Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Il Palazzo]

Come, now. We're not gonna give the game to Azthor just because some rando graph on the internet tells us to, right? For all we know, the data was compiled >completely objectively< by Big Fomoria. Follow the gold pieces, d'ya know what I'm sayin'?

[E. Albright]

I can't say I find the graph particularly interesting. It's suffering from the competitive Dominions community's fundamental brainworm: the sample size is too small to be treated as significant, and this drives a pile of conditional variables to be treated as independent to artificially inflate it. How well nations perform in a given game is dependent on the map they play on, scales/bless/pretender taken, the win conditions, the other nations in the game, the number of other players in the game, the player playing them, the players playing the other nations,  random map placement, how much experience each of those players have in general/with that specific nation/against the nations they're facing, any particular metarules in place on the game, the perceptions of comparative national & player strengths & weaknesses, etc. Oh, and let's not forget: what version of the game was played, what mods were used, and what version of those mods. To make things worse still, there are unacknowledged feedback loops at play as well - seasoned players versed in the meta-expectations of the community won't play nations that "aren't competitive" or that "are noob traps", which means over time the people that play "bad" nations will tend to be less experienced and thus lose more, reinforcing the notion that they're bad, etc. And to pile on to that, once we introduce mods, we have mod design being fed & reinforced by community perceptions. [Edit: it probably doesn't need said, but these perceptions also feed & reinforce map design & selection in a similar manner.]

All of that isn't to say that some nations aren't just plain better than others, it's to say that a graph like this isn't interesting b/c of how much data it omits in the interest of reaching an easily-quantifiable unequivocal universal conclusion. Which, again, is a problem that's been recurring in the competitive Dominions community for as long as it's existed: they want to treat this as if it's as easy to compare nations in this game as in games with far more streamlined variables - and even more so, ones with more data points. Quantity smudges out a lot of the above problems, but a sample size this small (and stripped of all data but nation + wins) makes this uninformative - again, not least b/c it does not even tell us if we're comparing the same nations since mods profoundly change how the game is played and how strong or weak given nations are. And that's the problem with collecting data for something like this - there aren't enough games available for us to compare only games where even just the mods used is stable. So we end up with what we have here: disparate data points being treated as interchangeable in order to drive up the sample size to a number that doesn't look completely arbitrary.

tl;dr: you can make an analysis like this, and it can be interesting, but not w/o a lot more information about the source data set.

[E. Albright]

Alternately, yes, it may have just been compiled by firbolg statisticians desperate to please Big Fomoria.

[Azthor]

I would guess it's like, the average number of wins you'd expect them to have given the games they were in if everything was totally random? Like if you play in a game with 9 people, that's 1/9th of an expected win?

Edit: This is some interesting info. The bottom and the top sorta are what one would expect. A little bit surprised by some of the more middle things though. Pan being so low is the big stand out surprise to me. As well as maybe Vanheim. I guess maybe they are such classical examples of the "powerful sacred" nations that there's some combo of everyone trying to prepare for them in nation making/dog piling them/being attractive to newer players and less attractive to experienced players bringing them down?

Helheim's position should be mostly explained by its ignominious bless rush that has virtually no counter early on. Blessed mounted elves in numbers are ridiculously hard to counter without any research: decreased defense penalty, multiattack, glamour, and 19 base defense, on a good MR chassis.

And that is the main reason why, I venture, Vanheim does so much worse, though both have access to good battle magic: helhirdings play much better into the combination than vanhere, as a bless chassis.

Come, now. We're not gonna give the game to Azthor just because some rando graph on the internet tells us to, right? For all we know, the data was compiled >completely objectively< by Big Fomoria. Follow the gold pieces, d'ya know what I'm sayin'?

Alternately, yes, it may have just been compiled by firbolg statisticians desperate to please Big Fomoria.

It is alright. I've never played a proper game of Fomoria before this one, even in SP. And my experience with giants and elves is extremely limited, the same being true, to a lesser extent, with air mages.

I know how Fomoria is supposed to be played, but lack the experience to accurately calculate risks, and hence continuously err on caution instead of playing hyper-aggressively like a meta Fomoria would; because Fomoria is extremely unforgiving of miscalculated risks, to the point it can quickly turn unplayable.

That lack of confidence is also why my policy on being ganked early game is that I am accepting my death dishonourably and making sure one of the attackers has their game ruined with mine; it is a little cheap, but it is the best deterrant I have.

It is also 100% Fomoria lore-friendly.

[Karlito]

Surprised by how few of you I've found, even with my very poor first 6 turns of expansion. This map is big.

[Shootandrun]

Surprised by how few of you I've found, even with my very poor first 6 turns of expansion. This map is big.

We were unlucky with the distribution of players across spawn points, I think. Despite being, I suspect, one of the smallest players in terms of provinces, I border four people and have at least six capitals within six provinces of my own cap (maybe even seven). I suspect that the biggest continent and the far eastern isles (eastern from the Agarthan POV!) are more sparsely inhabited.

Relatedly, I somewhat regret meddling with the map spawn points: in my test, the two spawns closest to Agartha were not normally simultaneously occupied. I thought that was the game balancing out player distribution... But no, it was just dumb luck! This time, three players got placed in a space normally meant for two. Sorry, neighbors 

[Il Palazzo]

Yeah, there are always going to be a few players luckier than others on any map with fixed starts and less than the maximum number of spots filled.

[Karlito]

Hmmm yeah. A nearby location that has been filled in some AI tests looks to be empty this game. Hopefully all you squished together people can put aside your differences and gang up on me!

[E. Albright]

Surprised by how few of you I've found, even with my very poor first 6 turns of expansion. This map is big.

At the risk of mounting my favorite hobby horse, it's because the map's provinces have a low average connectivity. There's a lot more 2 & 3 connection provinces than you typically see in modern maps, which increases the time you need to reach (or see) other parts of the map and effectively increases the map size dramatically w/o increasing province count. By contrast, the ocean provinces (well, the coasts, anyway) have much higher connectivity (which is normal and a design assumption in how unit stats are assigned - the oceans on this map are much "bigger" than an outlier map like forex Valanis) and as a result my scouts have physically seen all but two nations (and I know where they are)

You are by far the most remote nation, BTW. The two start points closest to you both turned into thrones so there's A LOT of unclaimed real estate between you and your neighbors (who yes, are all piled on top of each other).

[Il Palazzo]

For everyone's information:
ThtblovesDF is currently unable to access his Bay12 account, so for the time being communication with Niefelheim via in-game messages only.

[Azthor]

ThtblovesDF is claiming to be unable access his Bay12 account just as he breaks a border agreement and initiates a war without forewarning.

I personally cannot see how someone would be able to run Dom5 and send emails but not access Bay12 on any account.

He started our first contact by making one-sided demands. He claimed this was his first MP game, but he was a player in the three first games of Dom5 on Bay12, and I don't believe he could just completely forget that to the point of believing he'd never played Dom5 MP.

He has gone for a dompush build, aggravating all of his neighbours, he was apparently known for backstabbing overly easily in previous games, and he has been literally falsifying communications between himself and his neighbours to other neighbours.

I say all that to one simple purpose: he will try and contact every single one of my neighbours to try and get them to fight the war for him.

When he does, I simply want you to consider this: do you want that person, in charge of an enlarged Niefel, and their dompush, as your neighbour?

That aside, I also warn I will leave Niefel alone even as they siege my capital and route my entire army against the first person to join him in the conflict.

It is not a matter of spite: I can't win a gank, so I am commiting to a doctrine that makes it undesirable.

Edit: I've read his direct communication now; I do not recall creating a new Bay12 account to require mod approval, but if it is somehow the case nowaday, I do take back the very first accusation.

[Shootandrun]

Hey all. Related to what Azthor has just said (but not just), I have received pretty convincing evidence that Niefel (1) has broken a border agreement with Fomoria and (2) has lied to me (twice, as far as I can tell!). And then there is the "first multiplayer game" thing. And allegations he broke NAPs in previous games...

Since ThtblovesDF can't receive PMs but can read the thread, I may as well state this here: I consider the NAP we had together void. I won't be held to any agreements which I don't have reason to believe he considers himself held to.

Diplomacy with everyone else still goes! Just don't lie, or if you do, don't get caught.

[EuchreJack]

I thought we used to have rules about agreements.
Something like "Break Agreements and you don't get to play anymore."

But then again, if no resources changed hands, maybe that doesn't apply. Oh well, not my problem 

[E. Albright]

The traditional stance on B12 has always been "only trades are binding, but trades ARE binding" as far as I've known. It's what I operate under, and have operated under since the first (miserable) Dom3 game we played here. That said, I certainly make alliances and engage in diplomacy, but my understanding of them is always that at some point we'll turn on each other if one of us gets into a position to go for the win, and we won't generally announce that we're gonna before we do.

In sum: the traditional boilerplate metarules we've generally listed before games always stipulated only trades are actually binding - tho' ofc there IS a social cost to being excessively underhanded.

[Il Palazzo]

I should have put some general rules in the OP. My bad. But yeah, the general idea has always been as described above: trades binding, other agreements more like, uh, a suggestion. Usually there was also a stipulation not to rush anyone in the first year, but I guess it's a bit too late to be bringing it up now.