Strategies
Build a trading strategy from blocks, the way Composer’s symphonies work, then test it on history with the checks most backtests leave out: no decision reads a price it could not have seen, every try counts against the result, and what ARCANE measures about the world (shipping, sanctions, prediction markets, the Pacific) can be part of the rule.
Presets computed Sep 26, 2026 with prices to Sep 25, 2026. Prices come from Yahoo Finance (unofficial chart API): Unofficial and not licensed for redistribution. A licensed vendor (Alpaca, Tiingo or Polygon) is a founder decision before the builder is sold.
Hypothetical backtest; not investment advice; paper trading only.Past results, real or simulated, do not predict future ones. ARCANE places no live orders from this page.
Build one
Start from a blank symphony or a preset: weight blocks, If/Else conditions, filters that keep the best few, groups. Backtest it, change a number, backtest again; the page counts every try.
Preset analyses
Each written down, with a sealed holdout, before it was testedNone of these passes ARCANE’s house backtest gate, and none beats simply holding its benchmark by more than luck explains. They are here as worked examples of the method, and as starting points.
Geopolitics
Hormuz disruption tilt
About a fifth of the world’s oil passes the Strait of Hormuz. When the ships counted through it fall well below normal, supply risk is real, not a headline, and energy producers and gold have tended to cushion that kind of shock. The counts arrive weekly and few investors read them.
- A year
- +13.5%
- Sharpe
- 0.73
- Held out
- 0.79
- Overfitting: Not enough to judge
- House gate: could not run
Risk-off on sanctions escalation
Bursts of sanctions activity mark geopolitical escalation: new programmes, new designations, new export controls. When OFAC publishes far more than usual, step out of stocks into Treasuries and gold until the burst passes.
- A year
- +7.6%
- Sharpe
- 0.50
- Held out
- 0.42
- Overfitting: Likely overfit
- House gate: not passed
Macro and prediction markets
Fed-cut rotation
Kalshi’s traders put a price on what the Fed does at its next meeting. When they make a cut likely, long Treasuries and growth stocks stand to gain most from lower rates; when they make a hike likely, short Treasuries and banks hold up better. The market for the meeting often moves before the rest of the market has finished adjusting.
- A year
- +15.5%
- Sharpe
- 1.08
- Held out
- 1.08
- Overfitting: Likely overfit
- House gate: not passed
Climate
El Niño farm tilt
A strong El Niño shifts rain away from India, Southeast Asia and Australia, and harvests there tend to suffer. When NOAA’s index says the Pacific is strongly warm, put half the portfolio in farm commodities while it lasts.
- A year
- +9.7%
- Sharpe
- 0.58
- Held out
- 0.71
- Overfitting: Holds up
- House gate: not passed
Classic rules
Trend filter: the 200-day line
The oldest rule in the book: own stocks while they trade above their 200-day average, and step into Treasuries when they fall below it. It gives up some upside in exchange for sidestepping most long bear markets.
- A year
- +8.2%
- Sharpe
- 0.71
- Held out
- 0.60
- Overfitting: Likely overfit
- House gate: not passed
Dual momentum
Own whichever of US or international stocks has done better over the past year, but only if stocks as a whole beat cash; otherwise hold bonds. Relative momentum picks the leader, absolute momentum keeps you out of long bear markets.
- A year
- +8.7%
- Sharpe
- 0.61
- Held out
- 0.65
- Overfitting: Likely overfit
- House gate: not passed
Overbought, oversold
The pattern behind many Composer community strategies, without the leverage: step aside when the Nasdaq-100 is very overbought, buy it when it has been sold hard, and otherwise follow the trend.
- A year
- +13.4%
- Sharpe
- 0.81
- Held out
- 0.79
- Overfitting: Likely overfit
- House gate: not passed
How ARCANE tests a strategy
- No peeking. A decision reads closing prices and ARCANE data only up to that day, and trades at the next day’s close. Composer’s convention, deciding and trading on the same close, is there to compare with and marked optimistic.
- Costs. Slippage on every trade (5 basis points unless you change it) and US regulatory fees on sales. Cash earns nothing.
- Held out. The last part of the history is reported on its own. It stays out of sample only until the symphony is changed after seeing it.
- Every try counts. Each backtest of an idea raises the bar the next one must clear (the deflated Sharpe ratio), and each result is checked against its own numbers nudged either way (the probability of backtest overfitting).
- Written down first. A strategy enters ARCANE’s house book only with a pre-registration: its claim, why the edge should exist, who pays for it, and what stops it.
- Paper only. The builder never places a live order.
What ARCANE adds to the blocks
Each value dated the day it became public| Data | For example | History |
|---|---|---|
| Chokepoints (56 series) | Panama transits vs normal: Ships through Panama a day, last 7 days, as a percent above (+) or below (−) normal. −50 is half of normal traffic. | revised history |
| Sanctions and export controls (8 series) | OFAC actions, last 7 days: Rules and notices from Treasury’s Office of Foreign Assets Control (sanctions) published in the Federal Register in the last 7 days. | point in time |
| Prediction markets (2 series) | Fed cut at the next meeting (Kalshi): The market’s probability, in percent, that the Federal Reserve cuts rates at its next meeting: Kalshi’s cut contracts added up, at the end of the previous day. | point in time |
| Climate (ENSO) (2 series) | Oceanic Niño Index: The latest three-month sea-temperature anomaly in the central Pacific, in °C. +0.5 and above is El Niño territory, −0.5 and below La Niña. | revised history |
| Earthquakes and storms (6 series) | M6+ earthquakes, last 7 days: Earthquakes of magnitude 6 or more worldwide in the last 7 days. | revised history |
| ARCANE Calls (2 series) | ARCANE Calls open: How many of ARCANE’s public, dated Calls are waiting to resolve. | point in time |
“Revised history” means the source corrects its past figures, so a backtest sees slightly better data than a trader had on the day. Nothing is filled in where a source has no history.