A balanced 1vs1 map is not symmetrical

[Rya]

Yesterday something got me really angry and I really want to write about it, to explain people my viewpoint.

To those who don't know me: I'm usually making Wesnoth maps and hosting them to observe people playing them. Sometimes I rebalance them according to suggestions and testing.

There are design-wise three type of maps you can do (X means 2 or higher, but it's the same for both sides):
X vs 1
X vs X
1 vs 1

X vs 1 maps are usually "coop scenario against cpu" but they can also be made for player vs player. How to balance them is not very clear at the start, but with some testing it's possible to make it fairly balanced (of course it can only go so far as the players who play them, since who will win here will pretty much depend on the teamwork on that bigger team). Those maps are never symmetrical, but nobody really complains about that so it's usually no problem.

X vs X maps are pretty straight forward. They can be symmetrical and you do the team order 1-2-2-1 and you create a fairly balanced symmetrical map (of course it depends on your mapping skills).

With 1 vs 1 maps however, symmetrical maps are not a good idea. Especially on small maps it will result in the first player having a huge advantage.
I made a tiny 1 vs 1 map and testing it quite a bit. The first player always seemed to win, so I nerved him quite a lot. The map ended up with the 1st player people able to recruit only two units per turn while the 2nd player can recruit four units per turn. Additional the 2nd player has one more village near him. The rest of the map is symetrical (except for one tile). Even though the map looks like it's highly in favour of the 2nd player, my testing showed so far that it's really well balanced like this. If I let the ai play the map the win/loss rate for each player is very close to 50%/50%. So I started hosting it like this on the official server and the people who played the map seriously till the end agreed that it's very well balanced. So far the outcome was actually that 1st player still won 7 times while the second player only won 3 times, but seeing the matches felt like they were really balanced (even though the map is small, some matches took very look because both were similar well playing).
Either way I encountered the problem that players basically DEMAND symmetrical 1vs1 maps. Often when I start the match someone said that the map isn't balanced because the first player can only recruit 2 unit, while the second can recruit 4. Usually I try to explain the "first player advantage" to them and that they should try to play it before they complain. Luckily people often give it a try and later say "Hey, you're actually right, this is fairly balanced!" which makes me quite happy. Unfortunately, they are also those who don't understand it at all. When I try to explain them the first player advantage they just leave without trying it (and sometimes insult my map).
This is really frustrating for me because it's quite hard getting to players to play your rather unknown map if you don't play yourself and then someone just leaves without trying it at all when I finally got two people to play it. And there's also the annoying bug that I need to rehost the map every time someone leaves in the lobby when the host isn't playing which makes it even harder to get together two people.

So what I want to tell players is that symmetrical 1vs1 maps are NOT balanced. The smaller the map, the greater the advantage of the first player gets and even if a tiny map looks like it's highly in favour of the 2nd player it might end up being balanced because it evens out with the first player advantage.

Think of it this way, if the first player can only recruit 2 units and the second player can recruit 4 units and we assume that the starting money is enough to recruit 6 units then the actual towns taken / total income will be like this:
#Units / Town / Income
-----------------------
1st - 1 / 0 / 2
2nd - 1 / 0 / 2
(disadvantage for 2nd player, because he moves second and all other values are the same)
-----------------------
1st - 3 / 1 / 3
2nd - 5 / 1 / 1
(disadvantage for 2nd player, because he has less income since he has more units but the same amount of towns and still moves second)
-----------------------
1st - 5 / 4 / 10
2nd - 6 / 6 / 12
(here is where the money disadvantage of the 2nd player evens out, but he still has the disadvantage of being second)
-----------------------
Assuming at this point all villages are occupied
-----------------------
1st - 6 / 4 / 9
2nd - 6 / 6 / 12
(here is where the battling for the villages starts, the 1st player still has the advantage that he moves first, the second player has 2 villages more in exchange, this reaches a quite nice balance, because the 1st player will be able to take 1-2 villages in his turn and when it's the turn of the second player he has an similar change to fight to get the lost villages back, also at this point both player were able to recruit all the units they could from their money, so the debuff that the 1st player can only recruit 2 per turn isn't even noticed anymore, not to mention that on small maps making a castle change is quite possible too)

In the end even though it looks like the 2nd player has a big advantage, the map ends up as being fairly balanced. People need to consider the 1st player advantage more, because it can have a quite large effect, larger than you might expect.

[Blarumyrran]

You may note that the core 1v1 maps are indeed not symmetrical. So what is the point.

[Captain_Wrathbow]

You may note that the core 1v1 maps are indeed not symmetrical. So what is the point.

Isn't that exactly what he means? He's saying that 1vs1 maps shouldn't be symmetrical and it annoys him when people demand that they should be.

[Blarumyrran]

Oh, I thought he meant the core maps should be made less symmetrical.

[Gambit]

Or worse, some guy demands that his 15x15 perfect hexagon 6p map is balanced because it is symetrical -_-

[Captain_Wrathbow]

Or worse, some guy demands that his 15x15 perfect hexagon 6p map is balanced because it is symetrical -_-

Sigh... And the worst is when he won't listen to reason or explanation and starts calling you a noob cause you say it isn't balanced.

[Tonepoet]

On the other hand though, having the map become asymmetrical for the purposes of balance can lend player 2 a long term advantage if caution isn't taken as to exactly what's done to fix the matter. The one thing I've noticed about the official maps is that most often player 1 has less efficient village collection, forcing them to make a choice between recruit less efficiently or sapping a good deal of movement points away from about half of their units or simply skipping some villages and hope that their initial income sacrifice gains them a profit in the long run. I think this method is more or less the best of the techniques one can use, since it only has an impact near the beginning of the game to help take the edge off of Player 2's plight and allow them to take control of more or less even ground.

Also, a map could theoretically be perfectly symmetrical but lend itself to the initial attacker losing, in which case the player 1 advantage wouldn't matter and hence be perfectly balanced. At least in the sense that the players have even odds of winning, only changing dependent upon how they decide to play. However this sort of map would be of the sort that also lends itself to inaction since nobody would want to disadvantage themselves by being the aggressor, so such a map would still be inherently flawed nevertheless. At least from the standpoint of good design values anyway. The only official map that ever came close to being quite like this in my recollection was Blitz due to the relatively largish void going down the middle and easy to set up defenses. However due to that same relative defensive ease, it was possible to amass an army at the opponent's gates until you could break through their defenses and gain a stronghold in their territory, so this wasn't quite the case in practice and would still technically lend itself to player 1 advantage anyway. I'm also unsure if it, or any official 1v1 was perfectly symmetrical by the time I started playing so it may not even be all that relevant of an example to begin with. (I'd check the 0.9.X maps myself but I'm too lazy, considering that they're obsoleted anyway. )

[grrr]

Please look at the official 1v1 maps in 1.7. It shows another possibility to nerf the FMA: P2 is simply given one initially owned village (the most awkward one to capture, basically), meaning P2 does not have to send a unit back to grab it but instead send it to the front immediatly. I liked the first post because it pointed out the real problem of FMA: P1 and P2 army size (that is, units in range) during the first encounter. And Tonepoet is right: do not remove FMA by making it impossible for P1 to win longer matches.

Spoiler:

The WML is rather easy - in your P2 side definitions use:

Code: Select all

[village]
    x,y = 17,3
[/village]

where 17,3 denotes the location of a village on the map.

[cool evil]

Maps can be symmetrical and fairly balanced as well, we just need to make it so that no side(s) can capture a village in the first turn and restrict the amount of castle hexes as to force the leader to stay on until turn 2 to recruit more units. This leaves players with two choices:

1) Do i recruit a full castle and then move my leader in order for it to capture a village on the second turn?

Consequences: as a countermeasure to the rush advantage, should P1 decides to take this route, he/she will be left with some unspent gold and the leader would only make it back to recruit more units at turn 4 in the earliest whilst having 1 more village than the opponent who does not utilize this tactic as well. P1 would also have less upkeep as a result of neglecting to recruit said units.

2) Do i stay on the keep and spend all my gold in turn 2, then move my leader to the frontlines?

Consequences: the village that you're supposed to capture would only be taken on turn 3 and may be open to capture by the opponent, not to mention higher upkeep costs. However, you are allowed to potentially recruit these extra units 2 turns earlier and also move your leader 3 turns earlier than than is possible with the first tactic.

The problem with this is that if both players choose the latter strategy, P1 would still have an advantage. This can be mitigated by placing villages apart in the middle of a map or wherever it is most likely for a clash to happen (also following the same theory of placing villages to avoid capture on first turn with a leader).

[ParadiseCity]

Disclaimer:
-I think that symetrical maps can be balanced.
-I think that assymetrical maps can be balanced.
-I do not think that there is currently a significantly large amount of imbalance in 1v1 matches that are symetrical, except in cases of a very well implemented rush. This is merely my personal view that I do not want to get into debating.

However, I feel it necessary to highlight a less noted aspect of being player two, that indeed does give them the advantage when attacking at the right moment.

This advantage shows itself in matches where both players are lawful or chaotic. Very simply, it works like this(for a pair of lawful factions, as an example):

-It is Afternoon.
-Minimal fighting has occured yet, both sides are in roughly the same position of power.
-Player 1 is not in a good position to attack because he would face significant retaliation damage. This is because he would be attacking against people with the same +25% bonuses as him, who would be able to counterattack with the same +25% bonus.
-Player 2, however, is in a better position to attack. He would be able to attack with +25% bonus, but the opponent would be forced to counter with no bonus.

Simply, player 2 has a greater ablility to survive a counterattack than player 1, given that they are equal in strength and positioning.

[Rya]

I don't think that fully symmetrical 1v1 maps can be fully balanced. Even though large maps or defensive maps make the first player advantage very small, it's still an advantage.

It is indeed true that the idea for my map "Give p1 only two recruit slots and one village less" will balance the first few turns very well, but might cause p1 to have a disadvantage at longer matches (though granted, my map is so small, that castle swaps can easily happen). To counter this problem I put the map into a scenario now and an event on turn 3 will make the map symmetrical. That one the first player advantage is balanced and yet from turn 3 on the map will be fully symmetrical and give nobody an advantage in longer matches.

[Blarumyrran]

Doesn't arbitrarily changing a 1v1 map sound like raping the intuition of the players to you?

[Rya]

If I tell all players that it will happen I think it's fine.

[Captain_Wrathbow]

Hmm, that is an interesting way to balance a map... I suppose it would work, but I have to agree with Blarumyrran, I certainly wouldn't like to have the map altering itself mid-scenario. Actually I suppose it depends on what areas are changed. Are you mutating the whole map, or just equalizing the size of the castles?

[Rya]

I just equalize the size of the castles and add the missing village (though I'm thinking to place the village right at the start because it won't make much of a difference, I'd rather have a non-village recovery tile there, but the only one I know is the oasis and it doesn't really fit into the theme).