Regime awareness
See the market for what it is now, not what it was then. Context first. Always.
Long-horizon research / 01
Hold with
a thesis.
Conviction is not a feeling. It is a test.
Research for buy-and-hold investors.
Built on out-of-sample evidence.
Model evidenceBuy-and-hold / dip-buy test
IS→OOS
— IN-SAMPLE (IS)
Model development and selection.— OUT-OF-SAMPLE (OOS)
True test on unseen market history.
Model development and selection.— OUT-OF-SAMPLE (OOS)
True test on unseen market history.
What you get
Out-of-sample proof
Every idea is tested where it was not built — so you know what survives contact with reality.
Paper-only safety
No leverage. No derivatives. Just clean, executable ideas you can hold with confidence.
Your downside / 02
Know the pain before you hold.
Model a simple drawdown scenario against your own capital and loss budget. The result is arithmetic, not a market forecast.
Scenario only · fees, slippage, funding and gaps are not included
Research integrity / 03
Pretty backtests lie.
Overfitting, look-ahead bias, and curve-fitting turn noise into narratives. In-sample performance is not a license to risk capital — it is a warning.
If it can’t survive out-of-sample,
it shouldn’t survive your portfolio.
Reality check / held-out window
Selection ≠ proof● IN-SAMPLE Model selection● OUT-OF-SAMPLE Unseen history
The evidence, in plain sight.
Each row shows the configuration that looked best in-sample — and what that same configuration did on the held-out window.
| Horizon | Leverage | Win rate | In-sample | Out-of-sample | Verdict |
|---|---|---|---|---|---|
| 1d1d dip-buy -7% · ride to target / wide stop (position) | 3× | 32% | +$8639 | $-5450 | REGIME_DEPENDENT |
| 1d1d dip-buy -7% · ride to target / wide stop (position) | 3× | 37% | +$8920 | $-3278 | REGIME_DEPENDENT |
| 1d1d dip-buy -7% · ride to target / wide stop (position) | 3× | 38% | +$7530 | $-1975 | REGIME_DEPENDENT |
| 1h1h dip-buy -4% · ride ≤30d to target / wide stop | 3× | 33% | +$144 | $-404 | NET_NEGATIVE |
| 1h1h dip-buy -4% · ride ≤30d to target / wide stop | 2× | 29% | +$-130 | $-88 | NET_NEGATIVE |
| 5m5m dip-buy -2% · ride ≤7d to target / wide stop | 2× | 47% | +$-19 | $-33 | NET_NEGATIVE |
| 5m5m dip-buy -2% · ride ≤7d to target / wide stop | 3× | 62% | +$5 | $-25 | NET_NEGATIVE |
| 1h1h dip-buy -4% · ride ≤30d to target / wide stop | 2× | 38% | +$-70 | $-12 | NET_NEGATIVE |
| 5m5m dip-buy -2% · ride ≤7d to target / wide stop | 2× | 50% | +$-13 | $0 | REGIME_DEPENDENT |
How it’s modelled: WITH hard stop (operator spec). A stop caps the normal loss, but a gap/crash beyond the ~0.9/leverage liquidation buffer at 2-3× still wipes the margin — modelled. Realistic frictions: 5 bps/side + 2 bps/side slippage + perp funding over the hold. Dip-buy entry on the bar low, no look-ahead. Configs SELECTED on in-sample only; OOS = held-out last 30%, the honest judge. 5m history ~2 weeks (30% targets rarely hit intraday). Demo/paper experiments — live stays BLOCKED behind the proof gate.
Move from hope to evidence.
Build, test, and stress-test your buy-and-hold thesis.