We played a WW2 game using "I Ain't Been Shot Mum!" (we'll just call it IABSM) yesterday. First impression was that it makes for extremely volatile games.
Most of us learned to count the odds playing the generation of wargames that came from the 70s and 80s. Whether it was Avalon Hill, or Steve Jackson Games (the first wargame I played was Ogre and G.E.V.), they now (in retrospect) shared certain similarities - known and calculable odds for resolution of events, and fairly transparent game mechanisms, via outcome tables and charts.
We learned to count the odds, applying our growing knowledge of secondary school maths and statistics to the games (most direct application of education ever, really). Game decisions were dominated by probability concerns: you went for the course of action that had highest probability of success. Graduating from tactical thinking to higher order, game-winning strategies, involved considering bigger probabilities - going for the result that would put you in a position more probable to win the game, rather than just winning a combat.
There's been a move away from these deterministic systems, mostly by gamers who dislike the mathematical precision to which these wargames can be reduced, and I see IABSM as one of these games. IABSM uses a card-based activation system, coupled with an umpire's discretion to add special event cards, to mix up the game a bit. Of course, any game that involves rules is liable to number-crunching, but card based activation games seem less liable to it for the moment - probably because the numbers aren't fixed (varying based on the umpire's choice of what cards to add in), and because there's always the "Tea-Break" card that reshuffles the entire deck. Probabilities are much harder to calculate here - compare this with what people have done for DBA, for example, or G.E.V.) - although I'm sure somebody out there can do it. Perhaps it's only a matter of time before, in a sort of intellectual arms race, someone publishes an article on playing the odds in card based activation games. Gamers who dislike number crunching can either write new rules with more obtuse mechanisms (so you can't figure out the maths behind it - which often means there's no consistency, which most people agree is a Bad Thing), more randomness (the game is designed to swing wildly, discouraging number crunching since there's no longer any point) or to go back the old tradition of free-kriegspiel (where the players give verbal orders, without knowing the mechanism the umpire uses to resolve conflicts - in other words, the mechanisms are opaque to the player).
Our first IABSM wasn't really representative, but it was demonstrative of the extremes to which the game can swing. I commanded 3 platoons of Russian infantry, 3 tanks, a mortar and asniper, while my friend commanded a similar-sized German force. At crucial points in the early game, the Germans "lost a turn" because the "Tea-break" card was drawn, forcing a reshuffle. The Russians got to act, while the Germans didn't. Despite the umpire compensating by loading the deck with special cards for the Germans, the Russians had got an overwhelming advantage by the middle of the game. The final ratio of German to Russian cards drawn wasn't that disparate, and the Germans did get some luck in the form of a counterattack, but the early Russian luck gave a definite advantage: by the middle of the game, the Russians were squatting on 2 out of 3 victory condition objectives.
The danger of having a "trump" probabilty (in this game, the card that forces an end of the turn and a reshuffle) means that there's an over-riding probability of an event that neither player can influence with tactics, or gameplay. This means that no matter what your tactics are, you have to contend with the fact that X times out of ten, it won't happen because of the trump. Whether this has a 'positive' effect on the gaming experience depends on where your're coming from, and what your idea of a 'good game' is. If you prefer games to be 'realistic', and consider confusion, fog of war, and messiness to be representative of actual combat experience, then this sort of randomness will suit you: if, on the other hand, wargames are mostly games to you, systems that can either be good systems, or flawed systems, then games like IABSM might not be up your alley.
As for me, I've always believed that a single gameplay experience isn't all that informative, so I'm looking forward to a few more games of IABSM. Besides, my Russians haven't had an outing in years ..