This is the best changelist I've gotten so far:
https://chromium.googlesource.com/chromium/src/+log/49.0.2623.0..50.0.2661.0?pretty=fuller&n=10000
It isn't tight enough to be at all useful.

It looks like this is fine in M51.
https://uma.googleplex.com/timeline_v2?q=%7B%22day_count%22%3A%22All%22%2C%22end_date%22%3A%22latest%22%2C%22window_size%22%3A%221%22%2C%22filters%22%3A%5B%7B%22fieldId%22%3A%22channel%22%2C%22operator%22%3A%22COMPARE%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%223%22%5D%7D%2C%7B%22fieldId%22%3A%22platform%22%2C%22operator%22%3A%22COMPARE%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%22W%22%5D%7D%2C%7B%22fieldId%22%3A%22version_tags%22%2C%22operator%22%3A%22CONTAINS%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%22D%22%5D%7D%5D%2C%22histograms%22%3A%5B%5B%22Event.Latency.TouchToFirstScrollUpdateSwapBegin%22%5D%5D%2C%22default_entry_values%22%3A%7B%22measureModel%22%3A%7B%22measure%22%3A%22%22%2C%22buckets%22%3A%5B%5D%2C%22percentiles%22%3A%5B%2250%22%5D%2C%22selectedFormulas%22%3A%5B%5D%2C%22allFormulas%22%3A%5B%5D%7D%2C%22zeroBased%22%3Atrue%2C%22logScale%22%3Afalse%2C%22showLowVolumeData%22%3Afalse%2C%22showVersionAnnotations%22%3Atrue%7D%2C%22entries%22%3A%5B%7B%22measureModel%22%3A%7B%22measure%22%3A%22percentile%22%2C%22percentiles%22%3A%5B%2295%22%5D%7D%7D%5D%7D

Lan determined that the regression only takes place on the renderer side.
It looks like this regression never happened on Canary - maybe it was merged to dev?

https://uma.googleplex.com/p/chrome/timeline_v2/?q=%7B%22day_count%22%3A%22200%22%2C%22end_date%22%3A%22latest%22%2C%22window_size%22%3A%227%22%2C%22filters%22%3A%5B%7B%22fieldId%22%3A%22channel%22%2C%22operator%22%3A%22COMPARE%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%223%22%2C%224%22%2C%222%22%2C%221%22%5D%7D%2C%7B%22fieldId%22%3A%22platform%22%2C%22operator%22%3A%22EQ%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%22W%22%5D%7D%2C%7B%22fieldId%22%3A%22version_tags%22%2C%22operator%22%3A%22CONTAINS%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%22D%22%5D%7D%5D%2C%22histograms%22%3A%5B%5B%22Event.Latency.Browser.TouchAcked%22%5D%5D%2C%22default_entry_values%22%3A%7B%22measureModel%22%3A%7B%22measure%22%3A%22%22%2C%22buckets%22%3A%5B%5D%2C%22percentiles%22%3A%5B%2250%22%5D%2C%22selectedFormulas%22%3A%5B%5D%2C%22allFormulas%22%3A%5B%5D%7D%2C%22zeroBased%22%3Atrue%2C%22logScale%22%3Afalse%2C%22showLowVolumeData%22%3Afalse%2C%22showVersionAnnotations%22%3Atrue%7D%2C%22entries%22%3A%5B%7B%22measureModel%22%3A%7B%22measure%22%3A%22percentile%22%2C%22percentiles%22%3A%5B%2295%22%5D%7D%7D%5D%7D

Also we noticed that the regression only happened to Windows X86 but not X86-64.
https://uma.googleplex.com/p/chrome/timeline_v2/?q=%7B%22day_count%22%3A%2290%22%2C%22end_date%22%3A%22latest%22%2C%22window_size%22%3A%221%22%2C%22filters%22%3A%5B%7B%22fieldId%22%3A%22channel%22%2C%22operator%22%3A%22COMPARE%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%224%22%5D%7D%2C%7B%22fieldId%22%3A%22version_tags%22%2C%22operator%22%3A%22CONTAINS%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%22D%22%5D%7D%2C%7B%22fieldId%22%3A%22platform%22%2C%22operator%22%3A%22COMPARE%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%22W%22%5D%7D%2C%7B%22fieldId%22%3A%22hw_cpu%22%2C%22operator%22%3A%22COMPARE%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%22x86%22%2C%22x86_64%22%5D%7D%5D%2C%22histograms%22%3A%5B%5B%22Event.Latency.Browser.TouchAcked%22%5D%2C%5B%22Event.Latency.TouchToFirstScrollUpdateSwapBegin%22%5D%2C%5B%22Event.Latency.TouchToScrollUpdateSwapBegin%22%5D%5D%2C%22default_entry_values%22%3A%7B%22measureModel%22%3A%7B%22measure%22%3A%22%22%2C%22buckets%22%3A%5B%5D%2C%22percentiles%22%3A%5B%2250%22%5D%2C%22selectedFormulas%22%3A%5B%5D%2C%22allFormulas%22%3A%5B%5D%7D%2C%22zeroBased%22%3Atrue%2C%22logScale%22%3Afalse%2C%22showLowVolumeData%22%3Afalse%2C%22showVersionAnnotations%22%3Atrue%7D%2C%22entries%22%3A%5B%7B%22measureModel%22%3A%7B%22measure%22%3A%22percentile%22%2C%22percentiles%22%3A%5B%2295%22%5D%7D%7D%2C%7B%22measureModel%22%3A%7B%22measure%22%3A%22percentile%22%2C%22percentiles%22%3A%5B%2295%22%5D%7D%7D%2C%7B%22measureModel%22%3A%7B%22measure%22%3A%22percentile%22%2C%22percentiles%22%3A%5B%2295%22%5D%7D%7D%5D%7D

It does show up a bit on 64 bit machines, but the regression is _much_ larger on 32 bit machines.

This could just be because in general 64 bit machines are more modern in other ways though.

This recovered, so down prioritizing.