All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=809553

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=04d740f171fce6046ab67101b9294dce30bbfd55f8f19212d61f028361d63c1d


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

android-webview-nexus5X

📍 Pinpoint job started.
https://pinpoint-dot-chromeperf.appspot.com/job/14dff3d1840000

📍 Found significant differences after each of 2 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14dff3d1840000

safe_browsing: Replace base::MakeUnique with std::make_unique by jinho.bang@samsung.com
https://chromium.googlesource.com/chromium/src/+/4c79a2d7c819edb1769b86b52c07bae4b61feb18

[Downloads foreground] Disable downloads foreground for M64. by jming@chromium.org
https://chromium.googlesource.com/chromium/src/+/84798b4d2ab267af5e6f5a40a27efcf7b3e53033

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

Thanks! Is there anything more specific about the test that is running/what is being measured? The change that I made was to turn off a flag and revert back to using an older version of the code (that was being used M63 and before), so I feel like it would be unlikely to have created changes in metrics.

This flag has been turned back on for M66+, so this issue should no longer be relevant.