But those cars would have been where they are because they might have had an unfair advantage. It seems right to me.

Additionally, if 2 out of 4 cars fail the test, maybe it’s good to test all of them.

I agree. If the sample has a 50%+ failure rate then maybe it should trigger a wider inspection.

I read that the FIA keeps an eye on porpoising and that is the reason HAM and LEC got selected for a test. Because a high degree of porpoising might result in high wear on the skid plates. So there is some kind of logic that makes sense there. I guess they had to check at least VER and NOR to make sure their logic held up.

Found the source (in Dutch): https://nl.motorsport.com/f1/news/diskwalificatie-lewis-hamilton-charles-leclerc-gp-amerika-fia-controle-auto-max-verstappen/10536672/

Google translate of relevant section:

For example, the FIA ​​informed this website a little later in the evening that it is making a reasoned decision. "Of course we are not blind to what is happening around us." It means that the FIA ​​looks, among other things, at the so-called porpoising matrix when selecting the cars. This porpoising overview shows the bouncing of the cars, which logically has an effect on the wear of the floorboards. Cars that stand out have a greater chance of being examined more closely than others. For example, the FIA ​​has the impression that Sainz and George Russell drove with a higher ride height than their teammates, which would mean they would be in a good position.

to disqualify 2 of the top cars just like this after the race kinda undoes the whole story of the race.

They can't check for plank wear before the race 😅

Additionally, if 2 out of 4 cars fail the test, maybe it's good to test all of them.

It's a random spot check. Not something that would be done to the entire grid. It's literally practically impossible to check for every rule on every car after or during every race, which is why random spot checks exist.