I just wish they could balance this goddamn game.
Seriously, Soviets are way too good early, or rather, German squads get gibbed too easily. If you take few losses with Soviets and play super well, usually you are looking at a float of 600-800 MP midgame, just retarded.
Having played a lot more 2v2s lately where I can get Panthers out, yeah, I now admit they're kinda ridiculous. Vet 2 is obscene, but very rare (unless they use Elite troops). SU85s can keep them at bay, but you really need to double team them with some guards unless you have an IS-2 or ISU.
In which case in the game I just played, I counter with Tiger Ace, or a Tiger to support a Panther and act as a damage sink.
I've said it before, but it bears repeating - you can't beat the M10 and Hellcats, and vet Airborne from vCOH to counter late Axis armor. Soviets need a mobile tank destroyer that you can micro and has the speed to chase damaged Panthers. I would like the game a lot more with M10s and no SU85s.
While early game might favor soviets a bit, it is nowhere near scewed as far as that late game. There has never been a soviet opening I haven't been able to defeat with a bit a manuevering and flanking, but out manuevering as the soviets in the late game is terribly unrewarding.
Just the other day I was playing a game against an opponent who I was beating pretty well until he brought out an elephant against my su-85s. I managed to force away his infantry and flank him from both behind and the front with fricking su-85s which isn't exactly easy too do, but the damn thing was able to turn kill one and then turn and kill the other while only taking a quarter of damage.
A lot of the tanks on the ostheer except the Tiger have way too much health for the amount of armor and sometimes speed they have. It feels as though they were balanced against a total different faction that actually has effective AT. The only reason I suggested my change in the first place is because it just takes way too long to kill even if you do outmanuever them.
There was some talk at variable penetration at ranges, which I think could be a better solution then even the one I suggested.