so no, it doesn't have to do with when you joined the game or how many points you have. that you're new to the site only means that the other players probably played here for longer and thus didn't play a country for quite a while.

Ok, thanks for your quick response. It does make sense, because I joined every single game, while none of my games resulted in a win, draw, or cancel, yet. So i guess my count marker is still at start.

In case of I lose all my points, am I allowed to have a second account? Because I really would like to test this patteren

no, you're not allowed to have a second account. that would result in an immediate ban, don't do it. however, if you run out of points, they're refilled to 100.

Hey Deus, welcome to the site! What kasi says is all on target. First and foremost, please don't make another account. Your points will returned to you as soon as you complete a game, you will always have a minimum of 100 in play.

Please take a moment to read the rules of WebDiplomacy.
http://webdiplomacy.net/rules.php

If you have any questions, feel free to ask around like now, or send them to the mods at [email protected].

Another important note: the mechanism only weighs against countries you have _finished_ games with. In other words, if you join the site and immediately join four games, and get Turkey in all of them, you are still _just_ as likely to draw Turkey again if you join a fifth game. The mechanism does not care how many games you're playing as Turkey; only how many you've _finished_.

Thanks for the confirmation, so i guess the reason is simply because turkey and austria are the least succesful powers to play with. Taken to its conclusions wouldn't that mean every time someone finishes with england (they survive nearly every time) the likelihood of these players to get Turkey increases?

@webdipmod: my intensions was to prove a pattern, but if points will be filled up, Iwill have to wait till someone else is facing the same issue.

@deus: finishing doesn't mean surviving. i just means that the game is over and not cancelled or in progress. so you didn't get turkey and austria because they always lose (and that isn't even true for turkey, btw).

So in other words, you are saying it was just a coincidence that i'm playing now with ca. 40 different players in 7 different games, who played turkey and austria more times than I did.

yes. also, it could still be pure coincidence :)

How are things calculated when players abandon games?
Are abandoned games counted?

For example, if a player consistently leaves games when they are forced to play Austria (to pick a country many players dislike), the likelihood that they will be assigned Austria in the future increases which means they will also be increasingly likely to abandon more games.

under which circumstances would the likelyhood increase, jeff? it would either stay the same if they're not counted (which is what i believe), or it would decrease.

Well, it would increase indirectly by virtue of the other games. The assumption is that games for the other six nations would be finished to completion and therefore drive those future percentages down forcing the percentages for the disliked countries up.

The system (and someone correct me if I'm wrong) takes into account all games started, so even games that are canceled or the player CDs are reflected in the country odds. Games finished has no bearing on the percentage because the player was already assigned the country in that specific game.

When I reflect your previous statements then there is still an unbalance, because no matter how I finish a game, the only thing that matters seems the country you have been assigned to when the game starts. So there might be rate towards country you have or haven't played that many times, but statistically everyone should have played the same number of countries so everyone should statistically have the same chances. If this assumption is a dead end, then I'm asking myself how balance work here? Because I would also like to pick a country orrandom at fair chances rathen than just wating what the precentage holds ready for me.

if you want to play a certain country, vdip has a feature for that. on wepdiplomacy, the random assignment is part of the board, as this is the official ruleset from the boardgame.
or you find six players who want to play specific countries as well and ask the mods to assign you said countries once the game has started.

"statistically everyone should have played the same number of countries so everyone should statistically have the same chances. "

False assumption. Everyone doesn't have the same chances for a few reasons: first, not everyone has the same number of games played. Second, these statistics are not calculated in a vacuum - your chance of playing Austria is effected by other players' odds of getting that specific country as well. If you're on a board with someone who hasn't played Austria for 20 games, and you haven't played Austria in 15 games, his chance of getting Austria is higher than yours even if you haven't drawn Austria in quite a while. It this weighting that takes into account other players' chances that doesn't make the selection perfectly random each game.

It's a while since I've read the country selection code, but I believe 2nd is correct in that games started is what's used to determine the bias towards countries you haven't played in a while. This includes games that were cancelled or where you were reassigned countries by the mods (your original country counts). I don't think it includes games where you take over a CD.

2nd is also correct about his comments on statistics. A good analogy here is coin flipping- yes, you have a 50/50 chance of a head or a tail, but if you flip it 10 times, it's unlikely that you'll get 5 heads and 5 tails.

Actually, there's about a 25% chance of getting 5 heads and 5 tails. I would definitely not call that unlikely. I get your point but next time you use this analogy you should go for 100 coin flips with 50 heads and 50 tails. Now that's unlikely.

http://www.playdiplomacy.com/forum/viewtopic.php?f=13&t=26542&sid=f8589da3afdd322d799bde3b82cee6ec&start=10

I think HPWehmeier describes it better thanI could, especially he already played numerous games and has cold hard facts about his played games. I'd like to take it from there, because there is a basis now for further discussions rather than making assumptions i cannot prove.

What you'd "expect" given random country assignment, and what you actually observe can be very different. For more information, have a read of this: http://en.wikipedia.org/wiki/Gambler's_fallacy

HPWehmeier, whoever he is, might make a relevant point if our country assignments were truly random. However, as I said in my last post, they are not truly random but weighted random. You have to take into account your probabilities of drawing each country as well as the probabilities for each other player.