LongTurn League

Current and future games
Post Reply
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

LongTurn League

Post by Corbeau »

I have made a post about this on the Freeciv.org forum.
http://forum.freeciv.org/f/viewtopic.php?f=20&t=89912

Am copying it here, but I will maintain the "League" there for the purpose of getting more attention, people etc.

Comments, ideas, criticism, all welcome ;)
Starting with LT40, I will be calculating and recording scores for the purpose of persistent ranking of individual players of Longturn-class games of Freeciv. The scoring system will be fairly simple: number of points awarded will be calculated using the following formula:

Points = ROUND( ("inverse rank" / No. of active players)^2 * SQRT(No. of active players)*10 )

where "inverse rank" is almost self explanatory: in a game of 10 players, the best player has rank 10, 2nd best is 9 etc. and last is rank 1. Rank will be determined by the in-game score at the moment the game ends. Modifiers to this score may be possible, but that should be determined by half-game at the latest, and preferably much, much sooner.

An calculated example of the scoring can be seen here:
viewtopic.php?f=20&t=75565#p81253
(see bottom of the post, "Option 2")

Number of active players will be determined by the half-game based on how many people actually played a significant number of turns, that is, enough to have been able to influence the game.

If nobody objects, for LT40 I will use 25 as number of players because even if there were players who quit early, I believe they were replaced quickly enough. Of course, we can debate the "enough" but a change of this number for 1 or 3 won't change much anyway.

Conditions for LTL scoring
There will be one hard-fixed condition: no alliance victory in the game.

The reason for this is that I want this scoring to be a show of individual skill of running your nation and negotiating with your neighbours and potential friends and enemies. You will be competing against your friends, too, which isn't possible if alliance victory is enabled. This changes gameplay drastically and would make a huge difference regarding your goals.
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

Post by Corbeau »

Kryon wrote:From the below quote I understand that games with alliances will not be counted? Why not? They can still be counted with the same formula. The ally members almost always have the top scores by the end of the game anyways. But, including those games with this system will cause extra competition/friction between ally members in terms of earning more in-game scores (e.g. friction about which alliance member will conquer which enemy cities?). This might make the alliance games more interesting :)
Well, for this scoring to work, people need to care about it, and alliance games will already be scored with the standard LT system. So, when there is a standard LT alliance-victory LT game, an alliance member may consider sacrificing himself for the team and still geting a lot of points in the LT scoring. On the other hand, he would get much less points in "LTL scoring", so, basically, he has to choose between the two. And the truth is, LT is old and established and I expect much more people to chose that one over "LTL" which hasn't even started yet.

Also, "alliance victory" in itself is an abstract concept. What matters is an end score, what you take out of the game after the game is finished. Again, even if the game is "experimental" and isn't in the "LT scoring", but promotes "alliance victory", again, people will have to choose between two goals: be able to "take one for the team" and go for alliance victory, or go for maximum "LTL score".

I believe having two different out-of-game goals in one game can mess up the game.

Also, when an alliance is formed, it can overrun some opponents who are actually better players. But if there is no "alliance victory" in play, then the allies will indeed have to see among themselves how they will split the victory which indeed is what I am aiming for. Having "alliance victory" in play will only hinder this.

I feel I wrote this a bit messy, but I hope I got my point through. Simply, when alliance victory is enabled, it makes people play differently from the goal I wish to set up with this scoring, which is "individual play": play according to your own preferences and capabilities, not (completely) based on who you are allied with. Of course, there will be alliances, and people within the alliances will shift within their internal hierarchy, but there will be a clear knowledge that in the end your score will depend only on your own actions.
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

Post by Corbeau »

Wait, you mean there has been no scoring whatsoever since LT34? I haven't been checking the lists lately, I saw that they are not really updated, but I thought someone is simply late with them and that they wil be updated sooner or later!
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

Post by Corbeau »

Kryon wrote:Yes, there has been no ranking update since LT34 because I did not have time to do it :) Therefore I assumed your ranking would become the main ranking.
I don't have the ambition to become the "main LT ranker" for a simple reason that there are at least two different "schools of thought" in LT and I have a feeling my scoring system heavily favours one of them which would leave the other one disinterested. I think there is definitely room for two parallel scoring systems. Wouldn't be the first time a game or sport has them :)

And, now I checked and definitely there is data about previous winners. For example, LT36 is here:
http://forum.longturn.org/viewtopic.php?id=645
User avatar
Lord_P
Member
Posts: 123
Joined: Thu Jan 01, 1970 12:00 am

Post by Lord_P »

What I would really like to see for LT scoring is a sort of MMORPG version..
A kind of perpetual, perfectly balanced game where your civilisation can collapse back to the stone age as easily as advance and new players can constantly join or idle players can be kicked. Like a CIV petri dish where empires grow, stagnate and then are consumed by others.

The aim wouldnt be to 'Win' but to see what you can achieve before you are inevitably wiped out by the super hostile world/players that surround you.
Then you could have ongoing top 10 rankings like 'most science researched' 'largest empire' 'most units killed' etc. So you can aim for whatever aspect of the game you consider an achievement.

You could possibly do this as a kind of long thin 'conveyor belt' world where new tiles and players are added on the east side and tiles drop of the west side.
Make unit upkeep and available tech heavily dependent on city improvements and to keep your empire going you would constantly have to fight your way east, capturing and rebuilding as you go, against endless hordes of less developed barbarians, until one takes your place...
Last edited by Lord_P on Wed Feb 28, 2018 11:50 am, edited 1 time in total.
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

Post by Corbeau »

I really like this idea.

(Except the "conveyer belt" part which... doesn't make sense to me... Maybe do something with natural disasters or enhance global warming, but with some "early terraforming" abilities, desert -> plains -> grasslands and ocean -> swamp -> grasslands. This way land has to be maintained or else it falls into disrepair and the nation is wiped out.)

Also, tech upkeep does exactly this: if you don't maintain your nation, it drops behind and votes Trump or Brexit or something.

The main problem would be that the few old and good players who would play persistently would be unsurpassed because they simply wouldn't collapse on their own. However, this is where the long-longturn idea comes into play: with new players and enough time passing, there would inevitably be new alliances formed, allegiances switched and old empires not destroyed, but removed from the throne.

And to help the newcomers, make horsemen almost free to produce, have some guerrilla units always globally available (Great Wonder style, if one player has it, all players reap the beenfit). See how Fortress Europe handles itself under hordes running away from climate change and empoverishment... ok, I'll stop here :D

I'm all for it.
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

Post by Corbeau »

Standard disclaimer: This is the scoring I am (will be) running according to my preferences and ideas what the game should represent. Everybody is free to completely ignore this and play the games I'm keeping the score for any way he pleases.

That said...

Looking at the example of just finishing LT40, I wish to introduce two changes for future games where I will be keeping score.

1. Random end

I am still not sure how to implement this. Suggestions are welcome.

I have noticed two problems. One is that, as a multi-month game is approaching its finish, things are accelerating towards the violent grand finale. The final few turns are starting to look like a frantic race for the score regardless of any logic and previous playstyle. Ok, I am aware that this is "just a game" and this happens in all games, but sometimes I have a feeling this is going too far.

On the other hand, this is just a minor effect and it only influences the relative position of two neighbouring players and maybe it doesn't have impact on the game as a whole.

But the other problem is more serious.

I know of at least two players who planned or offered to give their cities to someone else in order to alter the final game score. I can only guess how many others are/were willing or actually went through with it. Now, in itself, this doesn't have to be a huge problem. In some cases this can be seen as a reward for consistent playing, being a worthy member of an alliance and also a reward for selfless acts earlier in the game. Also, maybe it can be considered a dead-man's trigger, a final act of revenge against someone who stepped on someone else's foot.

These things make this last-moment gifting a more or less valid tactic.

However, what can also happen is a trans-game score trading, exchanging out-of-game favours and creating trans-game alliances. I don't have to explain why this is bad.

To solve these problems, I'm thinking about devising a system to record the score at a random moment at some point in the game after a certain "switch" has been activated. It can be "any player researches tech X" or "best player has more than Y land" or something like that.

As for a random moment, still thinking about how to do it. It could be something completely out-of-the-box, like, after the "switch" has been activated, the game ends / score is recorded in the moment "the first title on the New York Times web page starts with letter M". Or some other random "switch 2" can be devised, ideas are welcome.

There are problems with this kind of solution, but at this moment it is the best idea I have. Also, I'm still weighing out how much of a problem this may become.

2. Pollution

Firstly, I don't know what happened to negative score from the pollution. At least in Civ 1, and very probably in later versions, too, there was -10 points penalty for each polluted tile at the end of the game. There is no such thing in Freeciv, not even as a ruleset/server possibility.

At this moment, in the last turn, the world of LT40 is a complete and utter wreck. Huge patches of desert and swamped plains and who knows how many land tiles turned into sea. As far as I'm conecerned, there are no winners in LT40. Everybody has lost. Naturally, not everybody is equally to blame for this.

This is why, for the purposes of this scoring system (that, I repeat, everybody is free to ignore), at the end of every game from now on (not including LT40) I will make a printscreen of the final map, MANUALLY count all the pollution tiles and deduct them from the score. I'm still considering whether this should be -10 or a bit less, but the information will be known soon enough.

.....................................


These two things may exclude each other: if I decide to stop the stopwatch and record the score at a random moment, I will not be able to check pollution. So I'll probably have to choose between these two options. Currently I'm more inclined towards #2.

As always, ideas and opinions are more than welcome.
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

Post by Corbeau »

I am long overdue with copying the results of LT40 here (being the first game I was planning to include in the LTL), but being only one game, it's not that relevant yet. The scoring can be seen here:

http://i64.tinypic.com/30vhlqc.png

With the second game I was planning to include in LTL, LT43, it turned out that I misunderstood the pre-declared winning conditions. Those are that the game can be ended EITHER by "alliance victory" OR by "score victory", depending on the situation and the disposition of players. (More information here.

So, I am slightly re-declaring things:

In order to fall in line with the main condition that I set - no pre-declared alliance victory - LT43 will be included in LTL ONLY in case a "score ending" is declared. If it ends with alliance victory, it will not. AAnd, again, this will be decided by in-game actions of the players.

If the game is included in LTL scoring, the following adjustment will be made: for every polluted tile 5 points will be deducted from the in-game score of the player on whose territory the pollution is present and the ranking for this game adjusted according to that.

And, as always, standard disclaimer: all of this doesn't dictate in any way how people should play their games. This scoring doesn't influence gameplay in any way and everybody is completely free to ignore this.
Last edited by Corbeau on Wed Jun 13, 2018 1:20 pm, edited 1 time in total.
Corbeau
Member
Posts: 990
Joined: Thu Jan 01, 1970 12:00 am

Post by Corbeau »

Post Reply