Just to clarify, the sudden spikes is in ERR_CONTENT_DECODING_FAILED.  It's about a two fold increase in frequency.

If you look at Net.ContentDecodingFailed.FilterType this appears to be a spike in problems with GZIP.  Yeah, I'm confused by that too.

https://uma.googleplex.com/timeline_v2?q=%7B%22day_count%22%3A%22All%22%2C%22end_date%22%3A%222016%2F05%2F04%22%2C%22window_size%22%3A%221%22%2C%22filters%22%3A%5B%7B%22fieldId%22%3A%22channel%22%2C%22operator%22%3A%22EQ%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%224%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%22milestone%22%2C%22operator%22%3A%22GE%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%2248%22%5D%7D%5D%2C%22histograms%22%3A%5B%5B%22Net.ContentDecodingFailed.FilterType%22%5D%2C%5B%22Net.ErrorCodesForMainFrame3%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%22bucket%22%2C%22buckets%22%3A%5B%220%22%2C%221%22%2C%223%22%2C%224%22%2C%225%22%2C%226%22%2C%222%22%5D%7D%7D%2C%7B%22measureModel%22%3A%7B%22measure%22%3A%22bucket%22%2C%22buckets%22%3A%5B%22330%22%5D%7D%2C%22zeroBased%22%3Afalse%2C%22showLowVolumeData%22%3Atrue%7D%5D%7D

Hrm...If you remove the milstone, things seem a bit clearer.  Volume went down a lot for M49, and then M52 spiked up again, but still looks to be less than the previous volume.  (Bizarrely, the histogram goes all the way back to Chrome 0...which didn't even have this histogram...  I'm confused).

https://uma.googleplex.com/timeline_v2?q=%7B%22day_count%22%3A%22All%22%2C%22end_date%22%3A%222016%2F05%2F04%22%2C%22window_size%22%3A%221%22%2C%22filters%22%3A%5B%7B%22fieldId%22%3A%22channel%22%2C%22operator%22%3A%22EQ%22%2C%22study%22%3A%22%22%2C%22selected%22%3A%5B%224%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%5D%2C%22histograms%22%3A%5B%5B%22Net.ContentDecodingFailed.FilterType%22%5D%2C%5B%22Net.ErrorCodesForMainFrame3%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%22bucket%22%2C%22buckets%22%3A%5B%220%22%2C%221%22%2C%223%22%2C%224%22%2C%225%22%2C%226%22%2C%222%22%5D%7D%7D%2C%7B%22measureModel%22%3A%7B%22measure%22%3A%22bucket%22%2C%22buckets%22%3A%5B%22330%22%5D%7D%2C%22zeroBased%22%3Afalse%2C%22showLowVolumeData%22%3Atrue%7D%5D%7D

Erm...milestone, not millstone, though I suppose UMA data does become more exciting if you drop a millstone on it.

Agreed on the confusion on Chrome 0, but I basically trust the more recent results (specifically, I vividly remember the spike in December 2014 :-}), so I'm not inclined to worry about it biasing the more recent results.

Having said that, I'm still concerned about the spike upwards.  Even if the question is "Why did it drop in the M49-52 time frame?" it's still a disturbing shift (back up).  

Oh, the 0 does not mean Chrome 0, but the first version of Chrome we have data for...ok, that makes more sense.

It was confusing because 49 is Chrome 49.  It's actually Chrome 50 that has the regression, not Chrome 52.