All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=621564

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgwpTpqgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICggrv9qQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICggqbjswsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICggtKVuAoM


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

android-galaxy-s5
android-nexus5
android-nexus5X
android-nexus6

In the similar range, 400596 - 400598,
ChromiumPerf/android-nexus9/smoothness.sync_scroll.key_mobile_sites_smooth / first_gesture_scroll_update_latency / http___www.theverge.com_2012_10_28_3568746_amazon-7-inch-fire-hd-ipad-mini-ad-ballsy had a 93.9% regression.

This is a very small range, either one skia change or one catapult change. Suspecting the catapult change 91953db7ae2b087fdc2581c39d07e3862e31fa96 since it references scrolling.

bccheng, is this expected?

Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

bccheng@ - ping!

Sorry for not getting back sooner as I was travelling/on vacation.

I have reverted my change on 6/28 but the problem still persists, so the actual cause should be something else.

https://chromium.googlesource.com/external/github.com/catapult-project/catapult/+/d4464872051362c384127e2e21fd25557bbd8a83

Actually, the graphs all recover on 6/28. So it was exactly that :)