Not super normal, but certainly possible, especially if your opponents have recently finished games as Italy. The system is weighed so that you're less likely to get powers you've recently finished games with, so it should even out as you play more games.

@Romain
Change your name. It's obviously influencing the randomizer.

@Thomas : lol :)

I know who that feels. 5 classic games and 3 Italy too.

Just the luck of the draw. I love playing Italy, so I'm envious of both of you.

Well some hints would be welcome.

You don't seem to need help with Italy at all!

http://webdiplomacy.net/board.php?gameID=158768

Ahhh but that was luck in having Austria and Turkey go missing at point in which I was going to crash :-)

In my first eight games on PlayDip, I drew Austria four times and Germany four times. I think the algorithms they use have some of our user names hard-coded in them. I'm pretty sure that's true in the tournament.

so things are better here because you drew Russia!

The weighting function only takes account finished games, so if you started all five games before finishing any, it wouldn't help you.

I'm embarrassingly bad at statistics, but the chance of any one player getting any one of dive countries in a row is not so unlikely. It just happened to be you.

Polyanna, having had very little experience of Russia I thought it was a terrible draw. I checked stats on PlayDip and saw that it was 2nd only to Turkey in the number of solos. So, maybe it's not so bad.

We all know (or should) that flipping a coin ten times and getting heads ten times is no more or less likely than any other sequence. It just *feels* odd. We aren't really rational creatures. Surprise, surprise.

Since you haven't finished any games, starting five new games means that, ceteris paribus, you have an equal chance of playing all seven powers. However, there are six other players on the board who (presumably) have finished games under their belt, the probabilities for your own country selection are influenced by the *other* players' probabilities. In other words, your odds aren't 1/7. Since each other player's odds of selecting one of the other countries is probably better than 1/7, you're simply a victim of weighted probabilities.

If you stay on the site for a while, you probably won't play Italy for another year, so eventually it will even out :)

Grow where you are planted, You'll either figure out some good ways to make Italy a viable power, or some great ones to kill him off quickly :-)

I am actually pretty sure that the country selection uses started games as an indicator. So you just had bad luck (or good if you like Italy).

Just stick with it, I hated Italy at first but after playing it a bunch I learned that it is the most relaxing power. You can go either way, neither of your neighbors is overtly concerned with you and if you are good at diplomacy you can lie to your hearts content about what Turkey/Russia/Germany/England are planning to influence what Austria and France do. It is so much fun. "Hi France , Germany asked me to attack you I think he may be coming your way." "Hi Austria, Turkey wants me to feign a lepanto, I want to start the Lepanto but stick with it. Turkey will assume I am working with him but really I am working with you." Then when you get in a comfortable position you can either rip into Austria or continue through with the lepanto and take out turkey.... Italy is the nation with the most creative options and the most stimulating play.

"I am actually pretty sure that the country selection uses started games as an indicator. So you just had bad luck (or good if you like Italy)."

Oh, is this true? Can anyone verify?

it's the opposite of my current understanding but since this is an objectively answerable question I didn't want to make too much noise about it I figure we can just ask a developer. Does anybody know a developer?

I just checked the code and it looks like JMO is right. Country weighting is updated for each player every time they are assigned a country, not when the game ends.

As Abge says, the country selector uses your started games to determine who is who. I don't know where the confusion with finished games came from, but it seems pretty common.

Doing it with started games is better than biasing by only finished games, as that way your current games affect the outcome.

A couple of other things:

* started games of course includes cancelled games
* the bias is usually pretty small
* the system "rolls" for the person who has had the biggest imbalance of who they've played recently first. So your opponents affect how much your previous counties affect your next country.

~webDip dev team

(I didn't design the system, but I can answer other questions if there are any)

I find it interesting that polemarch has "3 finished games" with the following stats:
Defeated 1 game - 33%
Wins 1 game - 33%
Draws 1 game - 33%
Survival 1 game - 33%

Known bug with anon defeats

Ahh, I knew it had something to do with a game that wasn't finished yet, but I didn't try hard to solve the mystery.

And that was being squeezed as Italy :-).

I was drawn as Italy a lot in my first month or so here. I'm not a big fan of playing as Italy. I've been France a whole lot in the last two or three months. I like playing as France, but I'd prefer more variation. More Russia, Germany, and Turkey would be fun.

Maybe the algorithm can be changed to be weighted more, so that it sets variance as a higher priority and randomness as a lower one?

Honestly... if this guy has played 5 times and got Italy 5 times our system is messed up. He recently finished his first game (as Italy)... so his second game should be less likely to get Italy. If he joins a game with 6 other players that have all gotten Italy then okay maybe he gets Italy twice. What about his third game How does he get Italy again. And a 4th time .... and a 5th time... Come on that's 1 in 2401 times... that someone would be assigned the same country 5 times in a row. I guess it can happen but if the system is designed to be weighted it should be such that after some (reasonable number) of games you won't get that country the "next" game.

If it's a 1 in 2401 (or even if it's less because of the weighting) it's decently likely that it will happen at some point in time to somebody since we have a decently large user base. Someone's going to win the lottery, it's just unlikely that it'll be you.

That is with NO weighting... we apparently have a weighting system to curb this. From the weighting systems that I have seen there is some theoretical max at which that play gets or doesn't get the desired result. I guess my thought (though I know this would be a crap load of coding) would be that each player has there last country assigned along with a streak. For example my last game was Germany and your's may have been France. Maybe someone has a last country of Italy and a streak of 5 in this case. When assigning countries the assignment algorithm goes through and picks the player with the highest streak and assigns them first (to a country that they didn't most recently player. Then goes to the next highest streak and so on. Certainly this would have many OTHER flaws to it but it would in all likelihood eliminate all occurrences of a player getting the same country 5 times (and likely even 4 and 3 times) in a row.

But yes mromain you are our lucky player! Or sacrificial Italian lamb. Whichever you prefer.

PS - my english no good... I'll proof read next time before I post.

hey. Guess what! I just had another game start and I am Italy again too! 4 times from 7 games. Lots of practice coming up it seems

I hate when that happens. I was assigned Italy two out of three World Cup games in a recent series.

I started a new game and I have now : FRANCE. Yihaaa.

Finally it is good to play several times under Italy :)
http://webdiplomacy.net/board.php?gameID=159705#gamePanel

5 games without finishing one, so no weighting in the algorithm, it seems like it should be (1/7)^5 - but if players in your games had all recently played Italy, that chance can go significantly higher. It would be very unusual but someone playing their first 5 games without finishing any, the sample set would be vanishingly small.

So not any example of a bug in the system - very much within statistical probability. But I'd lay money on it not happening to the next 1000 people who join the site.

Oh, just read ATC's post, well then it was less likely, but still not out of the realm of possibility.

Oh dear. I just got Italy again. So I have got Russia 3 times too and Turkey once but none of the others yet :-)

I've gotten Turkey a ridiculous amount of times. I just want to try France, Italy or Russia, but noooo.

Wouldn't it actually make a little more sense to weight the system considering the started games instead of finished games? I mean, I too often get those streaks when I join several games at once.

In modern I honestly think I got like 5 times Italy and 2 times Spain out of 8 games I played : D

See ATC above (and I bet he looked at the source code to get the definitive answer). It does use started, not finished. But it's also still random, so that means you can still get a run of repeat draws in a row.

I've actually gotten Italy in every game I've played so far. 400 times almost. Is this a bug?

Hah. Pull th other one!

Italy, Italy, Italy
Italy is all I crave.
Start me off with Italy
Think of all the strife you'll save.

Finland bounced? Jugg announced? England trounced?
Stuff it where you stuff your Western Trip.
You can play her fast or slow,
Armies high, or navies low.
Italy shows off your craftsmanship.

It is not playing Italy. Rather starting out on the site and instead of getting an even chance of playing all the 7 nations in your first 10 games, getting Italy over 50%.

That said I get your poetic message

Italy is the most fun country to play