Apex research
32 tested relationships between markets, macro and capital — each with its sample, its result, its multiple-testing status and what would invalidate it. 21 of them have fallen and are published anyway: the record of what did not work is part of the product.
CONFIRMED Confirmed — passes the validation tests
- Gold against the real rate
−0.95% per 10 bp (p=0.001)
- The deeper the equity fall, the tighter bitcoin follows stocks
monotonic dose-response
- Brent has decoupled from everything
shift holds at 180 days (p=0.0016)
- SOL has its own memecoin link
+0.16 overweight vs −0.04 for ETH (n=729 days, p=0.0003)
IN SHADOW In shadow — evaluated daily, no capital
- Bitcoin after a Fed rate hike
35 % up at +5d/+20d vs 56–58 % baseline (in-sample, n=20)
- BRENT on a 24-hour horizon
63.3% at 24h
- Flight to safety signals a turn
67.6% vs 55.6%, median +0.83% (n=148, p=0.004)
- Bottom-fishing XRP
59.4% vs 48.7%, median +0.78% (p=0.004)
- BTC beta through altcoins
54% over break-even, +138 units
DISCOVERED Discovered — found in a sweep, not yet validated
- Bitcoin's gold link at a five-year high — and it is bitcoin that moved
BTC~gold +0.65 vs Nasdaq's +0.22 — gap +0.43, 96.6th percentile (p=0.011)
- Bottom-fishing BTC
60.6% vs 50.1%, median +0.52% (p=0.015)
FALLEN Fallen — broke its own criteria; kept on purpose
- BTC up next day under three simultaneous conditions
66% vs 50%, median +0.9% (n=85, p=0.001) · engine backtest 6y: +696, max DD −92, 4/4 years
- BTC moves as an entry for SOL/XRP/ETH
fell 14 Sep: simultaneity, not lead
- The dollar pattern against Brent
fell 13 Sep: does not hold outside the window it was found in
- Lead time between markets
0 of 28 daily pairs show lead time
- The debasement thesis
BTC vs real rate −0.28 (120 days)
- The GDI score as a predictor
+12.7 pp raw; p = 0.49 controlled
- Combinations of logged market state
0 of 253 survive
- Rotation as a 2–3 day directional warning
0 of 160 tests survive
- Doing the opposite of the signals
−297 actual against −1 767 inverted
- Bottom-fishing Brent
59.1 % vs 53.2 % (p = 0.107)
- Bottom- and top-fishing natural gas
best cell 53.7 % vs 50.7 % baseline (p = 0.38)
- Top-fishing: selling extremes
0 of 54 top signals survive
- Capital flow as a warning for the Nasdaq
1 test with p < 0.05 where chance alone gives 4.2
- Memecoin volume as a warning for Solana
53.2 / 45.2 / 51.5 % vs 48.4 % baseline
- Memecoin activity as a regime filter for BTC beta
SOL~BTC 0.65 to 0.88 across regimes (p = 0.0003)
- Intraday signals on SOL, XRP and WLD (hourly data, two years)
117 with p < 0.05 where chance gives 18 — and nothing tradable
- Confidence as a predictor
AUC 0.498 (p = 0.94), n = 465
- LONG/SHORT asymmetry
+2.1 pp; fails the half-split
- News sentiment as a directional signal
54 % after de-duplication
- Weekend as a filter
fails the half-split
- Breadth as prediction
coin toss; half-split flips
Engine rules (thresholds, penalties, windows) are not published here — they are the mechanism, not the evidence. Methodology · Learn