Nice! Surprised Italy is last though.

they have never been calculated :-)

Why surprised? Italy sucks. You can't win unless it's a charity game to raise Italy's percentage. Even in real life they managed to get defeated by Greece and Ethiopia...

I just won an Italy game yesterday. Another charity game. We're boosting in up.

I only win as Russia :)

That's not helping it balance out at all! Rebel! Win more Austrian, German, and Italian games.

Im surprised that Austria isnt in dead last... I can never see why the southern powers (Russia, Turkey, and Italy) dont just have a food frenzy over it.

Honestly, I feel like Austria is just a mobile Balkans...

No way, Austria can be a fun country to play.....Erm I can be a touch masochistic at times though.

I won one as Italy too though ...... woot!

Yyyeeesss . . . charity . . . that's right, that's why we let him win as Italy.

The low Italy win percentages are often attributed to both its position (inconveniently in the center), but more particularly its having two armies to start when it's much more of a naval power. Hence the popular Fleet_Rome variant, designed to give Italy a boost.

Austria does a bit better because it has the slightly weaker Italy as a neighbor, but more importantly, it's within easy reach of a large number of centers, particularly the Balkans. Austria has dangerous neighbors in Turkey and Russia, but if it can knock out even just one of those, it's in great shape.

The best fix for balance problems is just for players to be aware of them and compensate accordingly.

Aw man....couldn't just let me enjoy my delusion of granduer, could ya' :-)

"top 4 countries are russia, turkey, france and england, while Germany and Austria follow and Italy is last"

Things have changed a lot if Turkey is that far ahead; Germany is one of the best on the map. I guess poor-quality games are skewing the stats a bit from what I've noticed in high-quality postal games.

i wonder if it would be possible to do official numbers... I'd specially like to see the difference between 1-100pt buy-in games vs 101+ pt buy-in games... i wonder if the difference is stark or not?

and of course, a seperation of PPSC games vs WTA games as there should be a far lower solo % in the WTA games... i'd even estimate that under 50% of WTA games end in a solo...

can these numbers even be culled based on the data?

Check out http://www.diplomacy-archive.com/resources/postal/chances.htm. Their numbers are pretty close to most other sites I have found.

the percentages I posted refer to the last 1000 games on this site that I manually measured a few months ago

I disagree- I love to play as Italy.

I've never won as it (if you'll check my records I've yet to win period, but ah well... it'll happen someday, some great and glorious day after all that work... that and I play to a solid draw unless I smell blood often) but it can be fun to play as Italy.

I usually prefer fleets to armies, so England and Italy are my preferences, followed by Germany (I'm in a game now as it) and Russia (not bad... but too easily to go out first for my liking.)

Turkey's too bottled up for my taste- Turkey wins far more than Italy, but given the choice I'll go with the Land of Pizza usually. :)

France... eh, nothing strong for or against...

And Austria... I can't see how it's higher than Italy...

i was looking into that and doing the same thing. i found the same results. remember that sites vary. yes turkey was suprisingly among the top bunch!

the thing about italy is that it has a low solo while it ranks among the top in draw participants. italy is by no means a chairity country, you simply have to know how to play them. from what ive calculated, i think they also had the second lowest amount of eliminations with france slightly lower. austria was way out in front in eliminations even though they had a slightly lower solo pct.

austria was also the low man on draw participants so if you calculate draws and wins together, italy comes out ahead.

and, for the record, id rather play as italy than austria.

Austria is a much better power to play. Maybe people on here don't Anschluss or are generally just...not good.

Win percentages are pretty meaningless unless they're put into perspective. Since 1910 Spurs' win ratio against Chelsea is 39% won, 23% drawn, 39% lost. Pretty tough call when you want to predict the next game, huh? If you look at data since 1992 the ratio becomes 6% wins, 35% draws, 59% losses. That makes it little bit easier.

TheGhostmaker touches on a similar situation in his article "The Russian Bear’s Soft Heart" in http://www.diplomacyworld.net/pdf/dw107.pdf. The win ratios change as the skill of the players changes.

Don't get killed by skew.

Similar to the tiers in Super Smash Bros., right?

Percentages are very handy. in general they show a trend. everyone was new when they started and they get better as time goes by. this is shown in the statistics. there will be some skewing but when you look at over 1000 games, the skewing reduces (larger sample size).

the percentages can be used to show the difficulty of playing a country. since turkey and italy have the lowest elimination rates, it shows their defensive positions. since russia has a high win pct and a high elimination rate, it indicates that when russia gets a good hold in a game, the chance of a russian win is higher than any other country. richard sharp illustrated this when he noted that if austria is eliminated by 1903 germany comes away with a survival at best. it also can show which countries make the best alliances: for example, everyone fears a juggernaut, yet over the first 500 games i looked at, there were only 2 games in which a turkish/russian alliance ended with a 2 way draw. france and italy were far more likely to be involved in a 3-way draw with an eastern country than the juggernaut countries alone.

ultimately percentages have to be coupled with views of the samples taken and can show the effectiveness of particular openings, how well a country works with another country, it can even show how important a given zone is (ie when the north sea is compromised, england's win pct decreases, but what about MAO or StP?).

it can also show some of the differences between WTA games, gunboat games and public press.

One thing I must agree with Acosmist (and I must admit I frequently disagree with Acosmist), is that Germany is one of the best on the map to play. Probably not so great in terms of win percentages, but it is fun to play.
Btw I believe it has been calculated that on webdiplomacy that Turkey has the highest win ratio. (I have not examined the data myself, though. From my limited experience I would have said France).

Tru Ninja, percentages and statistics can use used to show whatever you like; I agree with you there. However, I think you've got it backwards when you say that a larger sample size equates less skew. Given my Spurs vs. Chelsea example no one in their right mind would put their money on Spurs winning after reviewing the smaller sample of games. Games in the early 20th century do not matter at all whereas recent games matter because the clubs still have some of the players that played in those recent games. If we ignore the fact that you can't choose your country on this site you need to take your skill level and your opponents into account when choosing your country

I know my argument is just as flawed as anybody else's but the point that I am trying to make is that you should not adopt a defeatist attitude simplyt because you draw Italy or Austria. Being viewed as a less important underdog can actually work in your favour. The statistics prove it. ;-)

gjdip, theres actually a calculation for the fact that a larger sample size results in a smaller skew. if you sample enough people, youll eventually sample the entire population. calculating a population gives you true statistics. whether the graphs are skewed one direction or another doesnt matter because its based on the complete whole. but if you sample every game on this site (9k+) its a little time consuming. thus its easier to sample a smaller amount.

youll get much more reliable statistics if you sample 1000 games than if you sample just 10.

i totally agree that you shouldnt adopt a defeatist attitude with austria or italy. im not promoting that. the statistics are simply a means of breaking down information so that you can analyze it better. italy and austria tend to be harder countries to play, ill admit that as would anyone who has played them but they by no means mean a defeat.

keep also in mind that when sampling all games, player skill level wont be much of a factor since games are played containing all skill levels. typically those who have stronger skill will play in higher pot games, games which lower skill players are unable to join.

why would you ever be defeatist as Austria

the Anschluss is a draw AT LEAST

In my opinion:

1) Turkey
2) England
3) Germany
4) France
5) Russia
6) Italy
7) Austria

I've made some quick & easy reports in Excel, that could be used on the database of Webdiplomacy , to allow players to access the statistics.
I would need to access the DB first (or a dump or the DB), but I'll ask TGM for it soon (after the whole Leauge thing is up & running I suppose)

This data is a bit old, but its still good information

http://spreadsheets.google.com/pub?key=tb4yVZbw2ZXGt4aPKCwgKAQ&output=html

Its split into G-Rating Average above 150, 200 and All.

Guys need to Anschluss more :madface:

Wow, great to see :D

@ Acosmist: You're right about Germany. People massively under-rate Germany. With good diplomacy Germany can get three builds in year 1 and even with reasonable play you should get two. Germany has many opportunities for early alliances and is in a much stronger starting position than Italy, Austria, or Turkey.

You are also right that a lot of people don't seem very receptive to the idea of Anschluss on here, which has surprised me too after coming to the game with experience of face-to-face play. Too many Germany players seem to think "ok, Austria is no threat to me because he will be devoured early on", but also refuse to ally with Austria because they see Austria as weak. This generally leaves Germany involved in an alliance with either France or England against either England or France, which goes okay for a year or two until France or England stabs Germany, and Russia picks up the leftovers.

I have won five games, two of them as Austria.

And yes that is a very poor win rate. Mock me.

Well, Jamie, since we're bragging, I've been Germany three times, got the Anschluss all three times, and finished well (2 solos, 1 three-way draw). Austria did well in all those games (second place survival in the first, 16 supply centers in the second [btw, he ought to have held me to a 17-17 draw but he botched the ending], and participation in the three-way in the third game). Italy also did fairly well in all those games. I think Italian and Austrian players would do well to convince Germany to Anschluss more often, and, indeed, it looks to me like it's a good way for Germany to win more. Obviously, events can conspire to defeat the Anschluss (England and France can ally from the start, there could be a Juggernaut, etc.) but overall, the Anschluss should AT LEAST give better stats than those numbers disclose.

Shame, really. Not only if the Anschluss good but it's fun. And, hey, Austria doesn't die immediately!

"not only IS the Anschluss good"

typing fail ;_;

Doh. Actually on checking, three of my five wins came as Austria.

And I wasn't bragging. I actually drew attention to my poor win rate and said 'Mock me'. If anything that's anti-bragging.

But I do totally agree with you about the benefits of Anschluss for both Germany and Austria.

I was teasing you, bro.