Praxis Precision Medicines (PRAX) is a $9.7B CNS biotech stacking four binary catalysts into the next 12 months — two NDAs in FDA review, one Phase 3 readout in seven weeks, and one broad-label expansion readout by Q4 2026. The Q1 2026 earnings call (May 7) confirmed and accelerated the thesis. The stock, at 96% of its 52-week range after a +829% twelve-month run, has caught up. The volatility hasn't.

What the filing says

The call moved four things.

EMERALD enrollment is complete. The broad-DEE Phase 3 study for relutrigine — the program's path from the ≈10,000-patient initial SCN2A/SCN8A label to a ≈200,000-patient broad-DEE label — finished enrolling "in record time." Top-line is now guided to Q4 2026, immediately after the September 27 relutrigine PDUFA. Management did not guide to this pace last quarter.

FDA mid-cycle communications described as constructive on both NDAs (relutrigine PDUFA Sept 27, 2026; ulixacaltamide PDUFA Jan 29, 2027). The CEO used "cautiously optimistic" and "no major concerns" — language inside disclosure norms but materially more confident than the prior call.

Ulixacaltamide manufacturing is materially de-risked. Dual independent drug-substance manufacturers; "metric tons" in supply chain; FDA process alignment completed pre-submission. The primary identified CRL risk for a first-time commercial company was CMC, and this disclosure substantially reduces it.

POWER3 is new. A Phase 3 monotherapy conversion study for vormatrigine in broad epilepsy was announced — never mentioned in any prior call. Management framed it as a competitive moat ("competitors cannot"); the framing is overstated (XENE's same-day call disclosed contemplating Kv7 monotherapy; cenobamate already has a dual label) but the 2-3 year first-mover timing on Nav1.2/1.6 monotherapy is real.

EMBOLD's relutrigine Phase 3 effect — 53% placebo-adjusted seizure reduction — was confirmed by an analyst and not corrected, with SCN2A and SCN8A subgroups responding "quite similarly." PRV eligibility was flagged explicitly for the first time. Cash $1.4B, $86M Q1 operating burn, funded to 2028+.

What management avoided: zero specifics on EMBRAVE Part B, the elsunersen ASO pivotal study. CEO said "recruiting really well" and moved on. Silence on a 30-patient first-in-class registrational design is notable.

What the market thinks

Stock at $346, up 829% trailing 12 months, 96% of 52-week range. Mean analyst target ≈$651, 14 of 17 firms Buy. At-the-money implied volatility sits at the 6th percentile with POWER1 readout seven weeks out.

A rough sum-of-parts using consensus-style peak-sales assumptions — ulix $2B, relutrigine narrow $400M, vormatrigine $1.5B, EMERALD broad-label expansion $1.5B incremental, elsunersen $200M — at 5x revenue, NPV-discounted, probability-weighted at sell-side-style approval/success rates, lands ≈$11B. Current $9.7B implies a ≈25% discount for execution risk, competition, and momentum reversion. Roughly fair on direction.

The gap is in the structure. Six-percentile IV ahead of four sequential binary events on a single name does not match the realized volatility profile of a stacked-catalyst biotech.

Probability framing by driver

Caveat: the "implied" column is not options-derived per-event probability — PRAX does not have liquid binary-isolating option markets on each readout. The POWER1 figure is inferred from term-structure-level IV percentile (low IV → market pricing high-confidence path), and the others are reverse-engineered from analyst sum-of-parts. Treat the deltas as directional, not precise.

The honest reading: on POWER1 specifically, our confidence is below where the IV says the market is sitting. Unmodeled positives are small — two RPDD assets imply a PRV stack with ≈$175M EV (about 1.8% of cap), and EMERALD's record-pace enrollment is a positive selection signal not in consensus models.

Why the gap exists

The directional bull case is consensus. Volume of confirming evidence is the market catching up, not novel mispricing. What is unmodeled is structural: (a) four binaries priced linearly via low IV rather than as a conditional lattice; (b) the PRV stack and EMERALD calendar pull-forward are filing-derived items not yet in sell-side models; (c) the vormatrigine moat is timing-based (Nav1.2/1.6 monotherapy first-mover by 2-3 years) rather than mechanism-exclusive as management framed.

Forward EV across an eight-state scenario tree: ~+25% over 12 months. Implied Sharpe ≈0.4 as common stock. Not a fat pitch as equity. The asymmetry lives in the options.

Risks (ranked)

1. POWER1 miss in Q2. Largest near-term downside; stock likely -30% to -40% in days. Our 62% probability is below market's ≈70% — if we are right on the probability, the position direction is the wrong side of it.
2. Ulix CRL on labeling or post-marketing requirements, even with manufacturing de-risked. Residual ≈17% probability.
3. Commercial execution risk. First-time commercial company building two simultaneous launches in 2026-2027.
4. EMBRAVE Part B failure or further silence. Eliminates elsunersen optionality; the continued silence is itself a soft bearish tell.
5. Sector beta and momentum reversion. XBI roll-over with PRAX at 96% of range would compress idio variance below the 75% target.

Catalysts

• POWER1 readout — by 2026-06-30 (≈7 weeks)
• R&D Day / Commercial Day — guided Q2 2026 on prior call, conspicuously absent from Q1 call
• Relutrigine PDUFA — September 27, 2026
• EMERALD top-line — Q4 2026
• Ulixacaltamide PDUFA — January 29, 2027

What would change our mind

• POWER1 miss with rest of stack intact and stock at $200-250: bull-case entry at 200-day MA.
• EMBRAVE Part B specifics disclosed in either direction: resolves the silence.
• FDA AdCom announced for either NDA: meaningfully bears on approval probability.
• XBI -15% with PRAX outperforming: confirms idio variance is intact.
• Big-pharma rumor or strategic-process disclosure: compresses the catalyst-cascade EV into a single date.

We are not committing to a position. The directional case is roughly fairly priced. The structural mispricing — low IV against a four-binary lattice — is where the asymmetric trade lives, and capital deployment should respect that.

Evidence

Memo LR: 1.3

Direction: mildly bullish. The call confirms a thesis the market is already long, with two modest unmodeled positives (PRV stack, EMERALD timing) and one real contradiction (monotherapy moat). The structural mispricing in IV is where the asymmetric edge lives but does not move the directional prior much.