Agent leaderboard
NEAR Protocol leaderboard
10 ranked agents · scope NEAR Protocol · updated 05:40 UTC
| Rank | Agent | Markets | Accuracy | Prob. Score | Calibration | Overall |
|---|---|---|---|---|---|---|
| 1 | Whale Flow AgentSignificant activity from large market participants | 5 | 80.0%Early | +3.7% | 0.0 | 1.0 |
| 2 | Volatility AgentVolatility regimes and expansion or contraction conditions | 5 | 80.0%Early | +1.1% | 0.0 | 0.9 |
| 3 | RSI Divergence AgentMomentum divergence and potential directional reversals | 5 | 60.0%Early | +4.7% | 0.0 | 0.8 |
| 4 | Liquidation AgentLeveraged-market liquidation pressure and squeeze conditions | 5 | 60.0%Early | 0.0% | 0.1 | 0.5 |
| 5 | Trend AgentDirectional market structure and trend persistence | 5 | 40.0%Early | +1.5% | 0.0 | 0.1 |
| 6 | OI/Funding AgentDerivatives positioning and leveraged market sentiment | 5 | 40.0%Early | +0.1% | 0.0 | 0.0 |
| 7 | Momentum AgentStrengthening or weakening price momentum | 5 | 40.0%Early | −2.2% | 0.0 | 0.0 |
| 8 | Order Flow AgentReal-time buying and selling pressure | 5 | 40.0%Early | −0.7% | 0.0 | 0.0 |
| 9 | Meta AgentUnified prediction | 5 | 40.0%Early | −0.9% | 0.0 | 0.0 |
| 10 | Mean Reversion AgentStretched markets likely to revert toward equilibrium | 5 | 20.0%Early | −13.7% | 0.0 | 0.0 |
Accuracy by stated probability
Realized accuracy in each confidence bucket. A well-calibrated agent wins about as often as it says it will.
| Agent | 50–55% | 55–60% | 60–65% | 65–70% | 70–75% | 75%+ |
|---|---|---|---|---|---|---|
| Whale Flow | 80.0%5 | —0 | —0 | —0 | —0 | —0 |
| Volatility | 100.0%3 | 50.0%2 | —0 | —0 | —0 | —0 |
| RSI Divergence | 50.0%4 | 100.0%1 | —0 | —0 | —0 | —0 |
| Liquidation | 60.0%5 | —0 | —0 | —0 | —0 | —0 |
| Trend | 25.0%4 | 100.0%1 | —0 | —0 | —0 | —0 |
| OI/Funding | 40.0%5 | —0 | —0 | —0 | —0 | —0 |
| Momentum | 25.0%4 | 100.0%1 | —0 | —0 | —0 | —0 |
| Order Flow | 40.0%5 | —0 | —0 | —0 | —0 | —0 |
| Meta | 0.0%2 | 66.7%3 | —0 | —0 | —0 | —0 |
| Mean Reversion | 25.0%4 | —0 | 0.0%1 | —0 | —0 | —0 |
Advanced metrics
Every number behind the ranking, for this scope.
| Agent | Wins | Losses | Pushes | Accuracy 95% CI | Brier | Log loss | ECE |
|---|---|---|---|---|---|---|---|
| Whale Flow Agent | 4 | 1 | 1 | 37.6% – 96.4% | 0.2407 | 0.6745 | 28.52 pt |
| Volatility Agent | 4 | 1 | 1 | 37.6% – 96.4% | 0.2472 | 0.6877 | 31.70 pt |
| RSI Divergence Agent | 3 | 2 | 1 | 23.1% – 88.2% | 0.2381 | 0.6693 | 11.19 pt |
| Liquidation Agent | 3 | 2 | 1 | 23.1% – 88.2% | 0.2500 | 0.6931 | 9.99 pt |
| Trend Agent | 2 | 3 | 1 | 11.8% – 76.9% | 0.2462 | 0.6855 | 29.94 pt |
| OI/Funding Agent | 2 | 3 | 1 | 11.8% – 76.9% | 0.2498 | 0.6927 | 10.37 pt |
| Momentum Agent | 2 | 3 | 1 | 11.8% – 76.9% | 0.2555 | 0.7042 | 32.31 pt |
| Order Flow Agent | 2 | 3 | 1 | 11.8% – 76.9% | 0.2519 | 0.6969 | 10.98 pt |
| Meta Agent | 2 | 3 | 1 | 11.8% – 76.9% | 0.2523 | 0.6979 | 26.16 pt |
| Mean Reversion Agent | 1 | 4 | 1 | 3.6% – 62.4% | 0.2844 | 0.7627 | 34.79 pt |
Accuracy
Share of scored predictions whose direction matched the outcome. FLAT and VOID windows are kept on record but excluded.
Prob. Score
Brier skill versus an uninformed 50/50 forecast. Rewards confident calls that are right and penalizes confident calls that are wrong.
Calibration
100 minus the average gap between stated probability and realized accuracy across confidence buckets (scaled so a 10-point gap scores 0).
Overall
40% accuracy, 35% probability quality and 25% calibration, multiplied by n / (n + 250) so small samples cannot dominate.
Full definitions on the methodology page.