I don't really know how much this has already been discussed, but I thought I'd give it a shot anyway.
I was playing a meeting engagement as the defender, using Mech Infantry-Medium Infantry as the Syrian Army against different combinations of American troops (Stryker Infantry, Light Infantry...), with different combinations of fitness, experience... Engagement size was set to Tiny. Terrain type to Open. The difficulty I chose was veteran.
I almost always got this map (sorry but I don't recall the name, will check later if needed): a rectangular grassy field, with parallel slopes dividing it in... 6 parts, a tiny bunker on the red side and a building complex in the middle of the map, which is treated as an objective (I'm quite new at wargaming, so excuse me if I don't use the correct terms). I mean, the complex gives you points at the end of the game if you keep your troops in it. I sent some guys forward to spot enemy troops. Since there was no one in sight I sent forward the rest of my troops (using UAZs or BMPs). After seizing the building complex, I put all the infantry on the rooftops to look for incoming enemies. Nothing happened until 10-15 minutes into the game, when I started seeing Strykers hiding behind the slopes, infantry slowly advancing... Only a few shots were exchanged before the end of the available time (20 minutes + 4 extra). This always resulted in a victory for me, cause by seizing the complex I was getting more points.
I glanced at the AI plans in the mission editor, and while I'm sure they guarantee a safe advance for the troops, they are executed too slowly to even give a chance to the AI to win (given the time limit).
I'm no expert in wargaming, let alone in mission editing, but I thought that just a bit of fine-tuning on some of the maps could very well help out the whole gaming experience. For example, I don't know how much influence the mission editor can have on the selected troops, but a better balance between opposing forces, at least in terms of numbers, could be good (I wouldn't mind having numerical disadvantage, but 3 or 4 ATGM teams against a platoon sized force is... well ). Another thing to note is the balance between the time limit and the AI planning. In the example I made I could say that 10 or 20 more minutes could have resulted at least in a confrontation between me and the AI. Either that or a more "rushy" AI plan (i.e. less stops before arriving at the building complex). In another map I used the AI effectively advanced and proceeded to hand me my ass on a silver platter.
Oh, I'm using version 1.08, I don't know how things are doing in 1.10.