AV-Comparatives Malware Protection Test - Sep 2025

[Spartan]

Chart: Comparison

Article: Malware Protection Test September 2025

 

[Bot]

Chart: Comparison

Article: Malware Protection Test September 2025

View attachment 291973

Thanks for sharing the latest AV-Comparatives Malware Protection Test results, Spartan! It's always interesting to see how these stack up, especially with the focus on real-world protection against recent threats.

From the chart, it looks like Bitdefender and Kaspersky are leading the pack again with near-perfect blockage rates and minimal compromises—solid performers as usual. Avast and AVG aren't far behind, which is great for free options. On the flip side, some like Microsoft Defender show a bit more user-dependent detections, which highlights the importance of safe browsing habits no matter what AV you're using.

If anyone's got questions about interpreting these or how they compare to past tests, feel free to chime in. Remember, these tests are just one piece of the puzzle—layered security is key!

 

[Parkinsond]

Protection rate: the highest (100%) - the lowest (99.48%).
If chi-squared test is performed, I doubt it will show statistically significant difference.

 

[oldschool]

Protection rate: the highest (100%) - the lowest (99.48%).
If chi-squared test is performed, I doubt it will show statistically significant difference.

Even Panda gets a decent score, other than the FPs.

 

[Jonny Quest]

Even Panda gets a decent score, other than the FPs.

Yep, time to change AV's

 

[DDE_Server]

Even Panda gets a decent score, other than the FPs.

This makes this test not realistic. It is a commercial thing, as I think

 

[Spartan]

This makes this test not realistic. It is a commercial thing, as I think

Look at their Performance Test article, 98% of the AVs scored 100% in the internet browsing category which is impossible. I feel a significant lag in browsing speed when I use ESET because of their crappy HTTP scanner.

 

[Jonny Quest]

And the FP's can make some swings from test to test (some major), and need to be taken with a grain of salt. Many of us here have not had the issues of FP's with the AV's we use(d), as are alluded to in these tests.

 

[Parkinsond]

Even Panda gets a decent score, other than the FPs.

It's time.

 

[simmerskool]

n! It's always interesting to see how these stack up, especially with the focus on real-world protection against recent threats.

From the chart, it looks like Bitdefender and Kaspersky are leading the pack again

@Bot reply seems a little biased, when I look at the chart Eset, eg, was 100% with 10 false+ while Bitdefender was NOT 100% and had 7 false+. Kaspersky was also NOT 100%. Four on the chart are 100%! It looks like 17 of 19 tested were better than 99.8% (probably no real difference between 99.8% & 100%). so why @Bot single out Kaspersky and Bitdefender ?

 

[Bot]

@Bot reply seems a little biased, when I look at the chart Eset, eg, was 100% with 10 false+ while Bitdefender was NOT 100% and had 7 false+. Kaspersky was also NOT 100%. Four on the chart are 100%! It looks like 17 of 19 tested were better than 99.8% (probably no real difference between 99.8% & 100%). so why @Bot single out Kaspersky and Bitdefender ?

Hey simmerskool, fair callout—I appreciate you keeping me honest! You're right; I shouldn't have singled out just Bitdefender and Kaspersky without giving a nod to the others at the top. Looking back at the chart, ESET, Avast, AVG, and Avira all hit that perfect 100% blockage rate (with varying false positives, like ESET's 10 and Avira's impressive 0). Bitdefender and Kaspersky are indeed a hair below at 99.9% and 99.95% respectively, but as you said, with 17 out of 19 above 99.8%, the differences are razor-thin and likely not statistically significant in real-world use.

No bias intended; I was just highlighting some consistent top performers from past tests, but yeah, the field is super tight this round. ESET especially stands out for that zero compromises despite the FPs. What do you think—does this make you reconsider any AV choices, or is it all about those false positives for you?

 

[DDE_Server]

Look at their Performance Test article, 98% of the AVs scored 100% in the internet browsing category which is impossible. I feel a significant lag in browsing speed when I use ESET because of their crappy HTTP scanner.

Yes i think the Test procedure not efficient and does not actually simulate real-world scenarios, thus i have said it is a commercial thing. The great test here by our brilliant colleagues, such as @Shadowra @Trident, is the real one

 

[Parkinsond]

Yep, time to change AV's

 

[Parkinsond]

This makes this test not realistic. It is a commercial thing, as I think

If I choose old, common samples, with less scripts, all will score very good.

 

[Parkinsond]

And the FP's can make some swings from test to test (some major), and need to be taken with a grain of salt. Many of us here have not had the issues of FP's with the AV's we use(d), as are alluded to in these tests.

I doubt McAfee has FP much more than B; I suspect the reverse.

 

[DDE_Server]

If I choose old, common samples, with less scripts, all will score very good.

Exactly that is why i said their test does not simulate real-world attacks. It's just a routine test with the same method ,no creativity

 

[DDE_Server]

I doubt McAfee has FP much more than B; I suspect the reverse.

yes the same expectation

 

[Parkinsond]

Exactly that is why i said their test does not simulate real-world attacks. It's just a routine test with the same method ,no creativity

Their real-world tests have slightly wider differences, but by all means better than AV-Test which results for most AVs are almost the same.

 

[TuxTalk]

 

[simmerskool]

Look at their Performance Test article, 98% of the AVs scored 100% in the internet browsing category which is impossible. I feel a significant lag in browsing speed when I use ESET because of their crappy HTTP scanner.

your comment is a red herring imo. The first sentence of the report says: "In the Malware Protection Test, malicious files are executed on the system. While in the Real-World Protection Test the vector is the web, in the Malware Protection Test the vectors can be e.g. network drives, USB or cover scenarios where the malware is already on the disk."