BL-SD-048: Buying Volatility When Conditioned

Status: RELEASED

Classification: PUBLIC

Revision: 1

Date: 2026-09-08

Venue: Deribit options, priced against Bybit-derived conditioning signal Programme: II Verdict: FALSIFIED Axis of death: Idea Transferability: UNIVERSAL

1. Identification

2. Economic thesis

If selling volatility is out of reach on capital grounds, as BL-SD-047 established, the inverse route is to buy it selectively: use the fragility predictor to identify when implied volatility is priced too low relative to the risk actually building in the market, and pay the insurance premium only then.

Who is forced to lose, if the mechanism works as stated. Volatility sellers who are pricing options without the fragility information this programme has already validated as a real, persistent cross-symbol signal. If that signal detects risk the option market has not yet priced, the seller on the other side of the trade is underpricing insurance they will regret having sold.

3. Falsifiable claim

Pre-registered as a direct test of the thesis above:

The volatility premium is lower, not higher, in the state the fragility predictor flags as high risk, because the market has not yet incorporated that information into implied volatility.

4. Mathematical specification

Notation per BL-D-011.
 
Fragility signal, per BL-P2-003 information map, resolved at t from data
observable at or before t. Quintile assignment:
 
  q_t = quintile_rank( fragility_t )   in {1, .., 5}, 5 = most fragile
 
Premium differential, high versus low fragility quintile:
 
  Delta_VRP = mean( VRP_t | q_t = 5 ) - mean( VRP_t | q_t = 1 )
 
Entry: none simulated. This is a characterisation of the pricing differential
itself, not a live entry rule; the thesis is falsified or confirmed on the
sign and significance of Delta_VRP directly.
 
Sizing: not applicable, no position simulated.
Cost:   not applicable, no round trip evaluated; the finding closes the door
        before a cost model is needed.

Causality holds: q_t is assigned from the fragility signal at t, which is itself built only from information at or before t, per the information-map instrument this signal shares with the risk-gate work in BL-SD-049.

5. Operating envelope

6. Instrument

The fragility predictor and its information map are the shared instrument behind this row and the production risk gate in BL-SD-049. No new instrument was built for this door; the measurement is a quintile split of the existing variance risk premium series against the existing signal.

7. Results

Both instruments move in the same direction, and it is the direction the thesis predicts should not occur. The premium is higher, not lower, exactly when the fragility signal flags elevated risk.

8. Verdict, axis of death, and transferability

Verdict: FALSIFIED. Axis: idea. Transferability: UNIVERSAL.

The sign is wrong on both instruments tested. The thesis required the option market to have missed information this programme's signal carries; instead, the option market is already pricing that information, and pricing it in the correct direction, before this signal is applied to it. No cost floor, no capital constraint, and no execution channel changes a sign that runs the wrong way at the source.

This is the more informative of the two volatility findings in this programme, because it says something about the conditioning signal rather than about the operator. The fragility predictor is real and it persists out of sample; what this measurement shows is that its information is already reflected in option prices by the time it fires, which is a property of market efficiency rather than of this operator's access or capital.

9. Cost and capacity analysis

Not applicable. The mechanism fails on sign before any cost or capacity question arises.

10. Envelope sensitivity and reopening conditions

No binding parameter exists to name, because the finding is a sign reversal, not a magnitude shortfall.

Reversal condition:  none computable.

Applying the BL-D-002 section 4 test directly: at any hypothetical cost, including zero, the direction of the effect still runs opposite to the thesis. There is no envelope substitution that turns a mispriced sign into a correctly priced one; the market, not the operator, holds the correct sign already.

What would reopen it: evidence that the fragility signal leads option market pricing at some other horizon not tested here, which would be a new hypothesis on a horizon this measurement did not examine, not a reopening of this one.

What would not reopen it: any change to fees, capital, or venue access. None of them touch the sign.

11. Reference implementation

Shares its instrument with BL-SD-049; see that dossier's reference implementation note. No separate code path was built for this door.

12. Evidence register

The quintile decomposition itself is reported in BL-P2-001 section 7 and in the kill ledger headline for this row; it is not broken out as a separate field inside E-0007, and this dossier states that plainly rather than implying a field that does not exist.

13. Lessons extracted

A conditioning signal that is real does not automatically produce a tradeable directional bet on the instrument it might seem to point toward. The fragility predictor validated elsewhere in this record persists out of sample and calibrates honestly as a defensive gate; here, tested as an offensive signal against a market that prices the same information, it fails cleanly. The lesson generalises beyond volatility: a signal's value depends on whether the counterparty market has already priced what the signal detects, and that has to be tested separately from whether the signal itself is real.

14. Cross-references

BL-P2-001 section 7, the door 5b narrative. BL-SD-047 for door 5a, the harvesting variant of this same premium. BL-SD-049 for the same fragility signal used defensively rather than offensively. Ledger row D5b.

Copyright (c) 2026 Bijective Labs. All rights reserved.