Reputable websites for antivirus top lists and tests

Try3 said:

In av-test's case, they rate detections overall without identifying which was found by which AV package.
- At the moment, they say the industry average detection is 99.4% but Defender & several others detected 100% of the samples av-test threw at them.
- If, in the June/July tests, Defender & Norton only detect 99.9% of the samples, av-test will not say which ones they fail to detect or whether or not they both fail to detect the same ones.

You can see the detailed results for each package by clicking on the little arrow next to each entry in their results tables

This is so funny my friend because logic is very simple...

Premise 1: out of 15 tested AV software average detection rate is approx 99%
Premise 2: A single test case uses same malware base
Conclusion: in 99% of the cases they find same malware

you can confirm premise 2 by following the steps you gave me and notice the following is same for all AV software:

Protection against 0-day malware attacks ...
326 samples used
Detection of widespread and prevalent malware ...
15,664 samples used
etc... the numbers are all the same

Therefore they're all tested against same malware base therefore it logically follows that in 99% of cases they're equivalently capable or fairly tested.
There is only 1% available for competition.
But that's absurdity and suggests that either malware base is too small or no custom written malware was used for testing.

@Try3
Btw. I didn't mean to say that your post is funny but rather that the logic by which these tests are done look funny for obvious reasons.

Which is that malware base is equivalent in order for the tests to be fair and no FUD malware is used in testing.
I don't know how they gather 326 of 0-day malware though in a single day.