From: Houston, TX
Evaporation happens whenever a unit loses cohesion. While history is replete with tales of units that fought to the last man, these are unusual cases. There are many more examples of units that simply dissolved, even if only temporarily, under enemy fire. Units, even good units, can only take so much.
In most cases evaporation is different from what most wargamers are used to calling "elimination". Unless the evaporating unit is isolated, all surviving troops and equipment are still available for distribution to other units. Isolated units (those not able to trace a line of communication back to a friendly supply point), on the other hand, are truly eliminated when they evaporate. Their troops and equipment are permanently lost - assumed surrendered to enemy forces.
Time allowing, many evaporated units will eventually be reconstituted from available replacements.
IMO double combat losses are more than sufficient to justify green units to lose cohesion... and evaporate.
I suppose this is what happens in RL.
Evaporation is one of the reasons I love TOAW.
If this happened the same way with attackers and defenders, I'd say that it is working as designed. In this case, I can see the code. If someone is defending, suffers mild losses and fail their quality checks, they are going to divide into three pieces and retreat. If the attacker fails his quality checks he was going to evaporate, now he's going to reorganize.
I agree with Sekadegas that we do not want to shield the attackers from deserved evaporations. But if Ralph has found a bug that causes undeserved evaporations, we do want to fix that. And my tests indicate that he has.
To do the test, I just sort of reversed my "Test Terrain" scenario. Instead of the big American units attacking the 1/3 sized German units in their various terrains, I set up the attacks by the Germans against the Americans. So all attacks were at 1:3 odds, with all defenders in clear terrain. This should give lots of chances for Attacker evaporations. There were 100 attacks altogether.
The first set of tests were with the AD still at 40. At that AD setting, the attackers weren't suffering enough losses to justify any evaporations. But several still did under 3.2 - none under 3.4. Similar with the AD lowered to 10. But when the AD was lowered to 6, the evaporations went up in both cases - indicating that they were now mostly deserved. Finally, at AD = 4, most attackers evaporated regardless of version. I think this shows that there was a bug where attacking units were suffering undeserved evaporations, and it has been fixed in 3.4. Attackers clearly will still evaporate when they deserve to, but only when they deserve to.