All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=821963

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=9be824cadcfa184d304c4343c1eb6cda77fec472a45627ae4e6b80c1bc57547b


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

chromium-rel-mac11-pro
chromium-rel-mac12

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14e42c41440000

📍 Couldn't reproduce a difference.
https://pinpoint-dot-chromeperf.appspot.com/job/14e42c41440000

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/11f40afe240000

This graph (https://chromeperf.appspot.com/report?sid=babde592701168f6e2251c0ceb8b87b93eca17d8b285b8a5b136cd4e897f6574&start_rev=538953&end_rev=546908) clearly shows that this was indeed a real regression, but the bisect was kicked off on another metric (https://chromeperf.appspot.com/report?sid=b97e405319fbbcfb0b2f11e52b63c5e757d6a5528c6df1e660d37f9ceee4e9df&rev=542711) that regressed much less. Even though this bug is ~3 months old, I kicked off a bisect on this graph and strongly suspect that the bisect will turn up a result.

📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/11f40afe240000

Add fieldtrial testing config for new media controls by steimel@chromium.org
https://chromium.googlesource.com/chromium/src/+/fdfc5ae4154a67e0539b9d0a77b3c5e2da38f40b
0.09709 → 0.1082 (+0.01109)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions