I've run two bisects with different parameters (for the same regression,  issue 633193 ):

Bisect 1: https://chromeperf.appspot.com/buildbucket_job_status/9005529157844552592
max_time_minutes: 20
repeat_count: 10

Bisect 2: https://chromeperf.appspot.com/buildbucket_job_status/9004900681642967744
max_time_minutes: 100
repeat_count: 100

Nevertheless, the benchmark was run the same number of times in both cases:

Bisect 1: https://bugs.chromium.org/p/chromium/issues/detail?id=633193#c3
Revision         Mean    Std Dev  N  Good?
chromium@407992  464392  1022.11  8  good
chromium@408012  463840  0.0      8  bad

Bisect 2: https://bugs.chromium.org/p/chromium/issues/detail?id=633193#c7
Revision         Mean    Std Dev  N  Good?
chromium@407954  463840  0.0      8  good
chromium@408012  465062  3456.34  8  bad

Given that I increased max_time_minutes 5 times and repeat_count 10 times, I would expect that the second bisect would run each revision ~40 times (5×8).