 | +7.23 | Elite | NGS model agrees: +2.98 (2024), +5.07 (2025) — accuracy is his carrying trait |
| IWT rate (FTN, INT-worthy throws) | — | 3.7% (17/456) | 3.8% (11/287) | Good/Concern boundary | Slightly elevated; INT ledger fair-to-lucky (12 INT vs 17 IWT in 2024; 10 vs 11 in 2025) → no positive INT regression owed — mild fade signal per §5 |
| Pressure-to-sack | — | UNVERIFIED (rate) | UNVERIFIED (rate) | Good-Elite signal | Sack rate 5.9% → 3.5%; FTN QB-fault sacks 12/31 (2024), 4/11 (2025); "top-5 pressure-to-sack ratio" (SF Standard, 2025, via team profile). PFR pressure table inaccessible (403, 2026-07-07) |
| Deep-ball rate (20+ air yds) | — | 10.8% | 12.3% | Good → Elite edge | Ceiling source is real when protected |
| aDOT | — | 8.58 | 7.60 | Elite band (7.5–9.5) | Healthy depth mix |
| Play-action rate (his dropbacks) | — | 19.8% | 26.3% | Good | Shanahan-owned free efficiency, stays in 2026 |
QB-owned traits (CPOE, pressure-to-sack, scramble ability) are all strong and the environment persists — believe the efficiency. What does not persist is the 7.0% TD rate and the rush-TD overshoot; both are anchored back to xTD in §2.
Context (from data/team-profiles/SF.md, updated 2026-07-07)
- Play-caller: Kyle Shanahan, 10th year, confirmed primary caller for 2026 (Kubiak retained as OC, situational calls only). Scheme family: wide-zone/PA (motion 68.5%, PA 22.3%, league-high 21 personnel). PROE series −2.2 / −2.2 / −0.9 — structurally run-lean; 2025 pace flipped to 5th-fastest, which held plays/gm at 64.7.
- Job security: absolute — 5-yr/$265M (May 2025); Shanahan chose a healthy Purdy over a 5-3 Mac Jones without hesitation. Backup is tier-A (Jones), which protects SF pass-catchers but is irrelevant to Purdy's own value.
- Pass-catchers (the 2026 swing): ~170 of 550 targets vacated (~31% — under the 40% continuity tripwire, so 2025 efficiency carryover is allowed with a trim). In: Evans (X, age-33), Kirk (slot), R2 #33 Stribling. Returning healthy: Kittle (age-33), Pearsall (PCL — "clear No. 2 WR if on the field"), CMC (23.5% target share, age-30). On paper the best healthy corps of Purdy's career; in practice three of the top five weapons carry age or injury flags, and the tree still funnels through CMC (30.5% RB+FB target share in 2025).
- O-line: 4/5 returning, PBWR 10th (ESPN, Jan 2026); LG is the one open spot; Trent Williams age-38. Good-not-elite protection with one leak.
- Game environment: win total 10.5 (BetMGM, 2026-07-07) — positive script lean; leading teams run (qb.md §7), which is exactly why the attempt projection stays ~31–32/gm despite the pace gain.
- Health (web track): fully recovered from the 2025 turf toe; full participant through OTAs, mandatory minicamp cancelled for perfect attendance (NBC Sports Bay Area / Niner Noise, June 2026).
Tripwires
- League settings confirmed as 6pt pass TD and/or superflex → re-run immediately; verdict moves toward HOLD/TARGET (Purdy gains more from those levers than almost any QB in this range).
- ADP drifts past ~110 (QB14-ish) → the round-plus discount has arrived; re-verdict toward HOLD/TARGET.
- Any camp/preseason news on the toe or throwing arm → games-risk input voids; re-run.
- Shanahan cedes primary play-calling or departs → environment-owned inputs (YPA, PA rate, PROE) void; re-derive.
- Pearsall PCL setback, Evans age-decline camp reports, or any CMC soft-tissue news → YPA/xTD inputs trim (CMC news fires every SF eval per the team profile watch list).
Sources
data/stats/2025/&data/stats/2024/—passing.csv,rushing.csv,ngs_passing.csv,pbp_summary.csv,ftn_charting.csv(nflverse via nflreadpy, pulled 2026-07-07): official lines, NGS CPOE/aDOT/TTT, team plays/pass rate- nflverse pbp 2023–2025 (REG) via nflreadpy, computed 2026-07-07 (scratchpad
purdy_pbp.py,purdy_ftn.py): dropbacks, designed/scramble/kneel split, RZ & inside-5/10 carries, empirical xTD (league TD rate by yardline bucket applied to Purdy's plays), EPA/dropback, CPOE, aDOT, deep rate, sack rates; FTN join for PA%, IWT, QB-fault sacks, sneaks; league-scoring QB PPG tables (2024–25) for ranks and comps data/team-profiles/SF.md(built/refreshed 2026-07-07): play-caller/PROE/pace/personnel series, OL (PBWR 10th), depth-chart claims, vacated-target math, win total 10.5 (BetMGM), Purdy contract/backup, inside-10 pass-rate series, SF Standard pressure-to-sack notedata/adp/adp_ffc_2026-07-07.csv— Purdy 96.3, QB12 among ffc-ppr rows (FFC PPR mocks, 2026-07-07); band anchors: Daniels 86.6, Hurts 93.1, Goff 90.1, Stafford 100.9, C. Williams 106.3, Nix 115.6data/raw/sleeper/players_2026-07-07.json— age 26 (b. 1999-12-27), 5th season, Iowa State, no injury designation- Web (2026 offseason context): NBC Sports Bay Area OTA report + Niner Noise OTA recap (June 2026) — full participation, minicamp cancelled; SI/ESPN (2026 offseason) — turf toe fully recovered; Fantasy Life & CBS 2026 outlooks (June–July 2026) — market case ("low-end QB1 in 1QB leagues"); PFR advanced passing UNAVAILABLE (HTTP 403, 2026-07-07) → pressure rate marked UNVERIFIED