All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=697445

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgpNWx-ggM


Bot(s) for this bug's original alert(s):

android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8986315655939787088

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8986156197633722032

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : speedometer
  Metric       : Total/http___browserbench.org_Speedometer_

Revision             Result                  N
chromium@453429      32052.3 +- 19495.4      21      good
chromium@453680      33891.4 +- 10106.1      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests speedometer

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8986156197633722032

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5294513234378752


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!

Nexus5x bot is going completely crazy it seems. Regressions happening also on ref and not other bot.

+martiniss, jbudorick: any changes you know of on the N5X perf recipe or underlying android infra on 2/28? It seems more noisy. The first change I thought of was the device swap for BattOr, but these two builds are with the same phones:
https://build.chromium.org/p/chromium.perf/builders/Android%20Nexus5X%20Perf%20%283%29/builds/3474
https://build.chromium.org/p/chromium.perf/builders/Android%20Nexus5X%20Perf%20%283%29/builds/3475

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8982624231109293552