I don't think you have enough number of games to take those kind of conclusions.

And I thought that Austria was the "worst" country, not Italy.

Austria's high risk, but if it survives 1903 then it gets as strong as any other country.
I think Italy is weakest because it needs to be played by a maniac to get the same sort of growth as other countries (e.g. farmerman in SoW4).

Also, Austria is stronger in FtF, because in FtF it is more easy to detect conspiracy.

italy getting three builds in 1901 is amazing. Italy and Austria are both weak on average, but if they work together can both survive.

however since they usually both prefer to attack each other at some point any alliance is likely to only last until the mid game.

Still i think seing italy at 2% of wins in 46 games is not statistically unlikely. especailly if you look at those games and see how many of them don't have italy and austria working together til at least 1903. (italy satbbing austria before then usually ends up with either russia or turkey gaining most)

statistically speaking italy's winning chance is about 7%. italy is ranked last while the first countries have a 12-13% winning possibility

@ hellalt: Yeah, I thought 7% was a bit low though. I thought it was more like 10% where as the others were all right at or above 1/7th, which would be about 12-13%.

it's a bit low indeed. but not as low as 2%. also Germany's and Austria's chances are about 9%

i take it that 1 out of 46 games was won by italy.

1/46 is 2.17%

but the error in a count is usually the square root of the count. The square root of 1 is one, so the error is +/- 2.17%. Which is a pretty big error. Basically your sample size of games isn't big enough.

two out of my six solos on here have been with italy. it's probably my favorite country to play.

The number of trials is 46 though, not one... so you divide the t-score by the square root of 46, not 1. And besides, you'd have to do a complete t-test, which is nothing like what you did.

I don't think so, though the probablilty of italy not winning would be
45/46 +/- sqrt(45). which is 97.8% +/- 6.7%

that's a much smaller relative error but a bigger absolute error.
My error estimate is the lower of those two which is fair because i'm estimating the minimum error.... not sure about t-tests, but any number of games where there is only one italy win still has a 100% error estimate (predicting 1 italy win +/-1 which isn't too useful...) Thus 46 games isn't enough.

Italy is good. But I still haven't figuredout the key to success as Russia.

Shouldn't you distill from the 46 games the drawed games, to leave only games where one country has won?

According to my statistics I've played France 4 times in a total of 71 games. Now, that's weird.

I suppose Ursa... Or count the ones in which italy has drawn as well.

They are two separate stats and could be wildly different for each country. So in games which end in a solo, what percentage of those were Italy? And then in games which end in a draw, what percentage of those included Italy?

hellalt - the statsitics from Tournament FTF play (the most 'pure' form of diplomacy) reveal each country has a 5-6% chance of winning with about 60% chance of draws occurring. so I dont know where you are getting your 7% and 10-12% statistics? which assumes a very low draw %...

also keep in mind, PPSC games vs WTA games have a WIDE differentiation.

Looks like in your games there were 6 draws of which Italy and Germany were only involved in 1 each. England was involved in all 6. France was in 4 and the other countries were each involved in 3.

As far as solo-win games, 1 out of 40 is still only 2.5%. It is still a pretty small sample size.

I prefer playing Italy in gunboat games to ordinary press games. That way you don't have to worry about what Austria is doing, and just head straight to destroying him. Then your only worry is whether France/Turkey/Russia will come after you, but that's only a small concern...

7% was a figure given in some dipPouch statistics, i think. Based on play be email games over about a decade iif i recall correctly.

sorry, play by mail, not email.

@Babak perhaps hellalt was referring to percentage of wins by a given country?

I feel like I should point out that unless you played all 46 games as Italy, then your win percentage isn't 2% with Italy. Nor does you doing badly with any particular county say a lot about its overall effectiveness.

those are tournament statistics. mine are from the last 1000 finished game in this site. and those statistics are approximately 20% draw, 11.5-13% england france turkey russia, about 9% austria and germany and about 7% italy

I'm saying that of the 46 games I've played, Italy has won one. Does no one understand what I'm saying?? I'm not saying I have a 2% win percent. I personally have never won as Italy. I'm saying of 46 (relatively randomly selected) games, Italy has won one, and I find that surprising that it is that low.

hellalt - i see... well the last 1000 finished games on this site include both ppsc and wta which should not be included in the same categorization since there are widely differing reward-mechanisms in the two different scoring systems... another words, player behavior is (and should be) different based on the variant.

rlumley - yes, that is a low percentage. what made you go and look?

I don't know if this information is useful to anyone but out of 33 games I've played in (both here and on fb-dip) winners and drawers are:

Wins:
Turkey 6 25%
Germany 5 20.8%
France 4 16.7%
Russia 4 16.7%
Italy 2 8.3%
England 2 8.3%
Austria 1 4.2%

Participate in draw:
Russia 5 20.8%
Turkey 3 12.5%
Austria 3 12.5%
Italy 2 8.3%
France 2 8.3%
England 1 4.2%
Germany 1 4.2%

@ Babak

I was playing a game as Italy, and realized that I'd never been in a game that I remembered where Italy had won. (The one game in which Italy won, I was defeated early on as England) I was wondering if I just remembered wrong or not.

Data problem: Is that 2% of all games? Or 2% of games that had a solo win? In how many games that ended in a draw did the Italian player survive? How many games did you play with CD Italy?

There are lies, damn lies, and STATISTICS!

This site has several issues that distort the effect of Italy
1. if it starts games with less only 5 or 6 players then Italy is not played.
2. there is a high percentage (25-30%) of games where there are No Moves Received. The most common gain in such a situation is Russia not Italy.
3. Italy is along with Austria are the two hardest countries to play on the whole for the new player or players with average experience, and many opportunities are lost as well as disasterous moves less forgiven. *France is the most forgiving country to play if you make an occasional bad choice.
4. php's mix of players rewards aggressive early play which is hard for the players to integrate in playing Italy without blowing up the Austrians in a manner that does not establish a run away path for Turkey or Russia to win the game.