false positive?

We just deployed server push on our site here:
The testing it seems to me we are getting a bad score because we are pushing 3 things.
Does this testing need to be updated to take into account that we are pushing ?
On the google mobile speed test as seen here we receive great scores.

thx Brian

Hey Brian - can you post a link the wpt result that is showing a bad score? And which score specifically?

Keep in mind google’s page speed insights scores don’t always mirror a better user experience (although I’m sure your server push is a step in the right direction)

This AM I had this test… IS this a good score? I have other tests that show different things as seen here: https://developers.google.com/speed/pagespeed/insights/?url=https%3A%2F%2Fshopbakersnook.com%2Fm5%2F1419--46-piece-deluxe-decorating-set-by-wilton.html
BUT WEBpage shows this:: https://www.webpagetest.org/result/180707_KJ_1621afc0f8461a04bd74384f4002163b/
running the testing today I am not getting the same results as yesterday. The scores seem to be not so bad…

I dunno what to think here…

Using lighthouse as testing we get really good scores so I just dunno what to think.
Which test is the best ??
Lighthouse? WBT? Google pagespeed insights?
Smoke signals?

That’s actually pretty typical. Page performance can vary widely from day to day on some servers. And depending on what public test agent you get sometimes they’re less powerful than they should be.

Ways around this are to have it run multiple tests and use the median result, set up a private instance and ensure you’re using beefy agents (c4.large or better if you’re on EC2), or use a 3rd party service to monitor regularly.