Here is the MRG-Effitas real-world test.
http://www.mrg-effitas.com/wp-content/uploads/2012/06/MRG-Effitas-Real-Time-Protection-Test-Q1-20131.pdf
You can see Microsoft Security Essentials fails again. Now I'm running into a problem at work with this. Our CIO is a big Microsoft guy and seems to be buying in Microsofts excuses for horrid AV test results across the board. Microsofts excuse on this is that their antivirus works based on what is attacking their users at the moment. This real-world test seems to disprove this as it is running a test based on recently detected malware for Windows systems and applies it via the most common way machines are infected. (Downloading a file via IE). He believes that MSE works in such a way, that tests such as these don't correctly report its detection rates. (I don't see how that is possible in this scenario). I am also hearing, "M$ owns the source code for Windows, so who better to develop an Antivirus for it?". (good theory, bad in practice)
I have always thought MSE has had its advantages in integrating into Windows well, updating via Windows updates, being lightweight, low false positives, and good cleanup rate on what it does detect. However the glaring issue, appears to be horrible detection rates, apparently due to horrible heuristic scanning andnot being aggressive enough.
Am I way off here? Is there something that is not being accounted for in these tests where MSE is concerned? It seems to me that M$ is simply using a thinly veiled excuse. Especially considering the army of machines that I have had to clean in which rogue antivirus applications have completely overrun a Windows machine running MSE. (and MSE still won't detect an issue). It seems to me that a Windows machine would be best served using an Antivirus with a proven track record (i.e. Emsisoft or Kaspersky) along with Windows Defender.