All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=669138

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg2528swsM


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

chromium-rel-mac11

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8994723208624372112

===== BISECT JOB RESULTS =====
Status: failed


=== Bisection aborted ===
The bisect was aborted because Step('Working on revision chromium@434248,catapult@4e71cc3932.buildbucket.put') failed with return_code 1
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@434183  1.5795   0.729278  8  good
chromium@434218  1.569    0.742945  8  good
chromium@434236  1.558    0.720775  8  good
chromium@434245  1.442    0.172864  8  good
chromium@434247  1.5105   0.400842  8  good
chromium@434248  1.59125  0.313349  8  good
chromium@434249  2.3804   0.16128   5  bad
chromium@434253  2.4574   0.217134  5  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 669138

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=ESPN smoothness.top_25_smooth
Test Metric: first_gesture_scroll_update_latency/ESPN
Relative Change: 55.58%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/1059
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994723208624372112


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5264020892811264

| 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 Tests>AutoBisect.  Thank you!

Ned, heads up that the bisect looks like crrev.com/434249 is the culprit, and the only related CL in that catapult roll is:
2016-11-23 nednguyen Update ts_proxy to latest commit

Kicking off another bisect.

Strange, if there is anything, that CL should make the metric improve.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8994715561561287712

===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Update ts_proxy to latest commit
Author  : nednguyen
Commit description:
  
For all the changes, see: https://github.com/WPO-Foundation/tsproxy/compare/6fae6141104d077c8ceb1136b6ddee5b98f30240...9ae7d2ff822cb7cef15bc54ff614273c99691744

The main change is "Reduced CPU impact when idle or running in bypass mode".

Bug: https://code.google.com/p/chrome-os-partner/issues/detail?id=60009

BUG=chromium:653519

Review-Url: https://codereview.chromium.org/2516973005
Commit  : 4e71cc39324b205febf5a4cfa3a089a1e2aeab74
Date    : Wed Nov 23 17:55:09 2016


===== TESTED REVISIONS =====
Revision                             Mean     Std Dev   N  Good?
chromium@434144                      1.61213  0.691431  8  good
chromium@434226                      1.59438  0.651497  8  good
chromium@434247                      1.56463  0.375614  8  good
chromium@434248                      1.4755   0.325026  8  good
chromium@434248,catapult@982d024dad  1.578    0.571468  8  good
chromium@434248,catapult@4e71cc3932  2.395    0.904442  8  bad    <--
chromium@434249                      2.4656   0.284923  5  bad
chromium@434250                      2.4518   0.187208  5  bad
chromium@434252                      2.4504   0.232184  5  bad
chromium@434257                      2.4802   0.160134  5  bad
chromium@434267                      2.3978   0.359192  5  bad
chromium@434308                      2.4002   0.21951   5  bad

Bisect job ran on: mac_10_11_perf_bisect
Bug ID: 669138

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=ESPN smoothness.top_25_smooth
Test Metric: first_gesture_scroll_update_latency/ESPN
Relative Change: 48.88%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_11_perf_bisect/builds/1067
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8994715561561287712


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5850170380517376

| 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 Tests>AutoBisect.  Thank you!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8982632303327607856