TL;DR: DeJoy's changes appear to have made USPS slightly slower (0.1-0.5 days) in some places, but much less than the delta caused by a holiday like July 4th (see note buried at the very end).
Yes, but it should be noted that average delivery times may hide the severity of the problem. At least from anecdotal information, it sounds like some letters are arriving as usual and a smaller subset are severity delayed.
I think the median and mode delivery times would be helpful numbers rather than just mean.
But they _are_ having a decidedly negative effect on efficiency, and they’re still destroying machines when they said they were going to pause all changes. We shouldn’t just wait for it to get worse.
This is only data for a specific subset of USPS shipments, it doesn't tell us anything about, for a few examples of important categories:
* Prescription refills
* Mail to/from overseas soldiers
* Bills and bill payments
* Regular letters
Those other categories could be much worse, or could be better than Shippos' data here. Most of the reporting out there claims that other categories (like prescriptions) are severely degraded.
The data does show getting a parcel from California to most points in the east will have up to a 40% chance of delay. It seems their internal routes are getting backed up and causing a ripple effect.
Is my summary inaccurate? The article is very careful not to extend any political slant to its data and findings, please extend me the same courtesy. (I'm not even American.)
You say "slightly slower". The metrics behind the delays are much longer and your numbers only seem to include time in transit.
Second reason your summary is inaccurate: DeJoy's changes are rolled out in small pockets of the country thus far. The article seems to be indiscriminate from where those changes took place. The averages here are hence better than from those places where the changes have done the most damage.
I think the median and mode delivery times would be helpful numbers rather than just mean.