Paper trading only. No real money and no broker connection. Results assume your fee and slippage settings.
Chart
Open trades
| Opened | Instrument | Side | Setup | Entry | Stop | Target | Risk |
|---|
Decision log
| Bar | Instrument | Regime | Setup | Decision | Reason |
|---|
Profitable trades
Only closed trades that made money after costs.
| Closed | Source | Instrument | Side | Setup | Entry | Exit | Costs | Net | R |
|---|
Flash desk so far
All closed trades, winners and losers
| Closed | Source | Instrument | Side | Setup | Exit | Net | R | Lesson |
|---|
Market scan
Loading scan results.
These are backtests on past prices, not forecasts or live trades. Testing many markets makes some look good by chance, so a market must pass every test below before it counts. Perpetual results leave out funding payments.
Long-history strategy tests
Loading.
Rules were fixed before the results were seen. A market passes with a year or more of data, at least 10 trades, a profit in both halves, a profit at double cost and a t-statistic of at least 3. The spot trend test must also beat buy and hold.
| Strategy | Market | Days | Trades | Total | Per year | Max drop | Sharpe | t-stat | At double cost | Tests |
|---|
All markets scanned
| Market | Venue | 24h volume | Trades | Won | Net R | Profit factor | First half R | Second half R | Net R at double cost | Tests | Latest bar |
|---|
A market passes with at least 20 trades, a profit factor of 1.3 or more, a profit in both halves of the period, a profit at double the assumed cost, and a t-statistic of at least 2. R is one unit of the amount risked per trade.
Results by setup
| Setup | Side | Venue | Trades | Won | Net R | Avg R |
|---|
How it runs
Every 5 minutes the model works through these engines for each instrument, on 15-minute candles from Coinbase's public feed. A new trade needs all four hard checks and at least two of the three soft checks.
- ScanTrend, volatility, 48-bar high and low, volume.
- SignalLiquidity sweep, breakout, engulfing candle or pin bar at a key level.
- PlanEntry at the next bar's open, stop past the level, target at 2.5 times the risk.
- RiskPosition size, costs, open-trade limit, daily loss limit.
- MemoryWaits if at least 75% of 4 or more similar past trades lost.
- MonitorCloses at stop or target. If both fall in one bar, it assumes the stop.
- ReviewSaves the full trade record and a lesson.
Gates before real money
The first four are measured on forward paper trades only. Backtests don't count.
What the trades show
Grouped by setup and market regime. Fewer than 30 trades in a group is too few to rely on.
| Setup | Regime | Trades | Won | Avg R | Net | Evidence |
|---|