Phase Space AI

Trade Construction

Coherent [COHR]

Coherent Corp [COHR] — Trade Construction

The memo issues no position verdict. This section states how a position would be constructed if the book chose to take one, and what would invalidate it.

Vehicle: EQUITY

Ladder rung Vehicle Argument required Argument given
1 Equity none — default, adopted
2 LEAP IV − RV, plus why convexity justifies the spread not offered — see below
3 Sub-horizon option catalyst before expiry not offered
4 Multi-leg constructability at the strike not offered

Required disclosure
Trailing 252-day realised volatility 80.5%
ATM implied volatility, 12–18m calls 93.0% – 94.5%
IV − RV +12.5 to +14.0 points
Quoted size at the strike intended $220C 2027-12-17: 28 × 32, bid 93.74 / ask 99.82, OI 18
Next expiry out $220C 2028-01-21: 149 × 147, OI 37

Why not the LEAP, explicitly

  1. The premium is measured and large. 12.5–14.0 points of implied over realised, held for 17 months. Against a 12-month target of +21.9%, the option must clear both the premium and the time decay before the equity's return is matched.
  2. The chain fails the strike-depth test. 18 contracts of open interest at the December-2027 $220 strike, and a 6.08-point bid-ask spread — 6.3% of premium — on entry alone. The framework's record already names HCA at 18 contracts and CRDO at 78 as uninvestable; this is the same magnitude.
  3. Horizon conflict. The liquid expiry (January 2028, 149×147) is 18 months out; the thesis horizon that the 12-month target addresses is 12 months. Where no expiry is both liquid and correctly dated, the answer is equity.
  4. A LEAP is embedded leverage in an unlevered book, and COHR already carries 1.7x corporate leverage that inverse-volatility sizing does not see.

Position construction

Parameter Value Reasoning
Instrument COHR common
Entry Scale, not single-tick Spot $222.19 is −8.6% below the $243.18 tick the screen used one session earlier. A name moving 8.6% in a day at 80.5% realised volatility should not be entered in one clip
Entry zone $200–225 The lower bound is roughly one 20-day ATR below spot; below $200 the 12-month target's upside becomes +35% and the case improves materially
Sizing input 80.5% realised vol → inverse-vol weight is low The book's inverse-volatility rule sizes this down automatically. Note the framework's own HALO finding: inverse-vol sizing misprices a fat left tail. The −77% bear case here is fatter than 80.5% vol implies, so the vol-derived size should be treated as a ceiling, not a target
Single-name cap 20% book (framework limit) Not remotely approached at this evidence grade
Correlation constraint This is the binding constraint, not the single-name one COHR, FN, LITE, AAOI, CRDO and CIEN are six exposures to one driver: hyperscaler optical-interconnect capex. The book must size the cluster, not the name. Two of these six names share a top customer (NVIDIA is 27.6% of FN revenue and a $2.0bn preferred holder in LITE)

Exit and invalidation

Thesis horizon: 12 months (to the FY2027 Q4 print), with the 5-year implied path as the ownership test rather than the trading horizon.

Trigger Level Action
Invalidation — mechanism FY2027 guide below +21.4%, i.e. FY2027 growth does not exceed FY2026 Thesis is directly contradicted by the company. Exit
Invalidation — accounting DIO rises above ~180d while YoY revenue growth falls below 10% The benign inventory explanation fails. Exit
Invalidation — concentration FY2026 10-K shows top-two customer share above ~30%, or a named customer disappears from the note Re-underwrite before adding
Invalidation — margin Non-GAAP gross margin prints below the 39% guided floor for two consecutive quarters The 6-inch InP mechanism is not working. Exit
Target $271 (+21.9%) Trim, do not exit — the 5-year path is the ownership question
Valuation stop EV/EBIT above ~85x, or price/TTM-revenue above the 5-year high The multiple is already 91st percentile; a further re-rate is the risk, not the reward
Drawdown ladder Per the book protocol; nothing name-specific overrides it

What would move this from FAIL to PASS

Named, so it is falsifiable:

  1. A quantified design-win or backlog disclosure. COHR states record bookings and orders placed "over a year" forward but publishes no number. A disclosed multi-year 1.6T backlog would convert the 29.18% required CAGR from narrative-supported to evidence-supported.
  2. A price decline. At $175 the required CAGR falls to roughly the demonstrated rate and the Valuation Criteria passes on the base 15.0x exit multiple, with no change to any assumption.
  3. A sustained gross margin above 41%. That lifts the terminal-margin bridge above 20.0% and cuts the required growth directly.