I did a compilation of statistics for the highest finished pot games, which I presume is available by clicking "Finished games" under "Games" above. The games were retrieved on 19 Feb 2008.
First, my premises:
-High pot games are probably most likely to be representative of the better players. While this is not necessarily guaranteed to be true once the pots dip below 700 (wherein a new player can join), these games are still more likely to attract top players than 35-point defaults, and at any rate, a new player will be just as likely to fall into any power, I presume, so the noise, if all powers are equal, should be cancelled out with enough samples.
-I presume that the games available for analysis (that is, without hammering the server!) are representative of the whole for competent players. This may be a slight reach, but I don't think it is much of one.
-I presume that the game will be best represented among better players, who are more aware of the strengths and limitations of each power. The top finished games that are displayed achieve this goal in my opinion.
-The top 100 games in terms of pot were used in the calculation. Believe it or not, only two powers (one France, one Turkey, I believe) finished in civil disorder and still alive out of all these games. I will count them as still alive for statistical purposes, but 2 out of 700 really doesn't make that much a difference either way.
-If a nation has 0 at the very end of a turn where someone wins, I count them as eliminated, even if they have centers left. (The game summarizes number of units, so I had to settle for this "approximation".) The alternative is extensive access to a database that I don't have.
Of the 100 games, 11 were draws, with the following splits once each (alphabetical order of powers in each draw):
AGT
EFT
AEG
AEFGT
FG
FT
EFRT
EGIT
AE
AEGIR
FIR
Victories per power (89 total out of 100):
Austria, 15
England, 13
France, 15
Germany, 9
Italy, 6
Russia, 21
Turkey, 10
I was not in the least surprised about Italy bringing up the rear. This situation has always been true for the Italian player, ever since the days of postal play. Russia's number surprised me, since France has historically been called the "best" power in terms of raw winning percentage.
Remember that these are out of 100 games with 11 draws, so Italy has a 6% chance of winning in well-played games, and Russia a whopping 21%.
Austria, Russia, and France combined win right about half of the games played.
I was really surprised that England was fourth. I expected them to be 6th, Italy 7th. (Austria has historically done extremely well in games they *survive*, but not that much. Still that wasn't really surprising.) Apparently unsupported convoys are not that big of a handicap. Turkey has the benefit and drawback of a corner position, but so does England, and England does better. I think England has more mobility than Turkey, and an ability to move fleets all the way around your home centers makes it easier to go from offense to defense and vice-versa, I would suppose.
At any rate, a nation's fair share of wins is 1/7th of those games won, 89/7 or about 12.7% of the games in this sample. This actually puts England right about (slightly below) average. Thus getting England is a fair cop.
Germany's performance was a little disappointing to me, since Germany isn't a nation that I remember being mentioned as a players' bane. Is it possible that the convoy support deficiency actually works in *reverse* for a German assault of the British Isles? France presumably can get more fleets out there to flank English units and surprise England with the exact location of the landing, while Germany can only dump it on one coast. I do know that Germany versus Russia is a more likely explanation, since Russia doing so well might imply German strategic weakness against them (but that's just speculation).
If anyone knows why the Great Bear does so well relative to the other powers, let me know. I think the 4 units thing is overrated, and while I have won my only table top game as Russia, I completely botched the diplomacy for my current game as the Czar. (It didn't help that Germany and Austria were new and both collapsed early without any spoils for me, leaving two double-nation alliances and... me.)
Survivals in the 100 games (including draws):
Austria, 43
England, 63
France, 59
Germany, 43
Italy, 48
Russia, 43
Turkey, 61
These are very surprising numbers. Austria getting eliminated in 57% of all games should surprise no one. Germany having the exact same percentage gives one reason to raise the eyebrow, but Allan Calhamer wrote some time ago about Germany being just as connected, and thus as vulnerable, as Austria. But Russia *also* getting the boot in the exact same amount of games? This is strange, especially since Russia wins more games than all other powers in terms of absolute numbers. England and Turkey have always had strong corner positions, and always survived a higher percentage of games. I expect that these numbers are lower than in the older postal Dip games due to the fact that no direct mechanism for draws are possible. 11% is not the normal percentage for draws "in the wild"; it is much higher, though I forget exactly what. Italy is also usually a good survivor, so the fact that they get wiped out of 52% of their games is a little surprising as well. France makes it to the bitter end of 3 out of every 5 games.
The percentage of games won wherein the country actually survived to the end should give a little more insight into what a strong Diplomacy player can expect against similar competition:
Austria, 35%
England, 21%
France, 25%
Germany, 21%
Italy, 13%
Russia, 49%
Turkey, 16%
(Note: Percentages should not add to 100%, since these numbers are biased towards victories. After all, all games you win are ones where you've survived.)
That's right; if Russia survives, flip a coin. Heads? They win. Austria's number is commensurate with postal history. Being in the middle stinks, until you're able to hit whoever is weak in the midgame. Poor Italy not only wins few games, but gets killed half the time, and only wins a small share of games in which they survive. (I bet this will be what most players use "Abandon game" for when it comes out!) England and Germany share the exact same percentage for winning survived games. The Western powers are fairly even, but Turkey shows what being backed up in the corner will do to you: it's hard for someone else to get in, but it's also very hard to get out to go win.
Now back to draws. Nations participated in the 11 draws with the following frequencies:
Austria, 5
England, 7
France, 6
Germany, 6
Italy, 3
Russia, 3
Turkey, 6
Italy and Russia can expect to win or lose; there is no draw. (This is good for many-win Russia; yet another knock against Italy.) The others are fairly even, but look at England again. While not exactly statistically significant, it is interesting that England drew more than any other power.
Draws as a percentage of survived games:
Austria, 12%
England, 11%
France, 10%
Germany, 14%
Italy, 6%
Russia, 7%
Turkey, 10%
(Again, these should not necessarily add to 100%, since draws are a rare occurrence.
Germany on top is something to note; maybe that should be their goal from 1901!
I wouldn't call this an expose', but it should shed quite a bit of light on the outcomes of the games we see on this server. I personally learned a lot doing this, and I suspect that others will as well. This is also a great newbie guid to let them know what to expect (and what others expect of them) in their new games, especially since playing each power is an entirely different game.
My take:
Italy: If a draw vote option ever becomes available, I would urge Italy to take it in Spring 1901 and keep it checked all the way through the game, since that's their best outcome. Winning is simply not a viable option. 1 out of 16 games is futile, especially of the other players know what they're doing. I would expect that an easily obtained draw would dramatically increase the number therein, which can only be good for Italy, since the alternatives are rotten. Then again, kill Austria, and... um, oh yeah, Russia is stronger. Anyone got a good idea here?
Germany: Play to survive and draw. This appears to be the strength of the nation, and drawing means you're (usually) stronger than when you started. Winning may well be out of the question, but at least there's points to be had by finishing, right?
Austria: Survive 1902 without getting stabbed. This is quite obviously hard to do, since the best in the world all know that Austria is weak in the early going and it still gets crushed so often. How many games have you played in where the talk is of partitioning a nation, down to the center, in 1901, it soon comes to pass, and that nation hasn't been Austria? At least you know that, if you draw Austria and survive through 1902 without the stab, that you are in great position.
Turkey: Not good. You don't win a bunch of games, you draw some, and you get clobbered some, but you lose a LOT. You sit there and lose a lot. At least with this information, you have the chance to ally with England against Russia (something that very few of the three powers think about), and England should be glad to do so. Given Italy attacking Austria with nothing else to do, and Austria holding them off but not getting anywhere, and you have a much better chance. Russia and Austria win 36% of all games, and they're right next to you, so get others to attack them. You can't really attack strongly until 1903 anyway, but get in position to do so. Any Turk unit beyond Bulgaria-Aegean-Black-Armenia-EastMed when Spring 1903 rolls around probably adds 5 percent to your chances to win. That should be your opening goal for the first two years.
England: Your best diplomatic tool, "I can't support my convoys", just got thrown out the window. The good news is that, since Russia wins 1 out of 5 games, there's something else to use to gather pity from other nations. Obviously you need to contain Russia in Scandinavia to have a better chance at victory. Germany getting stronger isn't as much of a threat as France, or especially Russia. England is a nice average power. You win some, you lose some, you draw a lot, and it's a fair selection.
France: You're good, everyone knows you're good, and unless E-G try to wipe you away in 1901, you get two free builds. You are one of the top 3 powers on the map; the other two are your only real threats in the long run. Unfortunately, neither are your neighbors, and while Italy and Germany stink it up, England is a reasonable power. Your strategy must be one of diplomacy if you want to be the "best of the rest" when the midgame comes around. Orchestrating the partitioning of Austria should be a simple task (I mean, we all know it's coming; even Austria knows it's coming, and yet it still keeps happening, so you probably don't need to do much here). Russia, on the other hand, needs to be controlled and attacked vigorously. An A-T alliance is vital to stopping Russia in 1901 from getting a southern build, but Austria gets eaten so quickly that you may need to look north, and to be patient. While an E-G looks extremely dangerous to France, it will be able to rip Russia into nothing starting in 1902 if France is not interfering. Of course, an E-G that heads east can just as easily head west, so you must do the best early diplomacy to get into strong midgame position.
Russia: Okay, so this actually is an expose' for one segment. Your gig is up; now everyone knows about your wins. Russia wins so much in phpDip, I bet, because it can cross the stalemate line (across the map from the Baltic area to North Africa through Switzerland) in Scandinavia with ease, and that few powers understand this or use it, since draws are so rare. To win, though, you have to survive, and it is extremely difficult to do so on two fronts, and against 4 powers (EGAT). What's worse is that France is outside your early sphere of influence. I am obviously not qualified to comment on phpDip Russia play (since I botched it *so* artfully), but it seems that, now that I know this information, it would make sense for Russia to hold off Turkey, share Austria with Italy, then get the two to attack each other (in the Balkans, there has been infighting for over a hundred years straight already, and the centers do the same in Dip). Let the West sort itself out, even letting Germany take Sweden (since he can defend you better than attack you, and then England will thus have nothing much to gain with fleets from an assault on Russia), and hopefully swing your forces into Germany just as I-T go at it and one of the western powers bites the dust. You should be able to gain 18 centers without killing Italy or England if all works well, but any antagonism you show to a Western power in the first two years is going to haunt you. If you survive, you have a 50-50 shot. The timing, I bet, would work more often than not.
Any comments or suggestions?
Sparta game
)
)
)
)
)
)