Active
Shadow strategy experiments.
Every trading decision runs a second AI pass using an alternate strategy. The shadow decision is logged but never executed. This page compares what the shadow would have done against the primary.
Experiment overview.
3 pending
Shadow agrees with primary
Running since Mar 18, 2026. Latest shadow decision: 3 minutes ago.
Per-strategy results.
How each shadow strategy performed compared to the primary.
Breakout
1 decisions
Agreement
100.0%
Symbols
1
Evaluated
0
Support/Resistance Flip
1 decisions
Agreement
100.0%
Symbols
1
Evaluated
0
Mean Reversion
1 decisions
Agreement
100.0%
Symbols
1
Evaluated
0
Per-symbol breakdown.
Shadow experiment results for each coin in the universe.
| Symbol | Decisions | Evaluated | Agreement | Shadow Win Rate |
|---|---|---|---|---|
| BNB | 1 | 0 | 100.0% | Pending |
| ADA | 1 | 0 | 100.0% | Pending |
| BCH | 1 | 0 | 100.0% | Pending |
Recent shadow decisions.
The latest shadow decisions and how they compare to the primary call.
BCH
3 minutes ago
HOLD
vs
HOLD
Agree
ADA
36 minutes ago
HOLD
vs
HOLD
Agree
BNB
about 1 hour ago
HOLD
vs
HOLD
Agree
How shadow testing works.
Every trade decision runs the AI twice: once with the primary strategy, and once with a shadow strategy. Only the primary is executed. The shadow is scored against actual price movement.
Two decisions, one trade.
The shadow strategy runs in a background thread at the same time as the primary. It sees the same market data, indicators, and risk parameters — only the strategy lens is different.
Shadow never trades.
The shadow decision is stored alongside the primary but never submitted to the broker. It exists purely for comparison and learning.
Evaluated against price.
After 4 hours, the shadow decision is scored against actual price movement. A BUY that preceded a price increase scores positively. The evaluation is automatic and objective.
Keep following the experiment
The point is to make the lab inspectable.
You should be able to watch the portfolio, see how and why decisions are made, learn from it, and understand when I got something wrong.
If you spot wrong data or want to suggest an improvement, send feedback here.
If having your own personal lab sounds useful, leave your email on the waitlist. If enough people are interested, it is worth building.
This is a public paper portfolio. Past performance does not guarantee future results. Hugo's Trading Lab publishes rules-based analysis with AI-assisted decision summaries for informational purposes only, not financial advice.