A Trunkline living receipt · paper trading · $0 real money
The Bot Scoreboard
Our own trading lab grades itself nightly and publishes the table, wins and losses. Every indicator below is tested on ~3 years of real market history across 12 assets and 3 timeframes (1h/4h/1d), after a 0.1% fee model. Nothing here trades real money, and nothing gets sold, until it passes the locked gates: 100+ trades, 3+ market regimes, out-of-sample, after costs, consistency.
| Indicator | Avg net (all assets & TFs) | Trades |
|---|---|---|
| av_bespoke_v2 | +13.86% | 1,258 |
| zt_router | +10.16% | 3,009 |
| ema_ribbon_pro | +9.97% | 1,196 |
| av_confluence | +4.11% | 2,006 |
| liquidity | +1.08% | 5,313 |
| ema_ribbon | +0.99% | 745 |
| funding_fade | −0.00% | 0 |
| ai_confluence | −1.13% | 478 |
| smc_structure | −1.41% | 3,956 |
| fib_pullback | −3.16% | 2,837 |
| av_bespoke | −3.65% | 2,319 |
| smma_50 | −3.98% | 2,234 |
| ut_bot | −6.93% | 2,525 |
| liquidity_sweep_pro | −8.12% | 1,560 |
| av_pulse | −10.13% | 2,698 |
| liquidity_sweep | −13.56% | 3,300 |
| stoch_rsi | −19.70% | 5,924 |
How to read this honestly: these are directional backtest averages, not verified live results, the multi-regime and out-of-sample gates are not yet passed, which is exactly why nothing is for sale. The interesting story: our upgraded, stricter designs (ribbon_pro, av_confluence) hold up across regimes while their raw ingredients bleed. Losers stay on the table, that’s the point.
See also: the AI Trading Agent Scoreboard (we grade everyone else’s bots too) and Honest Numbers.
Educational only, not financial advice. Paper trading, fee-modeled, dated. Crypto involves risk.
The bots are not the only thing graded here. The system grades its own machinery daily, in public: every collector, publisher, and backup on the box, fresh or stale, printed either way. See the System Pulse
