Houston Texans DST — 2026 evaluation
Verdict
FADE (medium confidence) at ADP 107.0 (FFC PPR mocks, 2026-07-07) — DST4 by ADP, a round-9 pick in a 12-team league, costing a startable skill player: George Kittle (103.0), Khalil Shakir (105.6), Ricky Pearsall (103.6), Kyle Monangai (103.8), Makai Lemon (107.2). The market's case is fair and strong: HOU was fantasy DST2 in 2025 (164 pts — FantasyPros, fetched 2026-07-07) on the league's #1 yardage defense (277.2 ypg), #2 in both points allowed (17.35/gm) and EPA/play allowed (−0.109), with total continuity — DC Matt Burke enters year 4, all five Pro Bowlers return, Anderson and Hunter are locked up, and the 9.5-win offense in front of it committed the fewest turnovers in franchise history. Why the market is wrong: it is paying a round-9, top-4 DST price for last year's finish, and roughly half of that finish (80 of 164 pts) came from the two columns this system regresses hardest — 28 takeaways (3rd) and 4 defensive TDs — while the repeatable signal quietly slipped: charted pressure rate fell from 6th (2024) to 15th (2025) and ESPN team pass-rush win rate ranked just 21st. And the weeks 1–4 the pick actually buys are close to unusable: Josh Allen (23.5 implied, DK opener), Burrow, Daniel Jones, and Dak — zero rookie QBs, ~22.7 avg implied — with the premium matchups (Cam Ward week 5, JAX-London at 20.0 implied week 6) arriving exactly when streaming is free. The regressed median (~137, DST3–5 range) beats a replacement DST12 by ~1.1 PPG in a scoring regime showing no verified inflation — that cannot repay a pick that costs Kittle or Shakir. FADE is price, not risk: this is a genuinely top-5 profile, and on a slide past ~130 (DST6-or-later territory) it flips to TARGET; if the league table lands turnover-inflated (≥3/TO), re-run immediately — HOU is the biggest beneficiary on the board.
Bull case
- The best takeaway engine in football, and it's structural: 28 takeaways in back-to-back seasons with identical 19 INT/9 FR splits, powered by a returning All-Pro CB duo (Stingley 43.3% comp allowed; Lassiter) and Anderson — the one unit where above-mean takeaway projection is defensible, and the single biggest winner if the league table lands turnover-inflated (≥3/TO).
- Continuity the methodology explicitly rewards, plus a rising offense: same DC scheme year 4, all five Pro Bowlers back, zero contract noise, 9.5 win total, and an offense that scored 404 with a franchise-record-low 12 giveaways — leads force opponent dropbacks and protect the PA brackets from short fields.
- Proven DST1–2 ceiling: this exact unit already sustained elite output across two seasons and finished DST2 (164). If takeaways hold near 28 again and the def-TD luck repeats even partially, the ceiling (~168) is DST1 territory — and 3 of the first 4 are at home.
Bear case
- Half the price is regression fodder: 80 of 164 points (28 TO × 2 + 4 def TDs × 6) came from the two noisiest columns in DST scoring; regressed per methodology (TO → 25, TDs → 2) the median sheds ~22 points, from DST2 to DST3–5 — while the market pays the DST2-memory price.
- The signal is trending the wrong way and the rush is two men deep: charted pressure fell 6th → 15th, team PRWR is 21st (concern band), edge depth (Barnett, Autry) left without replacement capital, and Hunter is 31 entering season 12 — one edge injury from a mid-pack rush whose sack column already depends on coverage.
- The pick buys September weeks it can't start: wks 1–4 = Allen (23.5 implied), Burrow, Jones, Dak — ~22.7 avg implied, zero rookie QBs — while DET/BAL/CHI cover those weeks off waivers for free and the round-9 slot costs George Kittle or Khalil Shakir, startable every single week.
Projection & comps
Bottom-up, standard DST scoring (assumed table above), 17 games:
| Component | Floor (p20) | Median | Ceiling (p80) | Basis |
|---|---|---|---|---|
| Opp. dropbacks | ~600 | ~620 | ~640 | 2025: ~623 (47 ÷ 7.54% — def_summary, pulled 2026-07-07); 2024: ~638; 9.5 win total (BetMGM, 2026-07-07) = leads force opponent passing |
| Sack rate → sacks | 6.0% → 36 | 7.0% → 43 | 8.0% → 51 | Two-year sack rates 7.68%/7.54% (4th/8th) haircut for the pressure slide (charting 6th→15th; PRWR 21st) and thinned edge depth (Barnett/Autry out — SI, 7/2026); conversion 20.9% already at the ~21% mean, so no luck to sell — the coverage-sack component (Stingley/Lassiter) persists with full secondary continuity |
| Takeaways → pts | 20 → 40 | 25 → 50 | 29 → 58 | 1.65/gm in both 2024 and 2025 (28 TO each — def_summary) regressed ~50% toward 1.3/gm ≈ 1.47/gm ≈ 25; the two-year repeat and elite CB duo argue the true rate sits above the mean, but per dst.md §1 never project 28+ to repeat |
| Def/return TDs → pts | 1 → 6 | 2 → 12 | 3 → 18 | Base rate 1.5–2.5 per dst.md §1; 2025 actual: 4 def TDs (NFL.com team stats via search 2026-07-07; Anderson/Stingley pick-sixes, Rankins/Togiai fumble returns — houstontexans.com) — a ~12-pt regression drag on its own |
| PA/gm → bracket pts | ~21 → 22 | ~18.5–19 → 30 | ~17.5 → 39 | 2025: 17.35 (2nd) realized ~35–37 bracket pts (~2.1/gm, backed out of the 164 total); good own offense (404 pts scored, 12 giveaways — fewest in franchise history) protects the brackets; tough early slate + mild defensive regression push the median up ~1.5 PA/gm |
| Safeties | 0 | 1 → 2 | 1 → 2 | League base rate |
| Season total | ~104 (6.1/gm) | ~137 (8.1/gm) | ~168 (9.9/gm) | Floor ≈ DST18–24; median ≈ DST3–5 (2025 scale: DST5 = 136, DST10 = 127); ceiling ≈ DST1–2 (2025 HOU DST2 = 164, SEA DST1 = 179 — FantasyPros) |
Games/startability risk: medium. Driver: not literal games — the weeks 1–4 card (three implied totals ≥ ~22 est.) means the round-9 pick may ride the bench for much of September; secondary drivers are the top-heavy rush (one Anderson or Hunter injury drops the unit a full band — Hunter is 31, entering season 12; Sleeper, 2026-07-07) and takeaway variance week to week.
Comps (profile: continuity-elite, takeaway-built top-2 finisher priced top-4):
- HOU 2024 → 2025 itself — the bull comp: this exact unit already repeated 28 takeaways and a top-8 sack rate once, and improved to DST2 when the offense stopped gifting short fields. Persistence is real; the median credits it (takeaways at 25, above pure mean-reversion).
- DAL 2021 → 2022 — the good-case regression landing: a DST1 built on a 34-takeaway spike saw the takeaway column normalize but held top-5 on pressure and personnel (directional; totals UNVERIFIED).
- NE 2019 → 2020 — the bear comp: a historic takeaway/TD-driven DST1 cratered once the noise columns normalized and script flipped (directional; totals UNVERIFIED).
- PIT 2022–2024 era — the median case: an elite-edge-duo unit with mid team-wide rush win rate, chronically priced top-5 on name value, returning DST4–10 seasons (directional; totals UNVERIFIED).
Signal core (dst.md §2)
| Metric | 2024 | 2025 | Band / read |
|---|---|---|---|
| Pressure rate (charted dropbacks) | 35.4%, 6th (nflverse participation, 724 charted dropbacks; computed 2026-07-07 — charting defn runs higher than PFR) | 31.7%, 15th (710 charted dropbacks; same basis) | Good-not-elite, and the one-year trend is down — the core signal slipped from top-6 to mid-pack |
| Sack rate | 7.68%, 4th (49 sacks — def_summary 2024) | 7.54%, 8th (47 sacks — def_summary 2025) | Good band (6.5–8.0), remarkably stable |
| PRWR (ESPN, team) | UNVERIFIED | 35%, rank 21st (ESPN win rates, through Wk 18, 2026-01-06, fetched 2026-07-07) | Concern band (<38) — but Anderson's 26.2% led the NFL individually and Hunter posted 24.1% (ESPN): an elite duo atop a thin team-wide rush |
| Pressure→sack conversion | ~19.1% (49 ÷ ~724×0.354, computed) | ~20.9% (47 ÷ ~225 charted pressures, computed) | At/below the ~21% mean both years — no conversion luck to sell; the sacks-above-pressure gap is coverage-driven (Stingley/Lassiter), which persists with continuity but rests on health |
| Takeaways/gm | 1.65 (28: 19 INT/9 FR, rank 5) | 1.65 (28: 19 INT/9 FR, rank 3 — def_summary; lg mean 1.15) | Elite two straight years, identical splits — still regress ~50% toward 1.3 per dst.md §1; the CB-driven INT engine is the best argument in the league for above-mean true talent, but 28 never gets projected to repeat |
| Blitz rate / coverage | 27.9% blitz; man 50.1% / zone 49.9% (participation, computed) | 23.2% blitz; man 32.1% / zone 67.9% (same; Burke took over play-calling ~Wk 4 2025 — team profile) | Mid blitz (no boom/bust penalty); the 2025 zone-lean shell under Burke is the carry-forward scheme (dc_new: false) |
| Opp. implied PPG, wks 1–4 | — | 23.5 (BUF, DK opener 2026-05-14) / ~22.5 est. CIN / ~21.5 est. @IND / ~23.5 est. DAL — only wk 1 has a posted line | Concern-adjacent (~22.7 avg) — zero rookie QBs; three top-10-caliber offenses; the one mitigant is 3 of 4 at home |
| Base-rate xFP | — | rebuilt ≈ 137 std pts ≈ DST3–5 | Top-5 range — the market rank (DST4) is roughly *right*; the price in skill-player terms is what's wrong |
| EPA/play allowed | −0.054, 6th (def_summary) | −0.109, 2nd (def_summary) | Elite |
| PA/gm | 21.88, 14th | 17.35, 2nd | Elite in 2025 — and partly an own-offense output (12 giveaways, fewest in franchise history) per dst.md §1 |
2×2 read (dst.md §2): sacks (8th) above pressure (15th charted / 21st PRWR) = the mild sell cell — but with a twist: conversion sits *at* the mean (20.9%), so the usual conversion-luck story doesn't apply. The gap is coverage sacks manufactured by the NFL's best young CB duo, which is more repeatable than luck but less repeatable than front-four dominance, and it makes the sack column hostage to secondary health. Project sacks at ~43, not 47 — and note the direction of the pressure trend (6th → 15th) is the single most important number in this eval.
Context (team profile: data/team-profiles/HOU.md, built 2026-07-07)
- Continuity is the green flag (dst.md §3): Matt Burke enters year 4 as DC (2nd full season as defensive play-caller after taking over from Ryans ~Week 4 2025; retained after an ARI HC interview — staff announced 2026-03-07);
dc_new: false, so 2025 scheme stats carry: 4-down one-gap SF/Ryans-tree front, zone-lean (67.9%), mid blitz (23.2%). All five 2025 Pro Bowlers return (Anderson, Stingley, Al-Shaair, Lassiter, Bullock); Anderson's fifth-year option exercised ($21.5M gtd — ESPN 4/2026), Hunter extended (bengals.com preview 6/2026), no contract noise. Arrivals: S Blankenship (3yr/$24.75M), R2 #36 DT McDonald, DE Logan Hall (2yr/$14M). - The leak is edge depth, not the top: rotation pieces Derek Barnett and Denico Autry are gone, leaving Hall (possibly kicking outside), Dylan Horton, and Ali Gaye behind the stars (SI training-camp storylines, July 2026; camp opens 7/21 rookies, 7/28 vets). Combined with Hunter at 31 entering season 12 (Sleeper, 2026-07-07) and a team PRWR of 21st, the rush is two men deep. Off-ball/interior departures (Harris → ATL, Settle → WAS) barely register per dst.md §3. Minor pre-camp housekeeping: DB Ajani Carter and DE Xavier Thomas (both injured) waived 2026-07-01 (SI).
- Own offense (§5 multiplier): good and improving. Win total 9.5 (O −120/U +100, BetMGM, as-of 2026-07-07); 404 points scored (2nd-most in franchise history) with 12 giveaways (fewest in franchise history — team profile); Stroud locked through 2027, OL upgraded (Teller/Smith/R1 C Rutledge). Few short fields against this defense, leads that force opponent dropbacks — the complementary-football multiplier points up, with a rookie-center caveat early.
- Market context: the top-4 DSTs are bunched at the round-8/9 turn — SEA 98.2, DEN 100.8, LAR 104.6, HOU 107.0 (FFC PPR, 2026-07-07) — exactly the peak-hype pricing dst.md §6 says never to pay. DEN at 100.8 is already a system FADE; HOU has a better takeaway engine and a friendlier weeks-5–6 stretch, but a worse pressure trend, and the same structural math applies.
Early-season streaming picture (weeks 1–6) — handoff for /weekly-edge
Schedule: houstontexans.com / ESPN schedule release (May 2026, verified via search 2026-07-07). Three of the first four at home; week 6 vs JAX in London (neutral). QB rates from data/stats/2025/passing.csv (sack rate = sacks ÷ (att + sacks); INT rate = INT ÷ att), computed 2026-07-07.
| Wk | Opp | Venue | Early line / implied (as-of) | Opp QB (2025 sack / INT rate) | Read |
|---|---|---|---|---|---|
| 1 | vs BUF | Home | BUF −1.5, O/U 45.5 → BUF implied 23.5 (DK opener, 2026-05-14) | Josh Allen (8.0% / 2.2%) | Fails the anchor (≥23 is non-negotiable per dst.md §8) — real sack equity vs Allen at home, but start only on roster inertia |
| 2 | vs CIN | Home | no line — ~22.5 est., UNVERIFIED | Joe Burrow (6.2% / 1.9%; 10-gm 2025 sample) | Coin flip — home + sack equity vs a shaky CIN OL, but the implied total will likely fail the ≤20 anchor |
| 3 | @ IND | Road | no line — ~21.5 est., UNVERIFIED | Daniel Jones (5.4% / 2.1%) — re-signed 2yr/$88M but coming off a Dec-2025 torn Achilles, expected ready wk 1 (ESPN/CBS, 3/2026) | Thin — road divisional vs a low-sack QB; upgrade sharply if Jones' Achilles lingers |
| 4 | vs DAL | Home | no line — ~23.5 est., UNVERIFIED | Dak Prescott (4.9% / 1.7%; 4,552 yds 2025) | Sit/stream away — low sack, low INT, high implied |
| 5 | @ TEN | Road | no line — TEN implied ~19 est., UNVERIFIED | Cam Ward (9.2% sack rate, 55 sacks / 1.3%) | Best early start — the league's premier sack donor, year-2 QB, low implied total |
| 6 | vs JAX | London (neutral) | HOU −1.5, O/U 41.5 → JAX implied 20.0 (DK, 2026-05-13 via SI odds survey) | Trevor Lawrence (6.8% / 2.1%) | Start — at/near the ≤20 anchor with sack equity |
Likely-available pivots (FFC ADP 2026-07-07): DET (139.1) opens home vs NO — the cleanest wk-1 stream on the board (per denver-broncos-dst.md). BAL (154.6, system TARGET) has premium home weeks 2 (vs NO/Shough, 19.5 implied) and 4 (vs TEN/Ward, 19.5) per the BAL eval. CHI (166.0) opens @CAR (21.0 implied), home MIN wk 2, and a premium wk-4 home vs NYJ (18.5) per the CHI eval. Net: for weeks 1–4, free-agency units beat the round-9 pick in at least three of four weeks — the FADE in miniature. HOU itself becomes the asset in weeks 5–6 (Ward, JAX-London).
Tripwires (re-run this eval if any fire)
- League DST table gets filled in — the load-bearing unknown, and HOU-specific: a turnover-inflated table (turnovers ≥3, or INT-return yardage) disproportionately rewards this profile and likely flips the verdict to HOLD/TARGET at 107; a steep points-allowed-bracket table also helps (own offense + 17.35 PA baseline). Re-run §7 classification either way.
- ADP slides past ~130 (round 11+, DST6-or-later) — flips to TARGET: top-5 profile at zero meaningful opportunity cost.
- Anderson or Hunter injury/absence in camp or preseason — the rush is edge-concentrated with thinned depth; either loss drops the median a full band and hardens the FADE.
- Week 1 BUF implied total drops below ~21 (e.g., Allen injury) or the CIN/DAL lines open under ~21 — the early-card half of the thesis weakens; re-price toward HOLD.
- A meaningful veteran edge addition before camp (SI flags the depth hole; camp opens 7/28) — restores the rotation behind the stars; small upgrade to the pressure projection.
Sources
data/stats/2025/def_summary.csv,data/stats/2024/def_summary.csv— nflverse via nflreadpy, pulled 2026-07-07 (2025: 47 sacks/7.54% rank 8, 28 TO [19 INT/9 FR] rank 3, PA 17.35 rank 2, EPA −0.109 rank 2; 2024: 49 sacks/7.68% rank 4, 28 TO rank 5, PA 21.88 rank 14, EPA −0.054 rank 6; ranks + league mean TO/gm [1.15] computed 2026-07-07)data/stats/2025/participation.csv,data/stats/2024/participation.csv— nflverse, pulled 2026-07-07: charted pressure rate 31.7% (15th, 710 charted dropbacks) 2025 / 35.4% (6th, 724) 2024; blitz 23.2%/27.9%; man-zone 32.1/67.9 (2025), 50.1/49.9 (2024); conversion 20.9%/19.1% (computed 2026-07-07; charting definition runs higher than PFR — team profile's 23.6% figure uses a different denominator)data/stats/2025/passing.csv— nflverse, pulled 2026-07-07 (opponent QB sack/INT rates: Allen, Burrow, D. Jones, Prescott, Ward, Lawrence; Stroud line)data/adp/adp_ffc_2026-07-07.csv— FFC PPR mocks, 2026-07-07 (HOU DST 107.0 = DST4; SEA 98.2, DEN 100.8, LAR 104.6; skill range: Kittle 103.0, Pearsall 103.6, Monangai 103.8, Shakir 105.6, Lemon 107.2; pivots DET 139.1, PIT 148.4, BAL 154.6, LAC 161.5, CHI 166.0)data/team-profiles/HOU.md— built 2026-07-07 (Burke DC yr 4 / play-caller retained, staff announced 2026-03-07; scheme rates; personnel in/out incl. Blankenship, McDonald R2 #36, Hall; Anderson $21.5M option — ESPN 4/2026; Hunter extension — bengals.com 6/2026; own offense 404 pts / 12 giveaways; win total 9.5 O−120/U+100 BetMGM as-of 2026-07-07; 277.2 ypg #1, five Pro Bowlers — bengals.com preview 6/2026; Stingley 43.3% comp allowed)data/raw/sleeper/players_2026-07-07.json— Hunter age 31 / 11 yrs exp; Anderson 24; Stingley 25; Lassiter 23- ESPN 2025 win-rate rankings (espn.com id 46138675, through Wk 18, published 2026-01-06, fetched 2026-07-07): HOU team PRWR 35% rank 21st (team PBWR 56%, 30th); Anderson individual PRWR 26.2% #1 in NFL, Hunter 24.1% (via search 2026-07-07)
- NFL.com HOU team stats + houstontexans.com "Every Texans Defensive TD in 2025" (via search, 2026-07-07): 4 defensive TDs in 2025 (Anderson/Stingley pick-sixes; Rankins/Togiai fumble-return scores)
- houstontexans.com / ESPN / FBSchedules (2026 schedule release, May 2026, verified via search 2026-07-07): wks 1–6 = vs BUF, vs CIN, @IND, vs DAL, @TEN, vs JAX (London)
- DraftKings Network (dknetwork, 2026-05-14, via search 2026-07-07): wk 1 BUF −1.5 @ HOU, O/U 45.5 → BUF implied 23.5
- SI betting odds survey "Odds for every NFL game we can bet on so far" (as-of 2026-05-13, fetched 2026-07-07): wk 6 HOU −1.5 vs JAX (London), O/U 41.5 → JAX implied 20.0; no posted lines for HOU wks 2–5 (implied totals marked est./UNVERIFIED)
- ESPN / CBS Sports / colts.com (March 2026, via search 2026-07-07): Daniel Jones re-signed 2yr/$88M; recovering from Dec-2025 torn Achilles, expected ready for the season opener
- SI Texans "Training Camp Dates Announced: 5 Storylines" + "Texans Cut Two Injured Defenders" (July 2026, via search 2026-07-07): camp 7/21 rookies / 7/28 vets; edge depth hole behind Anderson (12 sacks, DPOY runner-up) and Hunter (15 sacks) after Barnett/Autry departures; Carter/Thomas waived 7/1
- FantasyPros 2025 NFL DST statistics (fetched 2026-07-07, via denver-broncos-dst.md): HOU DST2 164 pts; SEA DST1 179; DST12 NO 118; DST5 136, DST10 127 (scale anchors)
evaluations/players/2026/denver-broncos-dst.md,buffalo-bills-dst.md,chicago-bears-dst.md,baltimore-ravens-dst.md(2026-07-07): inflation-test computation; DET/BAL/CHI pivot cards reused for the streaming picture