Skip to content
module · research·lab notebook · falsifiable·we retract in public

Hypotheses, with falsifiers attached.

Short, dated notes from the desk. Each one comes with the specific observation that would force us to walk it back. Some get shipped into tools and playbooks. Some get retracted — and we keep those visible too.

notebook · status
7
notes total
3
shipped
2
active
1
retracted
$ tail -f /research/log
// methodology · how a note moves through the pipe

Four states. No quiet deletes.

Every research note starts as a draft and ends in one of three places: shipped into a tool or chapter, kept active with a live falsifier, or retracted in public.

  1. step 01

    Pose a hypothesis

    Frame the edge as a specific, dated claim. “X happens Y% of the time when Z.” Vague claims don’t make it onto the notebook.

  2. step 02

    Attach a falsifier

    Define the observation that would force us to walk it back. If we can’t name one, the note is rejected on intake.

  3. step 03

    Walk-forward, in public

    Promote a draft to active once the in-sample work is done. Active notes run live for a defined window with the falsifier on the wall.

  4. step 04

    Ship, hold, or retract

    Notes that survive get shipped into a tool, playbook, or chapter. Notes that fail get marked retracted and stay visible — we don’t quietly delete.

// index · all notes · status × track

The whole notebook. Newest first.

Filter by status or track. Retracted notes stay in the index — we don't quietly delete what didn't hold up.

status
track
activerisk·rn-007

Regime-aware sizing beats fixed-fractional in walk-forward

A simple regime classifier (VIX bucket × value-area context) feeding a multiplier into fixed-fractional sizing beat the unmodified 1% rule on Sharpe across an 18-month walk-forward, primarily by under-sizing during chop.

falsifier

If forward Sharpe improvement compresses to < 0.10 over 60 sessions, the multiplier gets shipped as opt-in only, not default.

draftexecution·rn-006

VWAP+1σ band as overshoot filter for mean-reversion entries

Adding a 1-sigma VWAP band as a hard filter on counter-trend entries cut the sample size substantially and lifted average expectancy across two ticker baskets. Tradeoff: fewer trades, fatter average outcome.

falsifier

If forward sample expectancy regresses to within 0.05R of the unfiltered version over 200 trades, the filter is dropped.

activemicrostructure·rn-004

TTM Squeeze release direction tends to follow the higher-timeframe trend

On 5-minute charts, when the daily trend is up and a TTM Squeeze release occurs, the resolution has more often printed upward than downward over a 90-day rolling window. The asymmetry collapses near macro events and inside the 60 minutes pre-FOMC.

falsifier

If the alignment falls below 60% for 30 consecutive sessions outside event windows, demote to confirmation-only signal.

shippedrisk·rn-003

Daily loss caps reduce week-over-week variance more than position sizing tweaks

Replaying 2024–2025 trades with a hard 3% daily loss cap and an unchanged position-sizing model cut weekly P&L variance more than halving size with no cap. Caps win on tail control; sizing wins on average return.

falsifier

If a regime appears where the cap clips materially profitable revenge-trade recoveries (top decile day > +6R after a -3% morning), revisit.

shippedexecution·rn-002

Opening-range breakout R:R degrades after 09:55 ET

In our sample, opening-range breakouts entered early in the session (09:30–09:50 ET) reached a 2R target noticeably more often than breakouts entered after 09:55 ET, even when the triggers looked mechanically similar.

falsifier

If the post-09:55 cohort recovers to ≥ 30% 2R hit-rate over 60 sessions, the cutoff is widened or removed.

shippedmicrostructure·rn-001

Volume Profile POC reverts in balanced regimes; trends in imbalanced

Across two years of intraday ES data, retests of the developing POC tended to mean-revert when VIX was low and the prior session closed inside its value area, and tended to trend-continue when VIX was elevated and price closed outside value.

falsifier

If 12 weeks of forward data show < 55% directional bias in either regime bucket, the heuristic gets demoted to discretionary input only.

retractedregime·rn-009

Earnings drift no longer pays in single-name large caps for our cohort

Post-earnings drift trades on large-cap names underperformed expectations by a wide margin in the 2024–2025 sample. We retracted this as a primary edge in member playbooks; remaining mentions are explicitly historical.

falsifier

Re-instate as an active edge only if 2 consecutive 6-month windows show ≥ 0.30R median drift on filtered candidates.

·revised Sep 30, 2025
// open questions · what we don’t know yet

What we’re still not sure about.

A short list of questions the desk hasn't answered. If you have data, opinions, or a way to falsify these faster, we'd like to hear about it.

  • open · q01

    Does TTM-Squeeze direction-alignment hold outside event windows?

    Current sample is 90 days. We want 240 sessions and explicit pre-FOMC / pre-CPI carve-outs before promoting to a primary signal.

  • open · q02

    Where does VWAP+1σ filtering break down?

    The lift looks real on two ticker baskets. The honest answer is we haven’t stress-tested it on low-liquidity names where VWAP itself is noisy.

  • open · q03

    Latency → behavior — is the threshold platform-dependent?

    We saw the 250ms inflection on the desk web client. Mobile users may have a different sweet spot. Open.

Get the daily read — free

The market read before the trade, and the results after — win or lose. No hype.

Join the free Discord