Skip to content
FrontierPicks

Dossier · SAH · Dormant

SAH · Sonic Automotive, Inc. · Stock research

Last analysed ·

Resolved Graded and closed 2026-08-18 at low conviction — the published kill line fired. Coverage continued after the close; the read below is dated 2026-08-22 and is not part of the scored record.

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.).

Kill line

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.

Pick status

Invalidated resolved published kill line fired How this is scored →

Latest analysis and events for SAH —

As of 22 August 2026, the latest FrontierPicks analysis for Sonic Automotive, Inc. (SAH): 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.).

Kill line: 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.

Next dated event on file: — catalyst in 20d.

Current Thesis

Coverage 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.

Since 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.

The 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.).

Bullish and bearish views on Sonic Automotive, Inc.

The model's bull view on Sonic Automotive, Inc. (SAH), in brief: 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). The bear view: 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. Both cases follow in full.

Bull Case

  • 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).
  • 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%.
  • EchoPark volume is scaling: 19,601 retail used units (+17% YoY), revenue $582.9M (+15%), record Q2 gross profit $64.3M.
  • 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.
  • 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.
  • 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.

Bear Case

  • 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%.
  • 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%.
  • 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.
  • EchoPark footprint expansion begins in Q4 2026, adding fixed cost to a segment whose GPU guidance was just reduced roughly $300 at the midpoint.
  • 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.
  • The prior note's structural reference at $78 did not hold, and no consolidation has formed beneath it yet.

Setup & Price Structure

The 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.

Positioning and crowding observables, stated as observables:

  • 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.
  • 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.
  • 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.
  • No earnings date falls inside 30 days. The only dated company event before Q3 is the 2026-09-15 dividend record/ex date.
  • 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.

Catalyst Calendar (next 30 days)

  • ~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.
  • ~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.
  • 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.
  • 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.

What Would Change Our Mind

The 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.

In 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.

On 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.

Correlation Notes

  • 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.
  • 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.
  • Rate sensitivity arrives through two channels — retail financing affordability on the customer side and floor plan interest expense on inventory carried.
  • 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.
  • 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.

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.

Related · shared themes

See also · stocks to watch