Once more unto the breach with assaults, gents!
Has anyone ever seen troops actually surrendering to an opposing force? I can say I have now -- twice in the space of a few days. At first the event caught me off guard, as a defeated enemy unit I'd just attacked suddenly seemed to vanish without a trace. No message, no sound... just 8 guys going poof, and yet according to the on-map results I'd managed to inflict only 1 casualty. Looked for them everywhere but nothing, not even weapons on the ground. Couldn't make sense of it at the time and moved on...
Several days later it happened again after another successful attack, and again the enemy squad went missing. But this time a dim light bulb in my head went on. Could it be that I was witnessing the effects of the surrendering mechanic mentioned in the manual?
The information contained therein is rather vague:
"Demoralized defending units that have no valid way to retreat are eliminated and considered to have surrendered. This does not apply to Japanese units however." (p. 59)
So what does "No valid way to retreat" mean, exactly? I interpreted –like folks in
this thread– that retreating units could not retreat into hexes blocked by terrain/map borders or hexes that contained active enemy units. But my vanishing enemies disappeared under none of those conditions. So something was up. It took a few basic tests to discover what seems to be another undocumented mechanic governing retreats and surrendering.
Assuming the game still applies the standard procedure used to determine retreats (two morale checks, one against the defender's and another against the attacker's), here's what I've gathered so far:
a) As is by now common knowledge, Demoralized retreating units can and will ignore the "can not move in the direction of the enemy" rule. They will move in their general direction with no problem.
b) However, they will not move into ANY HEX THAT'S ADJACENT TO AN ACTIVE ENEMY. Those are the hexes that seem to trigger the surrender. So what we have here is an effective ZOC exerted by enemy units for the purposes of determining surrenders and retreats.
I was pleased by this, as it seemed to favor encircling tactics that are IMO diluted by the lack of isolation rules or facing for infantry. Alas, my satisfaction was tempered by the following discovery:
c) Defending infantry units will generally* NOT retreat if this means surrendering to the enemy (i.e., moving into an enemy's ZOC).
So this introduces a new wrinkle, again not mentioned in the manual AFAIK. As it turns out, attacked units that would otherwise retreat into an invalid hex or an enemy ZOC will instead become entrenched and steadfast.
Consider the following screenshot. Russian forces (Quality A) are attacking four demoralized full German squads (No Morale) from the right. According to the retreat procedure, and provided at least one casualty is inflcted, the Germans should retreat after each attack. What happens instead is that the topmost German unit will always move into the hex marked with the yellow arrow, but the other three just won't budge, because that would mean moving into the ZOC of either the Russian truck or MG unit on the left.
(*)So far I haven't got those units to retreat in any of my tests. But there must be a way to do so, for how can surrenders be triggered if not?
Assuming this is an undocumented mechanic and not a bug, it leads to an interesting tactical corollary:
++ If you absolutely need to dislodge enemy from a valuable hex (e.g., an objective), make sure to leave an escape route open for the fleeing troops! Otherwise they're likely to remain there and fight to the last man. ++
Is this common knowledge, has this behavior been observed in the past? I will keep digging and report updates if I get new information.