Paper-trading metrics
Snapshot as of 2026-07-07 18:02 UTC — the last time the bot published data to this repo (see the freshness banner above for whether that is current). Empty: as of the snapshot no strategy had live-fired, and no further data was published before the bot was retired.
| metric | all-time | today | 7-day | 30-day |
|---|---|---|---|---|
| starting balance | $1,000.00 | — | — | — |
| equity | $1,000.00 | $1,000.00 | — | — |
| all-time P&L | $0.00 (0.00%) | $0.00 | — | — |
| trades | 0 | 0 | 0 | 0 |
| wins | 0 | 0 | 0 | 0 |
| losses | 0 | 0 | 0 | 0 |
| win rate | n/a | n/a | — | — |
| sharpe | n/a | n/a | — | — |
| biggest winner | — | — | — | — |
| biggest loser | — | — | — | — |
| avg hold time | — | — | — | — |
| orders refused by safety | 0 | 0 | 0 | 0 |
| kill-switch trips | 0 | 0 | 0 | 0 |
Per-strategy attribution
| strategy | orders placed | orders refused | win rate | avg P&L | sharpe |
|---|---|---|---|---|---|
| no orders yet | — | — | — | — | — |
What these numbers would have told us
Had paper trading started, the numbers that would have mattered most:
- win:loss ratio, not win rate alone. A 60% win-rate strategy with a 0.4:1 win:loss ratio is a losing strategy.
- payoff shape: how asymmetric are the wins vs losses. Asymmetric payoffs (rare big wins + frequent small losses) beat symmetric payoffs.
- drawdown, not P&L. The worst peak-to-trough while in the position is what scares real money out. If the worst drawdown is 4% even with a 30% annual return, that’s a reasonable tradeoff.
- signal-fired-to-fill ratio: how often does the strategy see a setup vs how often does the setup actually become a trade. Below 40% fill rate is a hint that the order type doesn’t match the venue’s liquidity.