Yeah? That's how chokepoints work.
But you guys have a point. Thankfully, we aren't only able to decide between the options of 'all armies retreat', which will lead to permanent stalemates(NOT the point of it; stalemates but not infinitely long ones) if between two forces with enough HQs and chokepoints/defenses, and 'one army retreats' which won't actually make a difference, really, and is basically just a way of saying 'defenders win ties, but not as much as actually winning).
So, I will put forth two more ideas for how it could work, closer to the centerpoint.
The first: in combat, if the rolls are tied, the attacking army that tied and one additional attacking army Retreat, no longer being considered in the battle, but not destroyed.
The second: In combat, if the rolls are tied, the attacking army that tied and half of the total remaining attackers forces(rounded down) Retreat, no longer being considered in the battle, but not destroyed.
Variations on these are certainly possible, such as the first idea causing 3 armies to retreat, or the second idea causing a third of the attacker's forces to retreat. ((Personally, I like the 'one-third of total attacking forces(nobody actually gives a shit which army tied) Retreat', though maybe it should be whichever one causes less armies to retreat, X or Y)h
Additionally, I have a possible addendum to this, for the first idea, or really any of them, possibly. When combat ties, the Defender's army that tied Falls Back.
Which is basically Retreating, but if the Defender's end up losing, rather than being destroyed, the army is moved to an adjacent friendly province(whether it's automatic in the direction of the Nation's Capital or the player's choice is another question).
Finally, on a similar trend, I feel that not all of a defending force should die if the attacking force ends up winning(I don't think that all of an attacking force's troops should die if they fail, either, but that's more or less covered by the Retreat rule, though something similar to the following idea could be used as well). For this reason, I would put the following idea up for debate.
When a defending force loses, half/one-third(rounding up) of the original defending forces(not counting those that have already Fallen Back) Fall Back to friendly provinces.
Next Bit(to avoid scenarios where losing a large defensive stack actually means you have more armies left than if you had won with heavy casualties):
In order to be balanced(for reasons I will explain in a minute), this needs to apply to both Attackers and Defenders. In total, it would mean this: When an army stack wins a battle, one-half/one-third((preferably one-third, at least for attackers(defenders could well be one-half, but it's far more up in the air); in this way there could be such a thing as a pyrrhic victory)) of their casualties are restored(possibly only during the player's next turn, though, as it takes time for armies to regroup, and gives the other player a chance to counterattack). Fluff-wise, this can be explained as the scattered elements of the army regrouping and reforming, bits and pieces of the various defeated armies getting formed into new regiments and the like, new recruits coming in from refugees or inspired/fearful citizens encouraged/forced to rethink which side they're on after the victory/defeat. This has to be balanced, however, by this other next bit.
When an army stack loses a battle, one-half/one-third((preferably one-half; see above) of the casualties from the battle are restored. Why is this larger than when they win? A couple reasons. Fluff-wise/thinking about what's supposed to be going on in the game, a lot of the reason one army would beat another, particularly in such large numbers, is less a matter of who has more bodies to throw(WWI showed us that doesn't work all that well on small fronts; WWII showed us it can work just fine on large ones, however, and this is already handled by the game mechanics, in some ways) and more a matter of tactics and organization. Yes, technology plays a role, as having aerial superiority or better tanks or radios or what-have-you means you can maintain an advantage in some fields, but that too boils down to being able to disrupt key elements of the opposing force while keeping one's own forces intact. The rest are just tools within that. And just like having a better sword than your opponent can make it easier to beat him, it's still a matter of how skilled you are with the sword, and how skilled they are. As for gameplay balance reasons, it's to avoid the inevitable counterattacking stuff that would go on when someone threw their whole stack at a defending force, lost all of it, and the defending force blew through his provinces afterwards. Why have this at all, though? Simple. To avoid what we all end up dreading, for one thing: they break through your primary defensive line, and after a blitzkrieg, you've lost half your provinces; rather than being put in a disadvantageous position by having them hold the bottleneck rather than you, and being forced to spread your forces out among the provinces bordering it and so on so that it creates more room for strategical maneuvering and a bit more breathing room for diplomacy
Speaking of which: Additional Ideas/Notes! First, the reason why it has to apply to both winners and losers, Attackers and Defenders, in order to be balanced. For one thing, Attack is inherently super-buffed in this game, late-game. Which isn't the point of the idea of 'the best defense is a good offense'. The point of that saying is the way momentum in a battle/war works and the fact that the attacker chooses when he attacks and sometimes where he attacks(chokepoints/bottlenecks are a defender's best friend for that very reason). Second, I should say why I want it at all: 100% casualties for the loser are not realistic. In reality, the only time that usually happens is in small isolated 'fighting to the death' scenarios, when they have nowhere to go, or something similar might happen if they still have nowhere to go and end up surrendering. Even then, an element of the defending force might break through their line and try and survive a bit longer. And as for attacking? You might commit all your forces to trying to break their lines, but it's not like retreating isn't a possibility. And nobody wastes all their forces continuing to try to break through something that simply can't be broken through in that way. Definition of insanity. At a certain point you stop, retreat, regroup, and come at it from a different angle. And again, for defending; Germany didn't get conquered immediately after D-Day. They fell back and presented a series of defensive lines which had to be defeated one by one. They burned bridges on their way back, destroyed railways, and in general practiced at least a little scorched earth policy, though not to the degree the Russians did.
Oh, and, uh, third: because right now, with armies spawning at the beginning of the 'month', being farther behind in the turn order gives you an inherent advantage, at low power levels, for defending and counter-attacking. Example: I'm Defending province #1, and I go last in the turn order. If I get attacked by guys from province #2 and they go first in the turn order, I can attack immediately after they did, taking the empty province and possibly other stuff if there's a railroad, even if it had a miltary HQ, Airfield, Fortifications, and AA batteries. Meanwhile, I'm attacking province number five, which belongs to a third faction, who go second in the turn order. I fail to take the province, but since I have an HQ in the province I attacked from, I get an army spawned there at the beginning of the next turn, before they have a chance to counter-attack, and they can't take it without a fight.
That said, time for additional rules ideas, mostly optional!
First off, rather than doing fractions for the Retreat/Fall Back/Regroup calculations, we could have more randomness(sorta) and instead have, after every fight, a Casualty roll for every army that was destroyed in the fighting. If it's a five(or 6 or 4 or whatever numbers, doesn't really matter) or better, they survived, and Retreated/Fell Back/Regrouped instead. You might get bonuses/penalties based on whether you won or lost, or were Attacking or Defending. Certain techs might give you a bonus to it. Certain techs might give your enemies a malus to it(coughChemicalWarfarecough).
Second off, I believe Airfields can support army Already in the rules, actually, more or less. Airfields...heh. But what isn't in there: Having AA batteries should only cancel one Airfield's support of the enemy units. Having AA batteries as well as Fortifications cancels two(fortifications by themselves do not cancel any, however). Having Long range Radar with an AA battery in the province cancels another one. However. Those negated Airfields still have an effect; they can still counter enemy Airfield support(Having Long-Range Rader means they counter one less than the number of negated Airfields supporting the Attacking army; not having it means two less(this could be switched around a bit so that it was whether the defender had Long-Range Radar that mattered). At a maximum, only up to Three Airfields should be able to support an Army stack(only so much space in the sky...). What this means, as I'm probably explaining things terribly:
((also, as a quick note, in this case, Negating and Countering are different, as Countered Airfields cannot Counter anything other than the Airfield they're being Countered by, whereas a Negated Airfield might still be able to Counter))
Faction B attacks Faction C. Faction C has AA batteries in the province, but no fortifications, and one Airfield in the province to give support. Faction B also has three or more Airfields in or around the province to give support. Faction C's AA batteries negate one of Faction B's supporting Airfields entirely, but one of the two remaining Airfields counters Faction C's lone Airfield. Total: Faction B's attacking Armies get one Airfield of support.
However, what if Faction C had two Airfields of support, and both sides had Long Range Radar? In such a case, Faction B would have two of it's Airfields negated, and one of those negated Airfields would still be able to fight Faction C's. All told, neither army would get any support.
But let's look at a fully upgraded case, shall we? That's what we really care about, after all.
Faction B still has three Airfields this time, as well as Long Range Radar. Faction C has Long-Range Radar, three Airfields, Fortifications, and AA Batteries. Their AA batteries, together with Fortifications and Long-Range Radar, negate all three of Faction B's Airfields. However, two of Faction B's Airfields are still able to fight two of Faction C's Airfields, resulting in Faction C's defending army receiving only one Airfield in defense. If Faction B had not had Long Range Radar, then Faction C's defending armies would get two Airfields of defense, instead. If Faction C had not had Long Range Radar, then Faction C would still get one Airfield of defense; perhaps this means that rather than going top down, it should go bottom-up for determining whether negated airfields can still fight enemy airfields(rather than 'two less than the number of negated airfields...', it would read 'up to one of the negated airfields...').
Strategic Bombers should be able to bomb up to two provinces away(three with LRR), not just Adjacent, but not past an enemy province if it contains AA batteries(border-province Airfields cannot bomb enemy provinces past the one(s) they're adjacent to if it(/they) contain(s) an AA battery). Only up to three Bombing attempts can be made in a single turn by a single faction against an individual province; this includes any ones auto-cancelled by Airfields. Mountain Passes/Water Crossings count as an additional Province.
Outnumbering an enemy army should work in two ways, and the number used is whichever one gives a smaller bonus to the larger force.(aka at small army stack sizes, method 1; at large army stack sizes, method 2) Method 1: For every three armies more than the opposing force, +1 to Attack and Defense rolls. Method 2: For every 50% more armies than the opposing force, +1 to Attack and Defense rolls. The exception is this: if the enemy only has 1 army in the stack, and you have 3, you get +1 to Attack and Defense rolls.
I'm wondering if there shouldn't be either a rule, nation trait, or tech that causes every two Military HQs(Capital doesn't count) your nation has to add 1 to Max Army Size.