Running average score as scenarios accumulate. Shaded band = +/- 1 standard deviation.
Each point shows the mean score across all scenarios completed up to that point. The shaded region represents +/- 1 std dev across runs. Scenarios ordered: Crash Recovery (cr-1 to cr-4), Secret Containment (sc-1 to sc-5), Handoff Reliability (hr-1 to hr-4), Channel Security (cs-1 to cs-5), Cost Control (cc-1 to cc-5).