scenario,task_id,candidate_policy,baseline_policy,metric,direction,n_trials,mean_delta,ci95_low,ci95_high,win_rate
normal,normal,adaptive,heuristic,avg_reward,higher,1,-0.117724,-0.117724,-0.117724,0.0
normal,normal,adaptive,heuristic,avg_cost_usd,lower,1,8080.364292,8080.364292,8080.364292,0.0
normal,normal,adaptive,heuristic,blackout_rate,lower,1,-0.458333,-0.458333,-0.458333,1.0
normal,normal,adaptive,heuristic,constraint_violation_rate,lower,1,2.5,2.5,2.5,0.0
normal,normal,adaptive,heuristic,emissions_tco2,lower,1,99.0795,99.0795,99.0795,0.0
normal,normal,adaptive,heuristic,unmet_energy_mwh,lower,1,-93.724,-93.724,-93.724,1.0
normal,normal,adaptive,heuristic,reserve_shortfall_rate,lower,1,0.5,0.5,0.5,0.0
normal,normal,adaptive,heuristic,stability_event_rate,lower,1,0.0,0.0,0.0,0.0
normal,normal,adaptive,random,avg_reward,higher,1,-0.09893,-0.09893,-0.09893,0.0
normal,normal,adaptive,random,avg_cost_usd,lower,1,8195.056667,8195.056667,8195.056667,0.0
normal,normal,adaptive,random,blackout_rate,lower,1,-0.208333,-0.208333,-0.208333,1.0
normal,normal,adaptive,random,constraint_violation_rate,lower,1,0.041667,0.041667,0.041667,0.0
normal,normal,adaptive,random,emissions_tco2,lower,1,52.6687,52.6687,52.6687,0.0
normal,normal,adaptive,random,unmet_energy_mwh,lower,1,-18.804,-18.804,-18.804,1.0
normal,normal,adaptive,random,reserve_shortfall_rate,lower,1,0.333333,0.333333,0.333333,0.0
normal,normal,adaptive,random,stability_event_rate,lower,1,0.0,0.0,0.0,0.0
normal,normal,heuristic,random,avg_reward,higher,1,0.018794,0.018794,0.018794,1.0
normal,normal,heuristic,random,avg_cost_usd,lower,1,114.692375,114.692375,114.692375,0.0
normal,normal,heuristic,random,blackout_rate,lower,1,0.25,0.25,0.25,0.0
normal,normal,heuristic,random,constraint_violation_rate,lower,1,-2.458333,-2.458333,-2.458333,1.0
normal,normal,heuristic,random,emissions_tco2,lower,1,-46.4108,-46.4108,-46.4108,1.0
normal,normal,heuristic,random,unmet_energy_mwh,lower,1,74.92,74.92,74.92,0.0
normal,normal,heuristic,random,reserve_shortfall_rate,lower,1,-0.166667,-0.166667,-0.166667,1.0
normal,normal,heuristic,random,stability_event_rate,lower,1,0.0,0.0,0.0,0.0
shock,stress_shock,adaptive,heuristic,avg_reward,higher,1,-0.235076,-0.235076,-0.235076,0.0
shock,stress_shock,adaptive,heuristic,avg_cost_usd,lower,1,-16627.043733,-16627.043733,-16627.043733,1.0
shock,stress_shock,adaptive,heuristic,blackout_rate,lower,1,0.1,0.1,0.1,0.0
shock,stress_shock,adaptive,heuristic,constraint_violation_rate,lower,1,2.3,2.3,2.3,0.0
shock,stress_shock,adaptive,heuristic,emissions_tco2,lower,1,-1147.9199,-1147.9199,-1147.9199,1.0
shock,stress_shock,adaptive,heuristic,unmet_energy_mwh,lower,1,2344.847,2344.847,2344.847,0.0
shock,stress_shock,adaptive,heuristic,reserve_shortfall_rate,lower,1,0.3,0.3,0.3,0.0
shock,stress_shock,adaptive,heuristic,stability_event_rate,lower,1,0.6,0.6,0.6,0.0
shock,stress_shock,adaptive,random,avg_reward,higher,1,-0.256967,-0.256967,-0.256967,0.0
shock,stress_shock,adaptive,random,avg_cost_usd,lower,1,-15904.296767,-15904.296767,-15904.296767,1.0
shock,stress_shock,adaptive,random,blackout_rate,lower,1,0.1,0.1,0.1,0.0
shock,stress_shock,adaptive,random,constraint_violation_rate,lower,1,0.333333,0.333333,0.333333,0.0
shock,stress_shock,adaptive,random,emissions_tco2,lower,1,-1225.6502,-1225.6502,-1225.6502,1.0
shock,stress_shock,adaptive,random,unmet_energy_mwh,lower,1,2504.712,2504.712,2504.712,0.0
shock,stress_shock,adaptive,random,reserve_shortfall_rate,lower,1,0.366667,0.366667,0.366667,0.0
shock,stress_shock,adaptive,random,stability_event_rate,lower,1,0.666667,0.666667,0.666667,0.0
shock,stress_shock,heuristic,random,avg_reward,higher,1,-0.021892,-0.021892,-0.021892,0.0
shock,stress_shock,heuristic,random,avg_cost_usd,lower,1,722.746967,722.746967,722.746967,0.0
shock,stress_shock,heuristic,random,blackout_rate,lower,1,0.0,0.0,0.0,0.0
shock,stress_shock,heuristic,random,constraint_violation_rate,lower,1,-1.966667,-1.966667,-1.966667,1.0
shock,stress_shock,heuristic,random,emissions_tco2,lower,1,-77.7303,-77.7303,-77.7303,1.0
shock,stress_shock,heuristic,random,unmet_energy_mwh,lower,1,159.865,159.865,159.865,0.0
shock,stress_shock,heuristic,random,reserve_shortfall_rate,lower,1,0.066667,0.066667,0.066667,0.0
shock,stress_shock,heuristic,random,stability_event_rate,lower,1,0.066667,0.066667,0.066667,0.0
outage,outage,adaptive,heuristic,avg_reward,higher,1,-0.2587,-0.2587,-0.2587,0.0
outage,outage,adaptive,heuristic,avg_cost_usd,lower,1,-16252.427967,-16252.427967,-16252.427967,1.0
outage,outage,adaptive,heuristic,blackout_rate,lower,1,0.066667,0.066667,0.066667,0.0
outage,outage,adaptive,heuristic,constraint_violation_rate,lower,1,2.2,2.2,2.2,0.0
outage,outage,adaptive,heuristic,emissions_tco2,lower,1,-1125.7141,-1125.7141,-1125.7141,1.0
outage,outage,adaptive,heuristic,unmet_energy_mwh,lower,1,2310.671,2310.671,2310.671,0.0
outage,outage,adaptive,heuristic,reserve_shortfall_rate,lower,1,0.3,0.3,0.3,0.0
outage,outage,adaptive,heuristic,stability_event_rate,lower,1,0.633333,0.633333,0.633333,0.0
outage,outage,adaptive,random,avg_reward,higher,1,-0.294903,-0.294903,-0.294903,0.0
outage,outage,adaptive,random,avg_cost_usd,lower,1,-15877.802767,-15877.802767,-15877.802767,1.0
outage,outage,adaptive,random,blackout_rate,lower,1,0.2,0.2,0.2,0.0
outage,outage,adaptive,random,constraint_violation_rate,lower,1,0.466667,0.466667,0.466667,0.0
outage,outage,adaptive,random,emissions_tco2,lower,1,-1215.8771,-1215.8771,-1215.8771,1.0
outage,outage,adaptive,random,unmet_energy_mwh,lower,1,2456.939,2456.939,2456.939,0.0
outage,outage,adaptive,random,reserve_shortfall_rate,lower,1,0.433333,0.433333,0.433333,0.0
outage,outage,adaptive,random,stability_event_rate,lower,1,0.7,0.7,0.7,0.0
outage,outage,heuristic,random,avg_reward,higher,1,-0.036203,-0.036203,-0.036203,0.0
outage,outage,heuristic,random,avg_cost_usd,lower,1,374.6252,374.6252,374.6252,0.0
outage,outage,heuristic,random,blackout_rate,lower,1,0.133333,0.133333,0.133333,0.0
outage,outage,heuristic,random,constraint_violation_rate,lower,1,-1.733333,-1.733333,-1.733333,1.0
outage,outage,heuristic,random,emissions_tco2,lower,1,-90.163,-90.163,-90.163,1.0
outage,outage,heuristic,random,unmet_energy_mwh,lower,1,146.268,146.268,146.268,0.0
outage,outage,heuristic,random,reserve_shortfall_rate,lower,1,0.133333,0.133333,0.133333,0.0
outage,outage,heuristic,random,stability_event_rate,lower,1,0.066667,0.066667,0.066667,0.0
renewable_collapse,renewable_collapse,adaptive,heuristic,avg_reward,higher,1,-0.242249,-0.242249,-0.242249,0.0
renewable_collapse,renewable_collapse,adaptive,heuristic,avg_cost_usd,lower,1,-16758.002233,-16758.002233,-16758.002233,1.0
renewable_collapse,renewable_collapse,adaptive,heuristic,blackout_rate,lower,1,0.1,0.1,0.1,0.0
renewable_collapse,renewable_collapse,adaptive,heuristic,constraint_violation_rate,lower,1,2.2,2.2,2.2,0.0
renewable_collapse,renewable_collapse,adaptive,heuristic,emissions_tco2,lower,1,-1185.0863,-1185.0863,-1185.0863,1.0
renewable_collapse,renewable_collapse,adaptive,heuristic,unmet_energy_mwh,lower,1,2367.081,2367.081,2367.081,0.0
renewable_collapse,renewable_collapse,adaptive,heuristic,reserve_shortfall_rate,lower,1,0.3,0.3,0.3,0.0
renewable_collapse,renewable_collapse,adaptive,heuristic,stability_event_rate,lower,1,0.666667,0.666667,0.666667,0.0
renewable_collapse,renewable_collapse,adaptive,random,avg_reward,higher,1,-0.249149,-0.249149,-0.249149,0.0
renewable_collapse,renewable_collapse,adaptive,random,avg_cost_usd,lower,1,-16281.8909,-16281.8909,-16281.8909,1.0
renewable_collapse,renewable_collapse,adaptive,random,blackout_rate,lower,1,0.1,0.1,0.1,0.0
renewable_collapse,renewable_collapse,adaptive,random,constraint_violation_rate,lower,1,0.133333,0.133333,0.133333,0.0
renewable_collapse,renewable_collapse,adaptive,random,emissions_tco2,lower,1,-1252.6913,-1252.6913,-1252.6913,1.0
renewable_collapse,renewable_collapse,adaptive,random,unmet_energy_mwh,lower,1,2503.293,2503.293,2503.293,0.0
renewable_collapse,renewable_collapse,adaptive,random,reserve_shortfall_rate,lower,1,0.266667,0.266667,0.266667,0.0
renewable_collapse,renewable_collapse,adaptive,random,stability_event_rate,lower,1,0.733333,0.733333,0.733333,0.0
renewable_collapse,renewable_collapse,heuristic,random,avg_reward,higher,1,-0.0069,-0.0069,-0.0069,0.0
renewable_collapse,renewable_collapse,heuristic,random,avg_cost_usd,lower,1,476.111333,476.111333,476.111333,0.0
renewable_collapse,renewable_collapse,heuristic,random,blackout_rate,lower,1,0.0,0.0,0.0,0.0
renewable_collapse,renewable_collapse,heuristic,random,constraint_violation_rate,lower,1,-2.066667,-2.066667,-2.066667,1.0
renewable_collapse,renewable_collapse,heuristic,random,emissions_tco2,lower,1,-67.605,-67.605,-67.605,1.0
renewable_collapse,renewable_collapse,heuristic,random,unmet_energy_mwh,lower,1,136.212,136.212,136.212,0.0
renewable_collapse,renewable_collapse,heuristic,random,reserve_shortfall_rate,lower,1,-0.033333,-0.033333,-0.033333,1.0
renewable_collapse,renewable_collapse,heuristic,random,stability_event_rate,lower,1,0.066667,0.066667,0.066667,0.0