Phase Space AI

Trade Construction

Turning Point Brands [TPB]

Turning Point Brands [TPB] — Trade Construction

The memo issues no position verdict.

Vehicle: EQUITY

Equity is the default and needs no argument. No options structure is proposed, and there is a specific reason beyond the default: the dominant catalyst on this name — an FDA PMTA determination — has no date. MEMO_BRIEF.md requires that "anything shorter than the declared thesis horizon requires a catalyst preceding expiry." An undated binary cannot be placed before an expiry, so no dated option structure can be constructed against it, at any strike. That is a structural argument against optionality here, not an omission.

Trailing 252-day realised vol 54.2% (screen record). IV not measured, so no LEAP argument is made — an unmeasured IV/RV spread is not an argument.

Invalidation — quantified, checkable quarterly, from disclosure that already exists

  1. Stoker's segment operating income. Q1'25 $24.1m → Q1'26 $19.8m on +48.1% revenue. If Q2'26 again shows Stoker's operating income falling on rising revenue, the growth engine's incremental economics are structural, not a launch phase, and the thesis is dead.
  2. Stoker's "other segment items" as % of segment sales. 13.6% (FY23) → 15.9% → 20.8% → 31.5% (Q1'26). Above 30% for two consecutive quarters invalidates the 23.0% terminal S&M assumption and with it the 18.0% terminal margin.
  3. Zig-Zag revenue. −7.2% FY2025, −22.4% in Q1'26. Below −15% for a full year means the legacy asset is in terminal decline rather than being harvested.
  4. Inventory. +24.1% YoY against revenue +16.8%. If inventory growth stays above revenue growth for two more quarters without the modern-oral manufacturing lines coming online, the pre-launch reading is dead and it is a channel problem.
  5. Any FDA action of any kind, on any TPB product line — instant, full re-underwrite.

Sizing constraint

size_bucket: small, $1.47bn market cap, 54.2% realised vol, and a binary undated regulatory risk with no hedge available. Size constrains sizing, never admission — but the binding inputs here are evidence_grade B−, a FAIL on the implied-path test at −0.8pp, and the fact that the downside case is a −54.3% permanent impairment with no observable warning. Per valuation.md rule 5, that reduces size and does not touch the operating assumption.

There is also a $200.0m unused ATM facility against a $1,474.5m market cap — 13.6% of potential undated dilution, after the company already sold $100.0m of stock at $98.59 versus a $75.71 spot.