AMSC · American Superconductor Corporation · Stock research

Refresh of coverage first published 2026-04-22; advances the 2026-08-24 note.

AMSC — American Superconductor Corporation

Current Thesis

The prior update graded against a weekly close below $33. That broke on 2026-08-21 at $30.01. The next session took out that low: the 2026-08-24 close printed $28.92, the lowest close of the move and the first sub-$30 close since the de-rating began. The shares sit 56.6% below the $66.68 52-week high with a three-month price change of -45.6%.

RSI(14) reads 23.8 on the 2026-08-24 close, under the conventional 30 line. The same series read 45.1 one session earlier at the 2026-08-21 close — a 21-point single-session shift, which comes from the rolling window dropping mid-August up-days at the same time price made a lower low. That decomposition is inferred from two dated readings; it is not measured from order flow.

Company news flow remains empty. Nothing dated after the 2026-08-05 Q1 FY2026 results has been issued, and the last broker action of record is Roth Capital's 2026-08-10 Buy reiteration with the target cut to $55 from $58. Thirteen sessions have passed since the print without a fresh negative headline, and price has fallen in a straight line anyway.

Life-cycle label on the site's scale: dead. The narrative failed and the structure is broken. Four dates carry it: the 2026-05-28 beat-and-soft-guide that started the de-rating; the late-June 2026 Russell 3000E and Russell Microcap deletions that removed the passive bid; 2026-08-05, when record revenue of $94.073M, record quarterly orders above $130M and record twelve-month backlog above $300M were met with lower closes; and 2026-08-24, a new closing low. The broader grid/AI-power theme is a separate object still reading maturing — this name decoupled from it downward in late May and has not re-coupled.

What is on offer is a mean-reversion case on a compounding order book against a chart with no base. The order book is measured. The base is not there.

Bullish and bearish views on American Superconductor Corporation

The model's bull view on American Superconductor Corporation (AMSC), in brief: Q1 FY2026 (quarter ended 2026-06-30, reported 2026-08-05): revenue $94.073M, +30% YoY from $72.4M and $6.3M above the $87.8M consensus — the first quarter above $90M in company history. The bear view: The de-rating survives good news. Record revenue, record orders and record backlog on 2026-08-05 were followed by thirteen sessions ending at a new closing low of $28.92 on 2026-08-24. Price is discounting the guide, not the print. Q1 adjusted EPS $0.16 missed the $0.21 estimate… Both cases follow in full.

Bull Case

• Q1 FY2026 (quarter ended 2026-06-30, reported 2026-08-05): revenue $94.073M, +30% YoY from $72.4M and $6.3M above the $87.8M consensus — the first quarter above $90M in company history. Grid $76.3M, Wind $17.8M.
• Orders exceeded $130M in the June quarter. Excluding the single largest recently announced award, bookings still exceeded $100M against an approximate $70M quarterly average in the prior fiscal year (CEO Daniel McGahn, 2026-08-06 call). The step-up is broad rather than one contract.
• Twelve-month backlog above $300M and total backlog above $400M at 2026-06-30, versus roughly $280M twelve-month at 2026-03-31.
• The 2026-07-28 award — $25M turnkey integrated power-quality package for a North American utility serving a large-scale mining development (STATCOM, capacitor banks, shunt reactors, 138kV transformer), delivery in fiscal 2027 — is the largest individual mining-development order the company has booked.
• Operating cash flow $16.0M in the June quarter; cash, equivalents and restricted cash $153.1M at 2026-06-30 against $147.6M at 2026-03-31; total debt $4.00M (10-Q filed 2026-08-06). No financing event sits behind the drawdown, and no dilution has been announced into the decline.
• Revenue guidance has behaved as a floor rather than a forecast: the 2026-05-28 guide of ">$85.0M" produced $94.073M. The Q2 guide issued 2026-08-05 carries the identical ">$85.0M" wording against $87.55M consensus.
• Sell-side has not capitulated. Four analysts polled by S&P Global carried an average target of $61.75, low $54, high $71, per stockanalysis.com checked 2026-08-25; MarketBeat's coverage of the same four put the average at $62.50 on 2026-08-10.

Bear Case

• The de-rating survives good news. Record revenue, record orders and record backlog on 2026-08-05 were followed by thirteen sessions ending at a new closing low of $28.92 on 2026-08-24. Price is discounting the guide, not the print.
• Q1 adjusted EPS $0.16 missed the $0.21 estimate despite the revenue beat, and the Q1 guide had carried a >$3.0M / $0.07 GAAP floor. Margin, not demand, is where the model is breaking.
• The Q2 guide sits below consensus on every line: revenue >$85.0M vs $87.55M; GAAP net income >$1.0M ($0.02) vs $0.12; non-GAAP >$8.0M (~$0.17) vs $0.23. That is the second consecutive sub-consensus guide after 2026-05-28.
• The 50-day moving average stood at $39.12 and the 200-day at $38.17 on 2026-08-05, both above every close recorded since and both declining. There is no rising average under this price to pull back into.
• Passive demand was structurally removed by the late-June 2026 Russell 3000E and Russell Microcap deletions. Rallies now clear against thinner books.
• Trailing EPS of $3.05 and the resulting low-double-digit trailing multiple rest on FY2025 GAAP net income of $133.81M, which included a deferred-tax valuation-allowance release. That is not comparable to a peer operating multiple.
• The gap between the $61.75 four-analyst average target and the $28.92 close leaves room for further cuts that arrive as headlines; only one target has been revised since the print.

Setup & Price Structure

Reference close 2026-08-24: $28.92. The $30.99 post-print low of 2026-08-10 and the $30.01 weekly close of 2026-08-21 are both now overhead. The nearest structural reference below is the 52-week-low area near $25.28 in the adjusted daily series; above, the first meaningful resistance is the lost $34.50 early-July shelf, then the declining 50-day at $39.12.

Positioning observables, stated as observables:

• Short interest 3.40M shares, 7.27% of float, 3.95 days to cover as of the 2026-08-05 dataset. No post-break reading has been published yet.
• Institutional ownership 69.61% on the same dataset, against a register that lost its index-tracking component in late June.
• Five-year beta 3.28 — roughly 3x broad-tape sensitivity, in both directions.
• CEO and CFO share sales run under Rule 10b5-1 plans (CEO plan adopted 2025-08-15, modified 2026-02-26) covering tax withholding on vesting restricted stock. No discretionary open-market insider buying has been disclosed into the decline.
• Retail-sentiment coverage has thinned rather than clustered: no company release since 2026-08-05, one broker note since, and no new headline into the 2026-08-24 low.

RSI(14) at 23.8 marks an oversold condition, which is a statement about the indicator, not about a bottom. Nothing in the price series shows a higher low, a volume shelf or a reclaimed average.

Catalyst Calendar (next 30 days)

• ~2026-08-26 (est.) — FINRA short-interest publication covering the mid-August settlement date.
• ~2026-09-10 (est.) — following FINRA publication, end-August settlement. Two consecutive readings separate covering from new short demand at the lower level.
• 2026-09-18 — quarterly index rebalance and quadruple witching. A liquidity event, not a fundamental one, and the first since the Russell deletions removed the passive holder base.
• No company-scheduled fundamental event falls inside the window. The guided September quarter does not close until 2026-09-30 and Q2 FY2026 results are expected ~2026-11-04 (est.).

What Would Change Our Mind

The structure that broke is the whole case. The $34.50 early-July shelf went, the $33 weekly level went on 2026-08-21, and the $30 handle went on 2026-08-24 — three references lost in seven weeks with no reclaim of any of them. The next reference is the 52-week-low area: a weekly close below $25.28 would confirm that no base formed at the August lows and would leave the name with no prior support until the pre-2024 range. That is the condition being graded.

On the other side, the case reopens on evidence rather than on price alone. A reclaim of the $34.50 shelf on a weekly close, with the 50-day flattening from $39.12, would be the first structural change since 2026-05-28. Fundamentally, the ~2026-11-04 Q2 print clearing the >$85.0M revenue and >$0.17 non-GAAP floors and showing twelve-month backlog above the >$300M reported at 2026-06-30 would break the pattern of guides the market discounts. A third sub-consensus guide at that print does the opposite.

A second broker cutting below the current $54 low target would confirm that sell-side capitulation is still ahead rather than behind.

Correlation Notes

• Beta 3.28 over five years means sector risk-off arrives amplified. Drawdowns in the AI-power and electrical-equipment complex — GE Vernova, Powell Industries — transmit here with leverage regardless of the order book.
• The index linkage that used to provide a passive bid was severed in the late-June 2026 Russell reconstitution. Correlation to small-cap index flows should be weaker from that date forward; correlation to sector sentiment is not.
• Segment mix splits the correlation: Grid at $76.3M in the June quarter tracks utility and data-center power demand, while Wind at $17.8M runs on a lumpy, concentrated order cycle with no data-center linkage. A sequential Wind decline would show up as a revenue miss unrelated to the grid narrative.
• Comtrafo (Brazil) purchase-accounting charges, indicated by management at roughly $1.5M non-cash in the June quarter, add a reported-profitability drag that is independent of demand and tapers on its own schedule.