Posting like that, I'm afraid you are loosing your credibility...
In post from 11/3/2011 2:32:02 AM , you actually attacked during blizzard, not summer as stated in the post, and in town (I somehow find that clear+town is different to clear, but maybe that is superstition). Anyway, Germans are not that great attacking in blizzard (though I agree that the results is not that great). However, you attacked guards rifle corps - those guys have higher morale thus you inflicted lower losses. I found that losses depends on morale a lot. I don't know your morale - if it was not that great, that would explain everything...
In post 11/3/2011 2:35:13 AM , you shown examples of SU attacking and complain, about Germans who lost more than Russians. Actually, Russians attacked in snow, not blizzard. The SU also have huge advantage in guns and used guards corps in the mix which have high morale, so lower losses. Finally, in the second example, Russian player lost more than 1/5 of all the tanks put into attack (tanks were not guards, mind you). If that would happen to you in your blizzard attack, you would lose 170 tanks, not 56 [and in blizzard should be much more breakdowns]. Quite a difference eh? Also, the odds for the SU are much better than those for Germans in blizzard - 1:4.2 is actually much different to 1:5.7.
Finally, the amount of losses is depending on many more factors than just the combat odds, and you know it well; the morale of troops is most important. Thus the funnily low losses of the NKVD border regiments in 1.4.xx
Anyway, it appears to me that your arguments are not strong enough and not well substantiated. Even though I agree with you on many issues you point out (though I am a SU player, ironically), it seems to me that recently you started to stretch the truth too much. As such, I'm afraid, I will start soon ignoring your posts without real evaluation of merit, as many others do already... Trust me, by stretching the truth your arguments do not support anything and your voice will be ignored. Even worse, once you find a real issue nobody will listen. Actually, it already happens as I cannot say whether the issue you state above is real or not - I can only tell that the data you provided is not credible. But by providing not credible data, you make the whole issue to appear as at best unlikely.