Rostov is the second scenario I've played, and I was shocked (delightedly so) at the operational bonanzas that aggressive supply acquisition can open up. Must admit, I stumbled into what you see in the late game on this replay. Chalk it up to beginner's luck. The other thing to note is that the AI sometimes seems torn between its supply vs objective defense priorities. Looks like something to be exploited.
I'm surprised how much I'm enjoying this game. I've played many operational wargames, but I never stick with them because they lack strategic complexity, tactical options are usually challenging but limited, and once you learn them...they're predicable. I've played the Strategic Command series for many years, but Cater's designs, although among the best, lack what this game offers: the ability to make multiple discrete moves, within the context of a movement point system, with the option to expend combat at any time in the sequence. That, combined w/ this game's ZOC and supply systems, make for a tactical dynamic that's breathtaking compared to many other games. I do hope that the designers are thinking in terms of taking this game system to the next level. It would be nice, for instance to have larger, broader campaigns, that give strategic choices: race to Moscow vs the Caucasus, or something you make up yourself for conquering the Big Bear. And instead of a string of scenarios (not that individual scenarios aren't great), how about simply releasing prestige points on a real time basis as objetives are taken, perhaps depending on how quickly they are taken. Another option would be to award fixed prestige for capturing objectives (real time, i.e. the turn you take them), but the quicker you take them means that the Soviet army is deprived of prestige points sooner, hampering their ability to rebuild units and defend the motherland. Such an approach would add a whole new strategic dimension to the game system.
But the current system does need some improvements. I can think of two that need high priority.
*) Need an easier way to see the enemy supply network. This goes out the window if the plan is to add a fog of war system (I vote no).
*) Need a way to make trial, reversible moves for the purposes of evaluating prospective combat odds. This would of course only be one unit a a time. Again the need is to be able to evaluate combat odds before a move is chosen. Currently the only way to do this is save then make a move and evaluate combat odds. If you don't like the odds, then go through the inconvenience of doing a replay.