{
  "@context": "https://frontierpicks.com/schemas/dossier.v1.json",
  "ticker": "SAH",
  "name": "Sonic Automotive, Inc.",
  "url": "https://frontierpicks.com/dossiers/SAH/",
  "json_url": "https://frontierpicks.com/dossiers/SAH.json",
  "status": "DORMANT",
  "current_conviction": "LOW",
  "graded_conviction": "LOW",
  "archetype": {
    "code": "a4",
    "n": 4
  },
  "current_thesis": "GPU-normalization narrative broke at the 2026-07-30 print: revenue and adjusted EPS beat consensus, but adjusted EPS fell 17% YoY, franchised segment income −23%, and EchoPark FY26 GPU guidance was cut ~$300 at the midpoint. Shares round-tripped the entire June–July advance to −26.8% off the $112.66 high. Nothing company-specific resolves before Q3 (~2026-10-22, est.).",
  "invalidation_trigger": "A weekly close below $78 breaks the pre-advance shelf the June–July move launched from and turns the round-trip into a fresh downtrend leg; secondarily, a Q3 print (~2026-10-22, est.) showing consolidated new-vehicle GPU under the $2,850–$3,000 FY26 band.",
  "catalyst_date": "2026-09-15",
  "outcome": "INVALIDATED",
  "outcome_date": "2026-08-18",
  "invalidation_fired": true,
  "themes": [
    "consumer-discretionary-rotation",
    "ev-autonomous-mobility"
  ],
  "tags": [],
  "sources": [],
  "notes": [
    "Dual-class structure: Class A trades on the NYSE as SAH; supervoting Class B is held largely by the founding Smith family, so public holders do not control the vote.",
    "Floor plan debt is inventory working capital. Headline total-debt screens overstate leverage versus peers unless floor plan is excluded.",
    "Reported and adjusted results diverge widely here: Q2 2026 showed reported net income +226% YoY against adjusted net income -23% YoY.",
    "OEM franchise agreements carry consent rights over dealership acquisitions and change of control, so M&A pace is not fully management-controlled.",
    "Quarterly repurchase volume is only visible with the 10-Q, so the pace of the largest identified structural bid lags the tape by weeks."
  ],
  "body_markdown": "## Current Thesis\n\nCoverage of this name since 2026-07-30 has hung on one question: had per-unit gross profit normalization finished. The Q2 print answered no. Revenue of $3.934B beat the $3.772B consensus and adjusted EPS of $1.82 beat $1.74, yet adjusted diluted EPS fell 17% YoY, Franchised Dealerships segment income fell 23% to $70.7M, EchoPark segment income fell 38% to $7.2M, and EchoPark's FY26 total GPU guidance was cut to $3,100–$3,300 from $3,400–$3,600. Shares closed down 10.9% on 2026-07-30.\n\nSince the prior update the price structure has caught up with that. The $78 area the June–July advance launched from is gone: the last completed daily close was $77.45 on 2026-08-21, and 2026-08-21 was a Friday, so the week finished under it. Price sits 31.3% below the $112.66 52-week high, the three-month price change is −0.4%, and RSI(14) reads 25.7.\n\nThe leg still on offer is no longer \"normalization is over.\" It is a washout leg — fixed operations and F&I supplying more than 75% of gross profit, roughly $528M of repurchase authorization remaining, a $1.64 annualized dividend, and a quote at the floor of the published target range. Nothing company-specific tests that leg before Q3 results (~2026-10-22, est.).\n\n## Bull Case\n\n- The recurring layer held through the margin reset: fixed operations and F&I produced **more than 75% of total gross profit** in Q2 2026, with same-store fixed operations gross profit +2% YoY, warranty +3% and customer pay +1% (reported 2026-07-30).\n- Management raised the low end of FY2026 new-vehicle GPU guidance to $2,850 from $2,700 at the Q2 print, and kept EchoPark FY26 adjusted EBITDA at $35–40M while lifting the EchoPark volume-growth target to 12–15%.\n- EchoPark volume is scaling: 19,601 retail used units (+17% YoY), revenue $582.9M (+15%), record Q2 gross profit $64.3M.\n- Powersports compounds off a small base — Q2 revenue $73.5M (+53% YoY), gross profit $19.7M (+58%), segment income $2.3M against roughly breakeven a year earlier; five Harley-Davidson stores bought April 2026 add ~$100M annualized revenue.\n- The company is the identified marginal buyer of its own stock: ~2.1M Class A shares repurchased in Q1 2026 (a 6% reduction versus 2025-12-31), with $500M added to authorization in April 2026 and ~$528M remaining.\n- Published targets still sit above the market. As of 2026-08-19 the 12-month range spans $72 to $139 with an average near $98.55, against the 2026-08-21 close of $77.45.\n\n## Bear Case\n\n- Every per-unit line compressed in Q2 2026: same-store new-vehicle GPU −16% YoY to $2,872, same-store used GPU −13% to $1,401, F&I gross profit per retail unit −4% to $2,619, EchoPark combined used + F&I GPU −12% to $3,292. Consolidated gross profit grew 2% on revenue up 8%.\n- Adjusted SG&A ran at 72.0% of gross profit against management's low-70s ambition, and the denominator is the line under pressure. Fixed operations margin itself slipped 30bp to 51.0%.\n- The de-rating has been sell-side led since the print. JPMorgan cut its target to $76 on 2026-07-13 and again to $72 on 2026-08-04; Morgan Stanley cut the rating to Underweight from Equal Weight on 2026-08-07 with a $72 target; Barclays lowered its Equal-Weight target to $87 from $92 on 2026-08-19. The last three published actions all moved down.\n- EchoPark footprint expansion begins in Q4 2026, adding fixed cost to a segment whose GPU guidance was just reduced roughly $300 at the midpoint.\n- CEO David Smith framed affordability pressure on the 2026-07-30 call as pushing consumers to repair rather than replace — supportive for parts and service, a headwind to unit volume and new-vehicle GPU.\n- The prior note's structural reference at $78 did not hold, and no consolidation has formed beneath it yet.\n\n## Setup & Price Structure\n\nThe narrative is **dead** — the narrative failed and the structure broke, both on dated events. The GPU-normalization-complete story died at the 2026-07-30 print (−10.9% that session); the price shelf that story was standing on gave way on the weekly close of $77.45 on 2026-08-21. Between those two dates the stock went from the 2026-08-07 close of $82.44 to $77.45 with no company-specific news beyond one target reduction.\n\nPositioning and crowding observables, stated as observables:\n\n- Price is below, not above, its longer averages: 31.3% under the $112.66 52-week high with RSI(14) at 25.7 as of 2026-08-21. This is a de-rating profile, and an oversold reading is a condition, not a floor.\n- Headline flow over the trailing 30 days is entirely analyst-action items (Barclays 2026-08-19, Morgan Stanley 2026-08-07, JPMorgan 2026-08-04). No product, deal or consumer headline appeared — thin new attention.\n- The gap between the ~$98.55 average published target (2026-08-19) and the $77.45 close has been closing from the target side, not the price side.\n- No earnings date falls inside 30 days. The only dated company event before Q3 is the 2026-09-15 dividend record/ex date.\n- No insider transactions surfaced in the filing feed across the covered window; the repurchase program is the only identified structural bid, and its Q3 pace is unknown until the 10-Q.\n\n## Catalyst Calendar (next 30 days)\n\n- **~2026-09-04 (est.)** — August US light-vehicle SAAR from industry and automaker reporting. Tests whether affordability is now cutting units, after Q2 same-store retail new-vehicle units ran roughly flat.\n- **~2026-09-08 (est.)** — Cox Automotive Manheim Used Vehicle Value Index, August full-month read. Wholesale values set acquisition costs for both the franchised used desk and EchoPark, reaching reported GPU with roughly a quarter's lag.\n- **2026-09-15** — Ex-dividend and record date for the $0.41 quarterly dividend declared 2026-07-30, payable 2026-10-15. Confirms the $1.64 annualized payout survives the margin reset; it resolves nothing about GPU.\n- Beyond the window: **~2026-10-22 (est.)** Q3 2026 results (Q3 2025 landed 2025-10-16), and **~2026-Q4 (est.)** the start of EchoPark footprint expansion.\n\n## What Would Change Our Mind\n\nThe structural reference is already lost — the $78 shelf broke on the 2026-08-21 weekly close, so the question is now whether a base forms above the $72 low-target cluster or price cuts straight through it. A weekly close below $72 would take the quote under every published target on the board (JPMorgan $72 on 2026-08-04, Morgan Stanley $72 on 2026-08-07) and argue the market is repricing the franchise rather than the cycle.\n\nIn the other direction, the August break returns to noise only on a weekly close back above the $82–$83 zone the stock held on 2026-08-06 and 2026-08-07 ($83.11 and $82.44). Absent that, an oversold RSI of 25.7 is a description of the tape.\n\nOn fundamentals, the normalization-complete argument gets rebuilt on evidence if the Q3 report (~2026-10-22, est.) shows consolidated new-vehicle GPU inside the $2,850–$3,000 FY26 band, adjusted SG&A back below 72.0% of gross profit, and EchoPark segment income above the $7.2M it printed in Q2. A Q3 GPU figure under $2,850, or any reduction to the $35–40M EchoPark FY26 EBITDA range, does the opposite. The 2026-09-15 dividend date passing without incident changes nothing about any of this.\n\n## Correlation Notes\n\n- SAH trades with the franchised-dealer complex (AN, LAD, ABG, GPI, PAG). Because the Q2 compression is being priced as an industry cycle, a peer print or guidance revision reads across to SAH ahead of its own ~2026-10-22 (est.) date.\n- Two macro series drive the whole group: the Manheim Used Vehicle Value Index, which sets wholesale acquisition costs roughly a quarter before they hit reported GPU, and monthly SAAR, which sets unit throughput.\n- Rate sensitivity arrives through two channels — retail financing affordability on the customer side and floor plan interest expense on inventory carried.\n- Idiosyncratic exposures peers do not carry: EchoPark as a standalone used-vehicle format, and Powersports, where the five Harley-Davidson stores bought April 2026 add ~$100M annualized revenue.\n- The dual-class structure mutes the governance channels that operate on single-class peers: Class A trades on the NYSE, while supervoting Class B is held largely by the founding Smith family.",
  "first_seen": "2026-07-30",
  "last_analyzed": "2026-08-22T10:01:29+00:00",
  "last_synthesized": "2026-08-22",
  "last_update_source": "watchlist_research",
  "license": "Content © FrontierPicks. Cite the canonical URL."
}