Recommend
4 
 Thumb up
 Hide
63 Posts
Prev «  1 , 2 , 3  Next »   | 

Terra Mystica» Forums » Organized Play

Subject: Sign Up for Season 30 of TM Tour rss

Your Tags: Add tags
Popular Tags: [View All]
Socks Wielder
Canada
flag msg tools
mbmbmbmbmb
Socks Wielder wrote:

On another note how likely is it that 16VP gets me to D4? There are 6 players in 2nd over 16vp and 2 with exactly 16. 1 league isn't finished so lets's say worst case 7 over 16 and 2 tied. I got through on 9 better and 4-5 tied in season 26. But that was probably a lucky breakthrough, A couple people not signing up from D5 and D4.


Turn out it wasn't good enough for D4!
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Robert
Germany
Bocholt
flag msg tools
badge
I paid 100 Geek Gold so that you can read this! :-)
Avatar
mbmbmbmbmb
Socks Wielder wrote:
Socks Wielder wrote:

On another note how likely is it that 16VP gets me to D4? There are 6 players in 2nd over 16vp and 2 with exactly 16. 1 league isn't finished so lets's say worst case 7 over 16 and 2 tied. I got through on 9 better and 4-5 tied in season 26. But that was probably a lucky breakthrough, A couple people not signing up from D5 and D4.


Turn out it wasn't good enough for D4!
In D1, you'd have won with 16.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Socks Wielder
Canada
flag msg tools
mbmbmbmbmb
Exactly, so I consider myself better than D1 players whistle whistle

Right?

If I had a better understanding of Halflings I probably should have made it so I'm to blame in the end.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Ryan Feathers
United States
Madison
Wisconsin
flag msg tools
badge
Avatar
mbmbmbmbmb
Steve496 wrote:
I won't speak for anyone else, but I, personally, am not complaining about faction diversity in the upper divisions. I enjoy debating faction balance. I enjoy pulling fun facts out of the statistics. And I would definitely enjoy the discussions and debates that would arise out of a serious effort to find faction tweaks that improved the game. But there is nothing so broken in TM that I feel it *needs* changing at this point in time. And to the extent that something does need changing: the tournament is not the place to experiment with such things. The tournament is an established institution that should only adopt new and different things when they are well-established as improvements, not as an experiment because it might be better.


In response to James, I tend to be of the completely the same mindset as Steve.

The other thing is that I do play pretty often with the map balanced VP's over on BGA, and let me tell you I do not believe they would have the effect you think they will. It might shift around the win rates a little and it will have some impact on how things play out surely, but my best is we'd still moan about how black factions are in every single round (because even with -5vp Darklings are still silly versatile, and at 27vp Alchemists have a lot more room to play), we'd still see a lot of the same factions simply because even if you gift the Giants a few more VP, the sorts of scenarios where they are playable are still niche (few cult spades, no bon1, early SH, Bon6, etc).

Essentially I remain entirely unconvinced map balanced VP's offers a more interesting game or meta, certainly I haven't seen enough data or had the experience that makes me convinced we should mess up the really good thing we have going here to venture into a new unknown that may well have just as much convergence on a few top factions and matchups.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Wolfpacker
United States
North Carolina
flag msg tools
badge
Avatar
mbmbmbmb
Y'all have a point, but nothing ventured, nothing gained.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Hungary
Budapest
Budapest
flag msg tools
I am not a fan of the adjusted vp stuff but I dont see how is it fun to have like 6 "viable" factions and only 3 of them have a realistic chance to win. In season 29 D1, in 6 games the 4th player knew that he does not have a chance for the first place when he chose his faction, and in the 7th game (game #5) darklings were open but still came in 4th. Also, even though usually the player in the 4th seat does not have a chance to win or get the 2nd place, he can choose whether he wants to kill engineers (pick witches) or cultists (pick nomad).

Anyways, from the tournament perspective it can work, as everyone plays once from every position. But i think its not very fun to play 2 games where I fight for 1st place and 2 games where I fight for 3rd place.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Robert
Germany
Bocholt
flag msg tools
badge
I paid 100 Geek Gold so that you can read this! :-)
Avatar
mbmbmbmbmb
fiziqs wrote:
Also, even though usually the player in the 4th seat does not have a chance to win or get the 2nd place, he can choose whether he wants to kill engineers (pick witches) or cultists (pick nomad).
The main problem is that he cannot choose to pick something which kills Darklings.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve Haas
United States
Mountain View
CA
flag msg tools
I mean, he can't really pick to kill Cultists, either. A nomads pick still leaves Cultists with legitimate chances at 2nd. Expected points:

Engineers 4.2
Cultists 2.6
Darklings 2.6
Engineers 0.6


Picking into DEC isn't about picking who loses; it's about picking who wins. A witches pick strongly favors Cultists (with Darklings having chances); a Nomads or Mermaids pick favors Engineers (with Darklings and Cultists mostly fighting for 2nd), and so on.

And you're right, picking with your best-case scenario being 3rd place isn't a ton of fun. The kingmaking isn't ideal. But balance is not a static thing - people are continually experimenting with different factions and placements and strategies, and while no good solution has been found so far, that doesn't mean that no solution exists. MEND was once thought to be a fair matchup. And then it was thought to be an engineers-dominated matchup. And now its starting to look like it might be a Darklings-dominated matchup. The same thing could happen with the matchups we're currently struggling with.

And even in a world where that wasn't happening, how much would VP offsets help, recently? The average deltas to winner for the above matchup at current are

Engineers +12.3
Darklings +1.8
Cultists -0.6
Nomads -13.5


With the map-balanced VP offsets that becomes

Engineers +11.8
Darklings +0.3
Cultists -1.1
Nomads -11.0


How big an improvement is that, really? Nomads are still losing, and Engineers are still winning. And yes, maybe map-balanced VPs means the 4th player picks Fakirs instead of Nomads for the extra VP, but even there - the extra 14 points only barely larger than the delta from Nomads to Darklings/Cultists - Engineers are still clearly favored, and even if the Map-Balanced VPs fix the color imbalance and puts yellow on a par with brown and black... yellow still ain't the favorite for 2nd place, because they're still playing Fakirs against Darklings and Cultists.

Basically: map-balance VPs address average faction balance across all matchups; but any *particular* matchup is far more imbalanced than the average, and sometimes the "fix" isn't even pointing in the right direction. I'm not sure black and yellow aren't favored in Alchemists/Nomads/Witches/Engineers in the first place, and map-balanced VPs gives them *more* of an advantage relative to the already-crippled engineers and hindered Witches.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Wolfpacker
United States
North Carolina
flag msg tools
badge
Avatar
mbmbmbmb
Doing a static evaluation in a dynamic situation sometimes works and sometimes doesn't... I'm going with doesn't work here.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve Haas
United States
Mountain View
CA
flag msg tools
You know, it's comments like that that make my question whether you actually want to have a conversation about this, or if you're just spouting your talking points in the hope that we'll all eventually give up. Posting "no, you're wrong" provides no basis for discussions. Why do you think it doesn't work? In what way do you believe this analysis is flawed? What analysis could be done that would allow us to assess whether your thing is actually better, or to improve the offsets to the point that they actually are better?
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Wolfpacker
United States
North Carolina
flag msg tools
badge
Avatar
mbmbmbmb
Map Balanced VP is also a static solution in a dynamic environment. So I am agreeing with you that it doesn't quite work either, but it will definitely cause some differences in faction selection. A faction selection and auction using the Evo/Stockpile method is a dynamic solution: either take a selected faction away from another player for higher vp cost or take an not yet selected faction for any cost including 0 and the auction is over once every player has a faction. This dynamic solution is not perfect either because people have preferences and preconceived notions.

However, that dynamic solution will not be implemented on Snellman as said by Juho in the past. If someone else wanted to implement it, they could probably host their own site with those changes and see if people wanted to switch the tournament to an auction method.

The only way to really find a best fit static balanced VP is using a regression model something similar to:

VP= x1*Map + [xR1...xR6]*[R1scoring...R6scoring] + [xBON1...xBON10]*[BON1...BON10]

You also can't add these terms because it would change based on factions in the game:
+ xf1*faction_player1 + xf2*faction_player2 +...

Also, I know that we don't even have enough data for every faction either. A few times I have been actually able to find a random near exact setup duplicated that was off by a BON or 2.

If we had an AlphaGoZero available for Terra Mystica, I'm sure it would be able to find out a lot.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve Haas
United States
Mountain View
CA
flag msg tools
For all that I'd love to see AlphaMystica, I kinda feel like this proposal is a bit overcomplicated. Fundamentally, the problem is not that faction X is 5vp better than faction Y averaged across all games. The problem isn't even really that faction A always wins the specific matchup between factions A, B, C, and D. The problem, at it's core, is that the mechanism by which we select factions has strong attractors. If there were 50 matchups that would all reasonably arise when picking factions in high level games, it would matter less that many of them are theoretically predetermined - none of them would get played often enough to fully converge (and even to the extent that they did, keeping track of who's favored and how to play fifty different matchups is a lot harder than when there's 10).

And why are there attractors? I mean, the fact that there are a couple of factions that are just plain bad doesn't help, certainly, but I think even if Fakirs were a full co-equal partner of Nomads I'm not sure that would fundamentally change the dynamics of the game very much. I think the problem, at it's core, is that there are a couple of factions or combinations of factions that are viable in literally just about every game. There's no such thing as a game (or matchups) for Darklings that is so bad that even if they're available in 4th seat, it's wrong to take them (there might be a better option - maybe - but they're certainly not a bad option). Similarly, it's very hard to imagine a game where it's not worth picking engineers or witches in 4th seat if both green and gray are still available.

The result of this is: almost every game has Darklings. Almost every game has Witches or Engineers. Most games have Nomads or Cultists. And with those constraints... there are only so many matchups that can turn up in regular play.

So at it's core, the solution needs to be about making sure there are some games for Darklings that are legitimately bad, where if you pick them you probably lose. And some games for Engineers that are genuinely bad, even if Witches are not in the game. Such that faction selection is no longer principally game of deciding what to pick into DEC, DEN, and WDC, but uses a much larger portion of the matchup space. Ideally, this would also increase the diversity between factions of the same color so it matters more whether you're playing Darklings/Witches/Engineers/Cultists or Alchemists/Auren/Dwarves/Halflings, but even that is somewhat limited in it's matchup when you have colors with 99% representation across games.

And the key point is: you can't do that with a VP offset. The fix can't be making Darklings 2vp worse in 100% of the games; it needs to be making them 10vp worse in 20% of the games. Which is not to say that even if you came up with a good way of doing that, it would be enough - there probably do need to be other changes across the game. But until you take the steps to fix that underlying issue, I don't think you can meaningfully improve matchup diversity - it may converge on a different set of a dozen matchups, but it's still gonna be a pretty short list of matchups.

And yes, having matchup-specific victory points that tell you who's supposed to win and applies the average delta so they don't might do that, but it's hard to implement on a computer, never mind in a board game (and, I think would suck a lot of the fun out of the game). So rather than tweaking the scores of the game to make things line up, I think you need to actually change the mechanics of the game to create weaknesses for factions that currently don't have them.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Robert
Germany
Bocholt
flag msg tools
badge
I paid 100 Geek Gold so that you can read this! :-)
Avatar
mbmbmbmbmb
Steve496 wrote:
There's no such thing as a game (or matchups) for Darklings that is so bad that even if they're available in 4th seat, it's wrong to take them (there might be a better option - maybe - but they're certainly not a bad option). Similarly, it's very hard to imagine a game where it's not worth picking engineers or witches in 4th seat if both green and gray are still available.


Steve496 wrote:
The fix can't be making Darklings 2vp worse in 100% of the games; it needs to be making them 10vp worse in 20% of the games.
Of course a setup-independent variable VP distribution can be a fix to what you wrote in the previous quote! If Darklings started with 5 VP, it would be wrong to pick them in most setups. Likewise for Engineers and Witches and Cultists and ... There's the danger of throwing out the baby with the bath water (e.g. 5VP Darklings may have no niche at all), but we already have factions where this happened during game design (e.g. Fakirs), and we all love TM anyway.

Originally I felt that the variable VPs were too drastic, but I'm moving towards the conclusion that for some factions, they actually are not drastic enough. If I were to make an update to the variable VPs, I'd certainly nerf Darklings into underdog status, and at the same time cut Alchemists and maybe Auren down a few VPs as well, while providing different values for 3p/4p/5p. All this using the rationale that even a single OP faction is much worse for game balance than several underdog factions.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Saku Suru
Poland
flag msg tools
mbmbmbmb
A simple tournament rule change, which enforces a flat distribution of faction's choices, could bring more diversity in games possibly resulting in learning more about less frequently chosen factions. I bet this would add an interesting layer in metagaming as well. Of course, this rule would be a problem for players like FaikirsOnly (70% win rate in playing Fakirs, rating 1450).
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Robin Zigmond
United Kingdom
Durham
flag msg tools
mbmbmbmbmb
norkbes wrote:
A simple tournament rule change, which enforces a flat distribution of faction's choices,


How would this work? Even leaving aside the fact that you couldn't enforce it (except with some sort of judge watching faction picks and penalising players for going against the rules), what would you want the rules for faction selection to actually be?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Saku Suru
Poland
flag msg tools
mbmbmbmb
Well, already the game tables are populated in such a way that "Every player in a league will be in the same game as every other player exactly twice." and "Every player will play each seat (first, second, third, fourth) exactly once."

The rule, I'm talking about simply requires that "Every player will play with a faction exactly once. The factions picked by a player in a present season cannot be chosen in the following seasons until the pool of other factions is emptied. If a valid choice in a particular game cannot be done due to the choices of other players, the player is allowed to choose a faction freely."

Now, the tricky part is how to implement this new rule. I think one possible solution is to postpone the start of every game in the tournament until the choices of participating players are validated. The automatic validation can be run periodically (e.g. hourly) from the tournament server...

I know that implementing of this new rule requires some additional coding but I think additional work is worth it. In the past, I ran a small tournament for a dozen of so players and it worked very well. The validation of player choices was done manually though.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Phil Hannay
United Kingdom
flag msg tools
mbmbmbmbmb
This would add another layer of complexity to faction choice though. I'd now need to also consider what are all the factions all my opponents are allowed to pick are. Maybe I fancy playing Nomads this game, but one of my opponents must choose Swarmlings or Fakirs, so I take Swarmlings to force them to be Fakirs.

This adds a load of potential metagame, which in my book is bad - I want each tournament game to be a normal independent game as much as possible.

Of course ... it's probably better than my suggested fix of just banning Darklings full stop
5 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Chris Harris
United Kingdom
flag msg tools
If you want variety you can play a 14 factions series, or join the fire & ice ladder, both of which are highly recommended!

Or you can do what I do in the tourney and just pick the factions you want to play with the most...
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
ScottyMcSock
Germany
flag msg tools
On the topic of variety - I just noticed that this season league 2-1 is a five-coloured, six-factioned league.

Which led me to realise (a bit late, probably) that shuffle games could be extended to more than four players: Just copy the layout in all games and compel each starting player to select faction 1, each second player to select faction 2, and so on. Might also be a fun variation, but adding another layer of complexity by the possible need of synchronizing some of the first moves.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Manpanzee
msg tools
Centering this discussion on "should the tournament switch to map-balanced vp?" seems pretty silly to me. It's not going to happen, and there are much better and more reasonable options, which would almost certainly be more popular.

It's a common thing in games to have multiple formats that get played in rotation. There's no reason Terra Mystica can't take the same approach. Those pushing map-balanced vp should set a target like once or twice a year (i.e., every third or every sixth season). Data could be collected, and this could be abandoned or expanded as appropriate.

I do think more variety would be good. For me the purpose of map-balanced vps isn't to make every setup balanced (impossible), but to expand the size of the game. It sucks that there are multiple base factions that are mostly unusable. They're in the box, and I want to play with them!

Predicting how map-balanced vps would affect the meta is very difficult. But restricting yourself to analysis of currently-popular matchups seems too narrow. For example, did you know that map-balanced vps creates a DECx matchup where Darklings might potentially be looking at 4th place?

It's true! Here are the stats for Cultists - Engineers - Giants - Darklings with all players 1200+:

Cultists - 3.80
Engineers - 2.62
Giants - 2.21
Darklings - 1.36

Granted, this is only a 21 game sample, and the rankings don't hold up with lower rated players included. But the point is that the potential of map-balanced vps lies in the enabling of largely unexplored matchups, not merely in the tweaking of common matchups.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve Haas
United States
Mountain View
CA
flag msg tools
It is true that there are games that rotate between rule sets. And, for instance, the F+I ladder does it, and has great success with it. But I don't think the tournament should. I feel like part of the appeal of the tournament is that the rules don't constantly change - you can learn from what you do in one season and apply it in the next and thereby improve and optimize your play more than if you're always playing something different. In a larger sense: success in the tournament is not the product of any one season, but the aggregate effect of continued success, due to the incremental way in which one ascends divisions. And changing the rules constantly would disrupt that.

More generally, the fact that we have such a large corpus of high-level games, all with the same rule set, has been a huge asset in increasing our understanding of the game, and I'd be sad to lose that.

I would legitimately rather have the tournament switch entirely to map-balanced VPs full-time than to try them on a periodic basis. And I say this as someone who hates map-balanced VPs and think they have no place in the game.

(Admittedly, either change would likely make me stop playing the tournament entirely, so perhaps my preference between them is irrelevant).
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Wolfpacker
United States
North Carolina
flag msg tools
badge
Avatar
mbmbmbmb
I doubt that Juho or Daniel would implement this, but instead of boiling the frog with the current map balanced vp, just keep slowly raising the temperature.

That is just make slight adjustments in faction vp every season based on pick rate and finish and auto create a new map balanced vp map for the season.

If a faction is picked and finished greater than average then subtract a vp. If it is picked and finished less than average add a vp. If it is picked less, but finished higher or vice versa then there are no adjustments.



 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Robert
Germany
Bocholt
flag msg tools
badge
I paid 100 Geek Gold so that you can read this! :-)
Avatar
mbmbmbmbmb
JamesWolfpacker wrote:
I doubt that Juho or Daniel would implement this, but instead of boiling the frog with the current map balanced vp, just keep slowly raising the temperature.

That is just make slight adjustments in faction vp every season based on pick rate and finish and auto create a new map balanced vp map for the season.

If a faction is picked and finished greater than average then subtract a vp. If it is picked and finished less than average add a vp. If it is picked less, but finished higher or vice versa then there are no adjustments.
Now this is a cool idea!

I probably would implement it more simple, i.e. only based on pick rate. E.g. sort all factions according to the number of times it was picked, with the top three factions losing 2VP and the least picked three factions gaining 2VP, potentially with the #1 faction losing an extra VP while unpicked factions gain an extra VP.

The aim of such a moving VP table wouldn't really be "balance the factions according to strength" (though it would help to do so), but rather "balance factions according to their frequency of being played", leading to faction combinations outside the well-trodden MEND/DECx/... path.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Manpanzee
msg tools
Tbh I don't really get the level of aesthetic objection toward map-balanced vps. Like, I understand that the platonic ideal is 14 well-balanced factions who all start with 20 points. But I don't get this idea that having a vp offset is such an... idk, impurity in the game.

The way I see it is, there are already two factions (Witches and Halflings) that have faction abilities that only give vp. Yes, the payouts aren't flat, and these vp abilities add some play to your decision making process. But the fact remains that they don't do anything other than give vp. Is "Witches get 5 extra points per town" really that different from "Auren start with 7 extra points" as a faction characteristic?

Going further, why do the Darklings get 2vp per priest dig? Why do the Fakirs get 4vp per carpet flight? Why do the Dwarves get 4vp per tunnel? Why does the Cultists' SH give 7vp? The overarching answer: because tweaking vps has always been one of the knobs used to balance the game, from the beginning.

When people talk about hypothetical balance changes, one obvious suggestion that comes up is reducing the Darklings to 1vp per priest dig. Granted, this isn't a popular change, but that's because knowledgeable people see it as too big a nerf. The unpopularity has nothing to do with the fact that the change constitutes a vp adjustment. People intuitively see a vp-related tweak and consider it. There isn't any need to ask "is this in the spirit of how the game is balanced?", because it obviously is. Treating adjustments to starting vps as categorically different from such vp-related tweaks seems like splitting hairs to me.
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve Haas
United States
Mountain View
CA
flag msg tools
Last year I wrote up my main objections here, and they pretty much all still hold.

To elaborate a bit further: I think Alchemists illustrate the problem with map-balanced VP pretty elegantly. The problem with Alchemists is not that they're too weak. The problem with Alchemists is that they're too hard to play. The set of legitimate Alchemists experts is vanishingly small, but in the hands of those players they're already good, and with a 10vp advantage relative to the "good" factions they'd be crushingly strong. But for most players, even with a 10vp advantage, they're still a bad faction. And so I think the fix for Alchemists is not to make them better, but to make them easier to play.

Similarly: I wouldn't necessarily want to see Halflings made stronger - I'd like to see them have better options outside of R4 dig games. I wouldn't necessarily want to see Fakirs made stronger - I'd like to see them be less totally reliant on carpet flight and cluster scoring. I'd like to see CMs handle crowded boards better. And so on.

In short: the problems I'd like to see fixed are the ones that cause the distribution of usage of the faction across players and games to be skewed in problematic ways. If you make the averages line up without fixing the distribution, you're inevitably going to create games where factions are brutally strong (if they hit their existing niche) and otherwise mediocre, or theoretically decent but still not fun to play, or whatever. And that really doesn't seem like much of an improvement over what we already have.
3 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Prev «  1 , 2 , 3  Next »   | 
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.