KILLED · untestable core
Hypothesis
Institutions ("smart money") leave a footprint — order blocks, liquidity pools resting above/below swing highs and lows, and fair value gaps (FVG). Price returns to these zones to fill imbalance and grab liquidity; trading the zones front-runs institutional flow.
Math — the falsifiable kernel
The only objectively-codable pieces. A 3-candle fair value gap (bullish):
$$ \text{FVG} \iff \text{high}_{t-1} < \text{low}_{t+1}\quad(\text{price imbalance between } t-1,\,t+1) $$
A liquidity sweep / stop-run = wick beyond a prior swing extreme, close back inside:
$$ \text{sweep} \iff \big(\text{low}_t < \min_{k \text{swingLow}\big) $$
Method
Mechanized the falsifiable rules — liquidity sweep of a prior swing, then entry on return into the nearest FVG — and tested forward returns across 50 symbols with full costs, removing the human who normally draws the zones after the fact.
Results
| Stop-runs / sweeps are real | yes — they happen |
| FVG "fill" rate | ~90% — but price fills most gaps anyway |
| Mechanical sweep→FVG entry, net edge | ≈ 0 |
| "Order block" selection | unfalsifiable (drawn post-hoc) |
The testable kernel is real but unprofitable: liquidity sweeps (stop-runs) genuinely occur, and FVGs genuinely fill — but price fills almost any gap eventually, so the "fill" predicts nothing, and the mechanized sweep→entry has no net edge after cost. The rest of SMC — choosing which order block, which liquidity pool — is drawn after the move and cannot be falsified (same trap as Elliott Wave, N-093). A real phenomenon wrapped in an untestable presentation. Killed as a system.
SMC/ICT names something real (stop-hunts exist) and surrounds it with after-the-fact chart-drawing. The part you can test does not pay; the part that "works" cannot be tested. That is the signature of a narrative, not an edge.