Defending by running away megathread

[Seravy]

as the fact that you've avoided to answer any of these points

I did answer those you actually asked before but my answer was ignored.

people can see vids on the web of shitty run away to keep a city tactics, making your work look bad

This isn't a commercial game, the goal isn't to make it LOOK good, it's to make it BE good. I'm not earning money from it so I don't care if there are 10 more or fewer players. Their loss, nor mine.

I hope we all agree that the richness of the game, its complexity, is what attracts us to it. Clearly a tactic that is always used presents therefore a problem.

We agree in that BUT here is the problem : This isn't a tactic that's always used. This is a tactic that's always used AFTER something has gone wrong and the main, better tactic fails. A backup plan. A last resort, so to say. A game doesn't need many of those, in fact, most games don't have any. But MoM has two : the FLEE button which allows you to get away with less damage but guarantees losing the city, and running in circles which will result in taking a lot more damage (to the city but also to your units from enemy combat spells) but has a chance to keep the city. It's up to the player to pick one, and running in circles is not always the better option.

Furthermore, the point of flight, invisibility and high movement is to grant the ability to avoid engaging the enemies. That's their primary function. That's why you'd pick those units over the cheaper unit that has the same stats but none of these abilities. Invisibility does come with extras, which is why it's more rare and more expensive, but movement and flight only has one role in combat : avoid getting hit by an enemy (or prevent the enemy from avoiding getting hit). This is a core game mechanic that's mostly unique to this game and fairly important.
Whether you use these abilities to keep cities, destroy enemies with damage over time effects like Call Lightning and Wrack, or simply buy time to cast more spells (buff your unit, summon things, direct damage, etc) doesn't matter - what matters is they can avoid fighting while keeping YOU in the combat so you can do something about it - you have more options to choose from. And the presence of this mechanic makes ranged units more relevant, which is also a nice plus - you can't run or fly away if they are shooting at you - but you can be invisible.

freedom doesn't mean lack of consequences.

Losing up to 75% buildings and population is very far from my definition of "lack of consequences".

if it were a multiplayer game nobody would accept this, let's be honest.

Well, there is a reason why multiplayer games are bad. It's the people's fault - no one likes to lose so if there are winners, they have to be disposed of by nerfing their strategy. The result is a boring game.
But the truth is, if it is multiplayer, the other player can easily avoid this tactic. Catching the running unit might be hard for the AI but a human, especially if they also use their spellcasting in a smart way, can manage. Despite that, the result would be what you say, the less competent players would fail to do that and would demand the game to be changed.

In what version? Because I've just had it happen.

In the next version which is not yet released. There are too many open issues to do so.

This works so well that eventually I started to use it on attack.

Attacking multiple times with the same or different units and casting spells each time to kill defenders and win has been a staple tactic ever since Master of Magic 0.98 and one that makes the game MUCH more interesting and MUCH more playable for human players. Unfortunately the AI isn't smart enough to use it well, which is why humans can beat it on high difficulty despite the resource advantage.

If you look at the flee function, it's a simplistic 50% survive/die for each unit.
We can extend this approach to the flee function and have speed and movement mode (flight, etc) be considered for the odds of fleeing, replacing the simplistic dice roll.

I've explicitly stated before that I'm NEVER changing the fleeing mechanic again because it's that hard to change. It's not always 50% by the way, see the game mechanics thread.

Because if you did, then you'd be playing MoM with the fixes, and summoning a wraith to beat the game.

You don't seem to realize the difference between a tactic that outright wins the game without effort and one that merely reduces the negative consequences of an already bad tactic (why didn't you have more defenders in the first place? Why did you let the enemy attack there? etc) and doesn't really get yourself closer to winning, it merely delays defeat.

If I have a belief, and I can prove my assumption with reasoning starting from bases we both agree on

Our definition of the term "problem" seems to be different, so we didn't agree on all the bases - we missed the most important one. Unfortunately "problem" is pretty impossible to agree on - it merely means "I dislike this feature". Of course, if everyone or at least most players agree they dislike a feature then it's fairly reasonable to say it's a problem, but I don't see the large amount of people saying so.

Well, thread closing tends to prevent discussion. But that's not exactly productive,

Depends. I do believe if I was actually doing some work and getting closer to releasing 5.14, it would be far more productive that this time sink on what no one except you thinks is a real problem. Which is why I closed it, but it didn't seem to have helped.

If it isn't a problem, is it a feature?

Yes, I believe I have said so in the closed thread as well.
I answered all these questions but one last time, I might as well repeat it.

What does it add to the game?

This is actually answered in this post in the first half. Better balance of the melee/ranged units, more value to movement speed, flight and invisibility, a richer game. More emphasis on spellcasting - that's what running is mainly for, whether through buying more turns to cast combat spells, or by buying an entire overland turn to cast most spells in the next battle.

Why do you like it?

Because being able to defend a city and not losing it to overwhelming odds feels good. Because being able to take advantage of all sorts of tactics (movement can be used on offense as you say, or even when defending noncity locations) makes me feel like I'm playing a game where my effort matters and paying more attention (and spending more time on) a combat improves my results.

Would you still like it if the AI did it?

They do it already and they do it because I made is so. Obviously not to torture myself but because I want the enemy to be as good as possible, just like myself. It's not fun otherwise. (well sometimes it is, but one's not always playing for challenge, there are other ways to have fun, too.)

... Can you convince me that you're not just after the cheap satisfaction of beating a game that is famous for its difficulty using a trick that has been by now clearly outed as easy?

Is there a need to? Games are played to have fun. They are played to satisfy the player.
I don't believe this tactic is beating the game though. It helps a little, but what beats the game is your main tactic, whatever it is. This just enhances its performance by, idk, maybe half of a difficulty level? Considering the AI resource advantage, I don't see that as a bad thing, the human needs every little thing they can get to win on Master or Lunatic.

parties are exhausted, your defending units are pushed an adjacent square overland, if available (some or all).

I can't make the defender retreat exhausted, I already said that.

[Nelphine]

You've made a ton of statements.

You have not answered the 3 in my opinion that matter most: 
1) you don't need to use this tactic to beat lunatic.
2) you think this tactic makes the game look bad. I do not. Why is your opinion more important?
3) the AI uses this tactic. Keeping this tactic in existence allows abilities like flight to matter.

As a fourth important consideration: how would you teach the ai how to deal with your proposal in the late game when the human can have a teleporting unit that baits the ai (such as a hero with bad stats) and also invulnerable units (such as 50+ defense heroes) and the ai has only fliers (such as sky drakes) that your invulnerable hero can't kill? 

How does the ai learn to NOT lose their city in that case?

Solving a minor tactic, that you consider a problem in the early game, that opens up the ability to trivialize the end game, is Bad.

[zitro1987]

I'll add more counter arguments, which explain why my favorite adjustment (if any) is a simple gold and possibly fame loot reward for an exhaustion battle based on you reaching the city destruction cap.
*could even trigger fame loss to the defending side, for losing most of his/her city (50-75% loss is a big BIG deal).

So counter arguments:
5a) research has gotten more expensive in later versions - defensive spells like invisibility, flight, or create artifact (movement) must be worth it, especially combined with entangle, web, spells like wrack, to defend cities while possibly dealing damage.
5b) similarly, offensive spells or creatures than can handle it (true sight, fast flying creatures, nuke spells, call lightning) must be worth the research, in order to be more able to take cities.
6) Cavalry has 2 armor, 3 hp. Ranged units and spells can wittle them down. There are not many other units early on that have this role.
7) If you're really attacked, they'll have many surviving units and goodbye nearly everything you liked about your city. Might as well not have it! (my favorite adjustment is meant to reward the attacker for the effort)

[Nelphine]

The ai defends a city with 4 melee units, you attack with 2 ranged units. They're both the same speed. Your ranged units can outrun the AI units as long as they never fire. If the ranged units can attack with all of their ammo, they can kill one enemy unit. Under your system, your ranged units will conquer the city because the AI will retreat to defend the city and so will allow the human units to use all of their ammo without taking any damage. The human will need to attack 4 times, but the AI will lose. With current rule, the AI will chase the ranges units, and either catch them and kill them, or prevent them from ever attacking, and therefore never lose the city.

This scenario is far more abuseable and important than the 'problem' you are trying to solve.


In my opinion, you must always ALWAYS design rules around the human attacking - that will always be more abuseable and directly lead to winning the game - if the human can abuse things and successfully defended, that's not winning - that's just avoiding losing. And usually it's a by product of preventing the human from abusing things to win.

[Nelphine]

Zitro, gold only triggers on a city being conquered. The impression seravy gave is that is a coding limitation.

[Suriname]

The problem is that you bring the same kind of proposal with conquest trigger, which combines difficult code with opening up new abuses that would be prevented with more complex rules and code.

Well, at least suggest an abuse...

If there is an abuse you should also show that it's worse than now, now it's as bad as it gets: have cake and eat it.

My proposal is similar to you #1, in that it fixes non committal/non engagement without consequences, so I wouldn't mind your solution either, but I suspect that it's harder to code than a simple counter. OTOH, I appreciate that you're silently confirming that you find this to be a problem by proposing solutions. Please don't be silent though, I won't convince Seravy because I'm a contrarian asshole.

Idea #2 (attacking rewards, my favorite of the 3)
*Attacking party, if reaching the cap city destruction, gains gold based on what the town has, similar to razing. This amount of gold can be significant for big cities.  Maybe win some fame as well!

I like it, plus it is an additional component to any solution, not an alternative solution. Coding permitting...

*Population destruction cap being 99% rounded down.

That still leaves the cake and eat it. The city closeness to the stack attracts the AI to another month of combat in a half ruin, all while your really important cities thrive and build more reinforcements. In the meanwhile you can cast more save or die spells without consequence. Or with only losing a half damaged cavalry and city that you should have lost anyhow.

[Suriname]

Mmm too many answers.

I can't believe I'm putting this much effort to prove that if you ran away then you ran away.

I'll answer 1 poster at a time...

50-75% loss is a big BIG deal

It really isn't. This trick works best in your frontier towns, not the developed ones, if you're using it in the core you might be done for already. But in the small cities there's nothing to lose.

5a) research has gotten more expensive in later versions - defensive spells like invisibility, flight, or create artifact (movement) must be worth it, especially combined with entangle, web, spells like wrack, to defend cities while possibly dealing damage.

They still are? I wouldn't ever cast those expensive spells for the cavalry trick.

5b) similarly, offensive spells or creatures than can handle it (true sight, fast flying creatures, nuke spells, call lightning) must be worth the research, in order to be more able to take cities.

Sure, so why can't an AI hydra conquer any city?

6) Cavalry has 2 armor, 3 hp. Ranged units and spells can wittle them down. There are not many other units early on that have this role.

The AI targets ranged units first.

7) If you're really attacked, they'll have many surviving units and goodbye nearly everything you liked about your city. Might as well not have it! (my favorite adjustment is meant to reward the attacker for the effort)

Wait that actually supports the concept 

[Suriname]

1) you don't need to use this tactic to beat lunatic.

You don't need wraiths on turn one either, so?

2) you think this tactic makes the game look bad. I do not. Why is your opinion more important?

No, I don't think that my opinion is more important.

But it's defending by running away. It is ridiculous, just try to look at it from a newcomer's point of view.

It's also completely unnatural and counter-intuitive. It's just something that you learn and then keep using. Most of the people on this forum are used to it and considers it natural, including the modder, it's therefore difficult to change.

In short... It has all the staples of a nice and easy trick. But then why was wraiths an issue?

3) the AI uses this tactic. Keeping this tactic in existence allows abilities like flight to matter.

The AI loses a lot more to this tactic than it wins. Flight matters much more for overland movement, and in tactical it ignores walls, which is pretty good. And not counter-intuitive!

As a fourth important consideration: how would you teach the ai how to deal with your proposal in the late game when the human can have a teleporting unit that baits the ai (such as a hero with bad stats) and also invulnerable units (such as 50+ defense heroes) and the ai has only fliers (such as sky drakes) that your invulnerable hero can't kill? 

At that point there's so much casting skill in game that any unit based consideration won't matter anymore tbh. And even then, this proposal is actually friendlier to AI and current mechanics than the current situation:

• the AI is powerless against this right now
  
• it's also unfair to AI as the AI cannot do this: strategic combat always ends in a defeat for one side
  

So even if you're right, then the proposal is an improvement over current situation.
Besides, if you consider the invincible hero a problem then why would it matter only to AIs with flying units? That's a problem for each AI, including land based ones. Or, why would that be ok as soon as the hero finds a shadow weapon? This counterargument doesn't seem to work...

The aim defends a city with 4 melee units, you attack with 2 ranged units. They're both the same speed. Your ranged units can outrun the AI units as long as they never fire. If the ranged units can attack with all of their ammo, they can kill one enemy unit. Under your system, your ranged units will conquer the city because the AI will retreat to defend the city and so will allow the human units to use all of their ammo without taking any damage. The human will need to attack 4 times, but the AI will lose. With current rule, the AI will chase the ranges units, and either catch them and kill them, or prevent them from ever attacking, and therefore never lose the city.

Actually, I've not suggested any change in the logic of leaving or returning to cities. You may be speaking of a problem independent from the proposal?

In my opinion, you must always ALWAYS design rules around the human attacking - that will always be more abuseable and directly lead to winning the game - of the human can abuse things and successfully defended, that's not winning - that's just avoiding losing. And usually it's a by product of preventing the human from abusing things to win.

Except, that in difficult situations you are attacking on one side and defending on the other side.

This trick is for difficult situations, and makes diplomacy pointless. You can attack one AI and use the trick against its ally to avoid facing consequences of bad diplomacy rolls.

I'm not making this up. See the saves. I attacked yellow and its ally - blue - counterdeclared, despite being pacifist. I've successfully attacked yellow while using the trick to waste all the momentum of blue. The city of Canterbury has suffered some damage and has been attacked at least 5 times, one of which by a stack of 9 warbears - that almost got me, that has mostly been dealt with through confusion, despite me having between 30 and 40 skill. All I had to do was keeping feeding cavalries to that city, and they're pretty quick to move there. Red is aggressive too. So I'm happily managing a 3 fronts war. 

At lunatic. 

With humans.

[Nelphine]

Wraiths attack the ai and win. This tactic allows you to keep a city, assuming the AI doesn't have the correct spells to defeat it. By itself, this tactic does not win. By itself, wraiths do win. A misclick with wraiths still kills the AI. A misclick with your cavalry, and you lose the city anyway.

Ever fought Djinn in a node? or air elementals? AI wins LOTS with this tactic. You may consider frontier cities important; I consider nodes in 1403 important. The human is powerless against this tactic RIGHT NOW.

Any mention of quick combat doesn't help your case. Both I and Seravy (and everyone else I believe) agree that quick combat is a problem; however, it is a coding space problem and simply cannot be fixed.

My examples:

Example 1: if fliers can't hold cities, then the most important units in the game, in late game, cannot hold cities. The AI can't deal with this. Regardless of any unit restrictions (and yes, when you both have a ton of skill, units still matter; skill cancels skill, and units end up deciding the victory. As I usually play my lunatic games until 1415ish, I regularly play games where the AI has 600+ combat casting skill) by teleporting into the city on any given turn, I immediately win because you don't want fliers to hold cities. That simply does not work. (And in fact, would stop working as soon as uncommon summons came into play. Flight is only overpowered while the AI can't deal with it - the common tier of the game. By uncommon tier, flight is so common to all players, that it becomes the standard, not the exception, and is no longer overpowered.)

Example 2: currently, if i attack with 2 ranged units, then the ai chases me mercilessly with its 4 melee units. If I stand and shoot, I die. If I run away, then I retreat exhausted having gained absolutely nothing. Under your proposal, if the ai chases me mercilessly, he will not be in his city. Therefore, by a little judicious choice of where to run (of which the human will ALWAYS be better), i can either ensure that I conquer the city if the AI is not taught to return to the city; or if the AI IS taught to return to the city to prevent losing it due to 'not defending it', then my ranged units will eventually not be pressured, and can fire all of their attacks on the AI melee units. Repeat 4 times, and I win the city. In either case, I win the city in a scenario where currently, I could not win the city from the AI. (In effect, your suggestion actually makes ALL ranged units into fliers when attacking, which, is FAR worse than the current topic of cavalry running around; even worse, unlike the current running away tactics, the AI can't be taught how to replicate that at all. The AI can't be taught how to make those exact movement choices that will allow it to potentially steal the city with its ranged units; nor can it be taught to understand that it might be helpful NOT to stand and shoot for x number of turns.)

[Suriname]

Getting rid of some for readibility, if I miss any point repeat it pls

the goal isn't to make it LOOK good, it's to make it BE good. I'm not earning money from it so I don't care if there are 10 more or fewer players.

Obviously false, why are you engaging with your user base? And thankfully so. Because you're human you like company, despite what you think of yourself. You're just treating me badly because I'm a contrarian asshole. But have you read the yom kippur war pdf? Contrarian assholes are useful.

I hope we all agree that the richness of the game, its complexity, is what attracts us to it. Clearly a tactic that is always used presents therefore a problem.

We agree in that BUT here is the problem : This isn't a tactic that's always used. This is a tactic that's always used AFTER something has gone wrong and the main, better tactic fails. A backup plan. A last resort, so to say. A game doesn't need many of those, in fact, most games don't have any. But MoM has two : the FLEE button which allows you to get away with less damage but guarantees losing the city, and running in circles which will result in taking a lot more damage (to the city but also to your units from enemy combat spells) but has a chance to keep the city. It's up to the player to pick one, and running in circles is not always the better option.

Wait. Where's the choice?
- running in circles: no troop losses, some damage to the town but I keep the town
- fleeing: troop random choice, I lose the town.

Furthermore, the point of flight, invisibility and high movement is to grant the ability to avoid engaging the enemies. That's their primary function. [...] but movement and flight only has one role in combat : avoid getting hit by an enemy (or prevent the enemy from avoiding getting hit). This is a core game mechanic that's mostly unique to this game and fairly important.

Sure. But does that mean also avoiding the consequences of avoiding to engage?

Have I said that it's unfair to attack with an invisible unit to cast spells? Nope. Because it isn't.

Or to defend with an invisible unit to cast? Nope.

But if the invisible unit is defending then it's inside the city and it's easy to find. What the trick allows you to do is to keep the invisible unit out of harm's way and not face the consequences of your choice. Have cake, eat cake.

these abilities [...]buy time to cast more spells (buff your unit, summon things, direct damage, etc) doesn't matter - what matters is they can avoid fighting while keeping YOU in the combat so you can do something about it - you have more options

All of which is still possible under the proposal. You just don't get to keep the cake of the city while eating the options that you mention.

Losing up to 75% buildings and population is very far from my definition of "lack of consequences".

No it's not, it's exactly lack of consequences. There are no buildings in the frontier town you captured with a lucky cavalry, you would have no buildings in the town you'd lose. You can sell the few buildings of the frontier town if you see the stack coming. These are all trick perfectionisms, but it works even without them.

This trick is not for your capital - obviously you don't want to use it there. This trick is for the frontier town you'd lose anyway.

if it were a multiplayer game nobody would accept this, let's be honest.
Well, there is a reason why multiplayer games are bad. It's the people's fault - no one likes to lose so if there are winners, they have to be disposed of by nerfing their strategy. The result is a boring game.
But the truth is, if it is multiplayer, the other player can easily avoid this tactic. Catching the running unit might be hard for the AI but a human, especially if they also use their spellcasting in a smart way, can manage. Despite that, the result would be what you say, the less competent players would fail to do that and would demand the game to be changed.

Attacking multiple times with the same or different units and casting spells each time to kill defenders and win has been a staple tactic ever since Master of Magic 0.98 and one that makes the game MUCH more interesting and MUCH more playable for human players. Unfortunately the AI isn't smart enough to use it well, which is why humans can beat it on high difficulty despite the resource advantage.

I'm fine with that actually. You see? I can change my mind when confronted with good arguments.

Can you?

Because if you did, then you'd be playing MoM with the fixes, and summoning a wraith to beat the game.

You don't seem to realize the difference between a tactic that outright wins the game without effort and one that merely reduces the negative consequences of an already bad tactic (why didn't you have more defenders in the first place? Why did you let the enemy attack there? etc) and doesn't really get yourself closer to winning, it merely delays defeat.

Because if you did, then you'd be playing MoM with the fixes, and summoning a wraith to beat the game.

You don't seem to realize the difference between a tactic that outright wins the game without effort and one that merely reduces the negative consequences of an already bad tactic (why didn't you have more defenders in the first place? Why did you let the enemy attack there? etc) and doesn't really get yourself closer to winning, it merely delays defeat.

Right, assume I'm a moron. I don't mind. The trick can be used to protect yourself on one side while you attack on the other for example - see answer to Nelphine. It gives you a parachute from surprise attacks when you're lazy with scouting around your cities. It gives you a last option when you're in lunatic and the AI has x times your resources. Is that a bad trick? Sure, it's not as easy as the wraiths, but it's a thing that works with each magic book and retort choice and most races so one can say that it's even worse than wraiths, as it's really omnipresent.

Unfortunately "problem" is pretty impossible to agree on - it merely means "I dislike this feature". Of course, if everyone or at least most players agree they dislike a feature then it's fairly reasonable to say it's a problem, but I don't see the large amount of people saying so.

So you like having others to speak - "most players" - with after all

No, it doesn't work like that. Players are lazy. They like to win. If you go by consensus you'll never eradicate the easy tricks. The problem is that when they win they get bored. I bet that you are not entirely convinced of democracy - same thing. People votes with the belly not with the head.
The game designer has the harsh job of challenging the players, and of making the game more difficult, and face the complaints - this you agree on, it's a rewording of your opinions concerning multiplayer. Except, that you're the designer this time.

As a consequence, your job with the trick in question is to decapitate it. It's an easy trick that the complaints in this thread want to keep. Or did you write that because you think that I'm the one complaining because I want an easy victory? That'd make no sense based on all that I've said.

If I have a belief, and I can prove my assumption with reasoning starting from bases we both agree on

Our definition of the term "problem" seems to be different, so we didn't agree on all the bases - we missed the most important one. Unfortunately "problem" is pretty impossible to agree on - it merely means "I dislike this feature".

Well I hope I've covered a bit more ground than I don't like this.

The following is perhaps the most important part...

What does it add to the game?

Better balance of the melee/ranged units

Do you realise that the trick only works with non ranged units because they get targeted first?

more value to movement speed, flight and invisibility

The proposal is not reducing that value. Movement/modes are for overland and in tactic they let non ranged units get to melee earlier. Invisibility is just as useful.

a richer game. More emphasis on spellcasting - that's what running is mainly for, whether through buying more turns to cast combat spells, or by buying an entire overland turn to cast most spells in the next battle.

Richer? Something that you do in exactly the same way with any realm, race or retort?? It's not richer to get your cake and eat it. To be more precise:

•  buying more turns
  You still have that. You just don't get to keep the cake of the city too.
  
• Cast more spells in the next battle

You still have that. You just don't get the cake of getting the defensive first turn as well.
So what is the proposal really taking away from you really? Eating the game richness, or just the "keeping the cake" of the easy trick part?

Why do you like it?

Because being able to defend a city and not losing it to overwhelming odds feels good.

Even now that you have been shown that this is an easy monkey trick?

I actually agree with you on the overwhelming odds part. That's why I want to be forced to really be in that battle. I want my cavalry to fight and use their first strike to deal damage, not to run around with their pants on the head. High movement would still be useful, even leaving the city: you could run out and then engage the attackers when they get under the walls and retreat. Think about it. If we got rid of this stupid trick we could even raise cavalry speed! Speed, say, logistics ability, or an ability to attack using only 1 MP, or haste... Just random ideas don't fixate on them, yes these particular ideas are abusable, but what I'm saying is that getting rid of the trick opens up options for you.

Because being able to take advantage of all sorts of tactics [...] makes me feel like I'm playing a game where my effort matters and paying more attention (and spending more time on) a combat improves my results.

Yeah well, we agree. So why are you forcing me to click done 15 times per unit left in the battle instead of efforts that matter?

Would you still like it if the AI did it?

They do it already and they do it because I made is so. Obviously not to torture myself but because I want the enemy to be as good as possible, just like myself. It's not fun otherwise. (well sometimes it is, but one's not always playing for challenge, there are other ways to have fun, too.)

OK, we agree. But if one isn't playing for the challenge then easier game modes can be used. Doing lunatic and then using the trick just takes away the challenge. It takes away from this way of having fun.
The AI can do it? I've never seen that, maybe only with flying units? But not with cavalry for sure. I think. So, you haven't taught this to the AI, in its most easy to use part - cheap cavalries... It's unfair to the AI right now, and taking it away reduces the player advantage much more than the AI's - players can use sprites as well.

... Can you convince me that you're not just after the cheap satisfaction of beating a game that is famous for its difficulty using a trick that has been by now clearly outed as easy?

Is there a need to? Games are played to have fun. They are played to satisfy the player.
I don't believe this tactic is beating the game though. It helps a little, but what beats the game is your main tactic, whatever it is. This just enhances its performance by, idk, maybe half of a difficulty level? Considering the AI resource advantage, I don't see that as a bad thing, the human needs every little thing they can get to win on Master or Lunatic.

In my view it takes away more than that. More importantly, it still takes away half a difficulty level even in your view then. So it's not master nor lunatic anymore. It doesn't let you feel good about beating the game at lunatic - if you manage, you know that it's half a difficulty level less now.

parties are exhausted, your defending units are pushed an adjacent square overland, if available (some or all).

I can't make the defender retreat exhausted, I already said that.

OK, so if I get it right you would be less against the proposal if you could make the defender retreat exhausted but you can't. Well, fixing flee removes the need for that, and it's frankly easy. So, if the fleeing mechanic needs to be fixed first then that's a technical prerequisite on the proposal. In other words, I wouldn't mind leaving this thread open and concentrating on the flee mechanic to fix it first, and then moving back to this. 

But if we can't even agree on the fact that fleeing to hold a city is a problem then I'm not going to bother tbh.