Balance without a Mod? A Discussion (Long)

[Velusion]

Gandalf Parker said:
Right off the bat Im thrown off of the "discussion of balance" by the fact that you didnt define those parameters. I would think that the other discussions on it would have shown you that there tends to be a wide difference of opinion on what an "average game" is. And you cant balance for the small-map PvP blitzers, and the mid-map VP games, and the large-map solo games.

Standard Game = defaults settings when you create a game. Medium is the average between small and large (hence it is the average). Those are the settings I'm interested in seeing as far as envrionment variables. Since the designers designated them as defaults and averages I would also assume that these settings are what they consider to be the "average" game. I don't think that is an unreasonable asssesment.

Gandalf Parker said:
Once you pick a game-type then a starting point is usually to put all of the AI into the game with a score graph output. Then set it to process turns every 10 minutes for a couple of days.

I'm interested in multi-player - not AI. All that would tell us is which nation the AI is coded to play better against other AIs.

[Xox]

yeah, again I have to agree with velusion, AI play will not be too informative as to the relative worth of the nations in mp play.

And he already addressed the differences between map size game types and settled on the mid sized map game.

Let's all read through and hopefully keep the need to repeat ourselves to a minimum.

I still find myself amused by the use of the phrase "openly contemptuous". And I would call upon us all to try to work it into all our posts in this thread from now on. Not attacking anyone, just for fun.

And so, in that spirit, I would say that I am openly contemptuous of the ability of the AI to predict relative worth of the nations in mp play.

[quantum_mechani]

I think this idea is a pretty promising, if you could really get semi accurate nation-design-point-costs that would be pretty interesting. But I do have my doubts that issues like the one KissBlade pointed out, and the average experience level of players, could achieve that.

And, of course, balance between nations is a drop in the bucket compared to all the other kinds of balance issues there are.

[Velusion]

quantum_mechani said:
I think this idea is a pretty promising, if you could really get semi accurate nation-design-point-costs that would be pretty interesting. But I do have my doubts that issues like the one KissBlade pointed out, and the average experience level of players, could achieve that.

Well to address these doubts:

KissBlade's assumption I think is founded on the fact that there is an exact perfect number of points needed to supply a dual bless strategy (F9/W9 I assume). There is more than a simple bless stratgey that goes into a pretender. Other things you need to factor in:

Pretender Form
Dominion Strength
6 Different Scales
Extra Magic
Awake/Dorment/Imprisoned

Kissblade's assumption doesn't factor these in at all. It assumes that there is a nation out there that is worth dual bless with the very worst scales, the worst pretender, no extra magic with a 1 Domain. Are we saying that Helheim with ONLY a W9/F9 and the very worst in everything else can beat a Vanheim in a better position?

Personally I think it is statistically implausible if not impossible that two people would agree on the exact maximum bid on a pretender for any nation. Dominion is a huge swing... how low would most go? What about scales - can you take them ALL bad? When do you throw in the towel and decide another nation is a better deal?

At first it looks like all you need is a dual bless, but people often forget the minimum foundation needed for a successful pretender. Dual bless isn't enough - you need, at the very minimum, SOME dominion. This is where the disagreements will occur and the bidding will be interesting.

And some people are not exactly even bless-only fans - they won't even consider bidding that high!

As far as experience - the good thing about bidding is that provided you have a handful of experienced knowledgeable players they will keep each other in check - even if some are not as experienced. An experienced player will not let a perceived low bid go unmatched. Multiple experienced players will make sure that no one person takes advantage of the less experienced players through checks and balances.

quantum_mechani said:And, of course, balance between nations is a drop in the bucket compared to all the other kinds of balance issues there are.

Of that - I personally agree. I've looked at your Conceptual Balance mod and liked (for the most part) what you've tried to do. I suppose I'm a member of the camp that likes more options. Some might call the changes arbitrary - but its a mod, not the gospel. I hope to play it someday.

But at least a reasonable community evaluation of the relative worth of nations can be used now - by everyone - without a mod. I know that if such a community valuation existed it would helpful to newbies... not as a hard rule (it would forever be debated), but as additional informative data.

Thanks for the comments!

[lch]

Nice idea, but I think what this will effectively do is make the game harder and more inaccessible for new players again, upping the already very steep learning curve even higher.

[Velusion]

lch said:
Nice idea, but I think what this will effectively do is make the game harder and more inaccessible for new players again, upping the already very steep learning curve even higher.

This experiment/system isn't designed for new players at all - so I'm not sure what you are talking about...

Velusion said:
So who plays? Bidding naturally favors those who are familiar enough with all the nations to a point that they can make experienced evaluations of the worth of nations. Also for the test to be good, you need players who can utilize whatever nations they have. The closer that all the involved player’s skills are the better the test should be. So in that vein it would seem that experienced players would be desired. How experienced? I’m not sure. I would *guess* that Dom2 experience would be a plus, that 10+ multiplayer games would be a good minimum and often really experienced players can identify others so a “referral system” might be good. Obviously trying to limit a game to experienced players could be controversial. In the end it might simply be that the top few well known experienced players who are interested in playing each invite one or two people to the game, and those players in turn invite others. We would have to trust that players do the invitations based on their assessment of the skill in other player, rather than personal preference. I’m not entirely sure I should have a hand in determining who plays as I am certainty not experienced enough to take part in the game itself.

[danm]

One thing to keep in mind when assessing results from this format is that you might eventually be determining which nations can do best with limited design points, instead of which nations are "best" overall. The effect probably wouldn't be too pronounced, but there are certainly nations that are less affected by pretender/scale choices than others.

So, while for "teh ubar" nations, you may get a hard bidding limit at a dual bless barrier (for example) you might have other "less powerful" nations that get bid up extra high because they just don't "need" the points as much, and people are more willing to bid them up rather than accept a single-bless hellheim or whatnot.

Other than that, I think it's a wonderful idea, and I'd be fascinated to see what kind of stats would come out of it. Even if you never get enough games in the system to provide statistically significant data, it'd be an interesting exercise.

[Velusion]

danm said:
One thing to keep in mind when assessing results from this format is that you might eventually be determining which nations can do best with limited design points, instead of which nations are "best" overall. The effect probably wouldn't be too pronounced, but there are certainly nations that are less affected by pretender/scale choices than others.

So, while for "teh ubar" nations, you may get a hard bidding limit at a dual bless barrier (for example) you might have other "less powerful" nations that get bid up extra high because they just don't "need" the points as much, and people are more willing to bid them up rather than accept a single-bless hellheim or whatnot.

Other than that, I think it's a wonderful idea, and I'd be fascinated to see what kind of stats would come out of it. Even if you never get enough games in the system to provide statistically significant data, it'd be an interesting exercise.

I don't think the expereinced players are interested, so I'm not going to be running this anytime soon.

[jutetrea]

Interesting topic - I think it would get the most bang for the buck out of having a (hypothetical) full field of players. If you're option is to bid the maximum for your "best" build is overtaken by someone else you drop down to your "next-best" build and so on till you land. How much do you spend "NOT" to take a bad nation?

Some players view a basic nation as better irregardless of pretender and may keep just enough points for scales, or just enough for an early SC pretender they know will fade late game.

Maybe this would work out with some sort of mod that actually rewarded players for taking generally considered "worse" nations - maybe a 80/20 or 90/10 split. Actually give extra points to those willing to take MA agartha.