ACHC · Acadia Healthcare Company, Inc. · Stock research

Current Thesis

The binary resolved and the tape went the other way. Q2 landed after the close on 2026-07-28 and beat the company's own guide on all three lines — revenue $865.8M against a guided $835–850M, adjusted EBITDA $149.2M against $142–152M, adjusted EPS $0.38 against $0.30–0.40 — and the FY26 floor was lifted (revenue $3.40–3.45B from $3.37–3.45B, adj. EBITDA $590–615M from $580–615M, adj. EPS $1.45–1.60 from $1.35–1.60). Underneath, the quarter deteriorated year over year: adjusted EBITDA $149.2M versus $201.8M, adjusted EPS $0.38 versus $0.83, net income attributable to Acadia $10.9M versus $30.1M, same-facility revenue flat with patient days +0.8% and revenue per patient day -0.8%. The Florida money arrived as a $22.3M supplemental payment tied to the prior state fiscal year. Two desks raised targets after the print (Barclays to $32 from $25 on 2026-07-29, Guggenheim to $39 from $38 on 2026-07-30) while price fell from the $35.61 52-week high to the 2026-08-14 close of $30.83, -13.4% off the high with RSI(14) at 44.9. That leaves the stock 2.1% above the $30.20 prior 52-week high that the July leg broke through. The retest of that shelf is live and unresolved, and nothing on the company calendar forces it either way inside 30 days.

Bullish and bearish views on Acadia Healthcare Company, Inc.

The model's bull view on Acadia Healthcare Company, Inc. (ACHC), in brief: Guide raised at the print (2026-07-28): FY26 revenue $3.40–3.45B, adj. The bear view: The P&L got worse, not better: adj. EBITDA -26% YoY ($149.2M vs $201.8M), adj. EPS -54% ($0.38 vs $0.83), net income $10.9M vs $30.1M. A raised annual guide does not remove a quarter that shrank. A $28.6M reserve adjustment "for expected settlements" inside a $39.3M increase in… Both cases follow in full.

Bull Case

• Guide raised at the print (2026-07-28): FY26 revenue $3.40–3.45B, adj. EBITDA $590–615M, adj. EPS $1.45–1.60, operating cash flow $350–400M. Every point of the guided cash-flow range sits above every point of the guided capex range of $235–255M, which was itself cut from the earlier plan.
• Q2 cleared its own bar on all three guided lines: revenue $865.8M vs guided $835–850M; adj. EBITDA $149.2M vs $142–152M; adj. EPS $0.38 vs $0.30–0.40. StockStory scored the $0.38 as ~9.7% above consensus (2026-07-28).
• Normalized volume is not flat: same-facility revenue growth would have been +3.2% adjusting for supplemental-payment timing, per the Q2 8-K — the flat headline is partly a comparison artifact against a richer 2025 supplemental period.
• Capacity keeps landing: 240 licensed beds added in Q2 — a 144-bed JV with Orlando Health (Florida), a 96-bed JV with Methodist Jennie Edmundson Hospital (Iowa), plus two new comprehensive treatment centers.
• Sell-side band still sits above spot: high forecast $39, six raises between 2026-07-09 and 2026-07-30 (UBS $39, Keybanc $40, TD Cowen $36, Guggenheim $38 then $39, Barclays $32), rating split 8 Buy / 6 Hold / 1 Sell.

Bear Case

• The P&L got worse, not better: adj. EBITDA -26% YoY ($149.2M vs $201.8M), adj. EPS -54% ($0.38 vs $0.83), net income $10.9M vs $30.1M. A raised annual guide does not remove a quarter that shrank.
• A $28.6M reserve adjustment "for expected settlements" inside a $39.3M increase in professional-liability/general-liability costs (Q2 8-K, 2026-07-28). That is money reserved against claims management expects to pay, booked in the quarter the re-rate was supposed to validate.
• The legal overhang is still open and still expensive: $7.5M of government-investigation costs in Q2 alone; the DOJ grand-jury subpoena disclosed in 2024 and the SEC inquiry remain unresolved. A charging decision, deferred-prosecution agreement or consent decree is a discrete single-session gap risk that no guide raise offsets.
• $105M jury verdict (2026-05-12): subsidiary Fashion Valley CTC, $35M compensatory plus $70M punitive, under post-trial motions and appeal.
• No buyback into the drawdown: zero share repurchases in Q2 2026 and zero for the first six months, per the Q2 filing — the balance sheet was not put behind the stock while it fell 13% off the high.
• The Florida line is period-specific: the $22.3M relates to the prior state fiscal year. Recurrence at that magnitude is an assumption, not a disclosed run-rate.
• Price sits above the average target: $30.83 (2026-08-14) versus a $29 consensus average, with the dispersion running $13 low to $39 high.

Setup & Price Structure

• Reference close 2026-08-14: $30.83. 52-week high $35.61, -13.4% below it. Three-month return +19.4%. RSI(14) 44.9.
• The June-to-July advance ran from roughly $24.48 through the $30.20 prior 52-week high to $35.61. Roughly two-thirds of that leg has been surrendered.
• $30.20 is the structure. It was the ceiling for a year, it was the breakout level in July, and it is 2.1% below current price. A retest in progress, not a base that has formed.
• RSI 44.9 alongside a +19.4% three-month return: momentum has fully unwound while the trend-length return is still positive. That combination resolves in one direction or the other; it does not persist.
• Crowding observables, stated as observables: four PT raises in the nineteen days before the print versus two in the days after — flow positive but decelerating; price trading above the $29 consensus average; the sell-side raising targets on 07-29 and 07-30 into a tape that then fell 13%; no insider buying and no corporate buyback disclosed for the six months to 2026-06-30.
• Short interest is currently unverifiable from one consistent series: vendor data pulled in mid-August 2026 diverged sharply (one series ~24% of shares outstanding, another ~4% of float). The squeeze-fuel leg of the July story should be treated as unconfirmed until the next FINRA settlement-date publication reconciles it.
• The narrative is maturing. Dated by the 2026-07-28 print (the event the narrative was priced for came and went), the 2026-07-29/07-30 target raises (coverage still expanding, at half the pre-print rate), and the -13.4% fade from the 07-print high. It has not broken — that would require losing $30.20 on a weekly basis — and it is no longer accelerating.

Catalyst Calendar (next 30 days)

• ~2026-08-26 (est.) — FINRA/Nasdaq short-interest publication for the 2026-08-14 settlement date. Resolves the vendor disagreement on how much short base survived the July run.
• ~2026-10-27 (est.) — Q3 2026 print, outside the 30-day window. This is the next date that tests whether the Florida supplemental line repeats and whether PLGL reserves build again.
• Undated but live: any 8-K disclosing a DOJ or SEC resolution; a post-trial ruling on the $105M Fashion Valley verdict; Florida program approval for the current state fiscal year.

Elapsed catalysts

• No company-scheduled event confirmed inside the window as of 2026-08-16 — no earnings date, and no September conference appearance listed on the IR event calendar at time of writing. (passed 10d ago)

What Would Change Our Mind

The thing that breaks first is the shelf. The $30.20 prior 52-week high is what separates "July was a breakout" from "July was a spike into a resolved catalyst"; a weekly close below $30 surrenders it and re-labels the whole re-rate as a failed breakout, with the June low near $24.48 the next reference. Fundamentally, the read fails if the Q3 print (est. late October) shows no comparable Florida supplemental contribution and another PLGL reserve adjustment — that would make Q2's raised FY26 floor a timing artifact rather than an operating inflection. A DOJ or SEC charging 8-K, or an appellate affirmance of the $105M verdict, breaks it independently of price. On the other side, a weekly close back above $33 would put the post-print range's upper half back in play and argue the fade was digestion; absent that reclaim by the Q3 print, and absent any further target raises, the theme reads saturated rather than maturing.

Correlation Notes

• Closest read-through is Universal Health Services' behavioral segment: the same Medicaid state-directed-payment mechanics drive both, so a CMS action on supplemental payment programs moves them together regardless of company-specific execution.
• Florida concentration is the single-state factor. The $22.3M Q2 contribution came from one state program tied to one fiscal year; there is no disclosed diversification of that line across states.
• The DOJ/SEC leg is idiosyncratic and does not travel with the healthcare-facilities group — group strength offers no protection against a charging headline.
• The name is a domestic services business with no AI or semiconductor factor exposure; it will not move with the momentum complex that drives most of the current high-attention tape.