I've always thought that Turkey was more defensible from a tactics standpoint and that England was more defensible from a negotiating standpoint. If that makes any sense. Though, I do agree that France should also be in the conversation.

Ok, this is hilarious and shows you how careful you must be when drawing conclusions from statistics. I've been working on a program that calculates effectiveness of opening moves for the last few days. I fund that the most common Italian opening is Nap-Ion, Rom-Ven, Ven-Tyr. It results in a win in 5.72% of the games where it's used. In fourth place in popularity is the classic Italian opening Nap hold, Rom hold, Ven hold. It results in an Italian win in 5.98% of the games. NMRing is more efficient than attacking Austria, it seems.

I have spent the last hour going through the code looking for errors, and finally, I manually went through over 50 games where Italy NMRd in the first move and ended up winning. I can't find any error in the code, except for some minor kinks that I need to work out before I release it, and the finished code will have to exclude "bad" games (CDs, NMRs etc). But this was too funny, I had to share it.

HAHAHA italy is such a freak

Vaftrudner
Any chance of getting the player stats on average number of points per game by player by Game type ...... i.e. Classic Full Press & Non=Press/Classic Gunboat/etc ,etc

I know it's a very crude measure of Diplomacy skill but still interesting nonetheless

I'm not really sure what you mean by that. Do you mean how many points a player earns per game on average?

Note: Never attack Austria as Italy! If you absolutely must, enter all hold instead.

Yes, if they have played 100 games, regardless of whether they won how many SCs were they left with at the end of the game. Even if they leave before the game finishes and they have multiple SCs they still had to earn them so all games played could be included. I guess it won't show the games people played, then left and were replaced. I guess you may need to add a zero SC count for every player who was defeated.to give you an average per game.....

Yeah I can easily run that, here is yours:


+-----------------+---------------------+
| pressType | avg(supplyCenterNo) |
+-----------------+---------------------+
| Regular | 3.8169 |
| PublicPressOnly | 6.4167 |
| NoPress | 4.5522 |
+-----------------+---------------------+

Someone, I believe it is lando, uses the Italian NMR was a strategy in gunboat games. Sounds odd though.

Vaft
That's cool but it itself reasonably meaningless but in a table with all other people who have played 100 games or more it becomes more interesting. The only relevance to the 100 games is that the list a bit smaller :-)

Been saying for a while that France was the most defensible power! Turkey's defensibility is overrated, big time; I remember once last summer that Scmoo472 and I happened to draw Russia and Austria three times in a row in consecutive live games, and each time we had a different Turk who never really talked much, so both being fairly chatty players, we quickly ended up deciding to smash Turkey all three times. By the third time we'd perfected the art of KOing Turkey by 1903 unassisted by Italy. The thing about Turkey is that if it can keep Russia out of the Black Sea, it becomes infinitely more defensible... absent BLS, you need the Lepanto to work (which is no guarantee, as it's easily blocked) or else you're faced with slogging it out until 1904-05+. But if Russia manages to score the Black Sea in S02 (we pulled this off all three times), Turkey's position is broken, and it's just a matter of time.

I would say England is about as equally defensible, but not in the same way. Where Turkey is predicated on keeping hostiles out of 1 or maybe 2 specific places on the board, England is predicated on your ability to avoid an early convoy to the home island. Once that army goes it's basically all over - it's rare for England to keep its army at home, so your army has a massive advantage in horizontal movement (able to jump immediately from Yorkshire to Liverpool, and support moves to Edinburgh from Liverpool), and England, like everyone else, has some very weak non-SC spots (Yrk, which covers all 3 SCs, and both Cly and Wal which touch 2). England is probably more defensible than Turkey in that while invading is easier due to having five easily-accessible sea provinces to use as launching sites for the invasion, actually killing it once the marauders land is much more difficult.

And then there's France. I've never seen France die before 1905. I've managed effectively to kill it by 1903, but both times were as Germany with a willing Italian ally, and involved breaking both Burgundy and Piedmont DMZs on the first turn; I challenge you to find a country that could survive 2 hostile neighbors breaking DMZs right off the bat! France has the frontal access of Turkey with the internal defensive cohesiveness of England, and an extra SC, to boot.

I completely agree with you PE, that France is the most defensible country. In that last example, I'd say France made the mistake of agreeing to the DMZs in the first place... Better organize at least one bounce in either of the two provinces, right?

I've gotten to the point that I always arrange a bounce in Burgundy on the first turn, or announce my intention to support myself into Burgundy. It's just too dicey not to make sure something's happening there. Italy doesn't often attack, so not as big a deal there, you can deal with it when you face it.

Define defense

NigeeBaby, I wrote a script to do just that just to see if I can. Does anyone know where I can just upload an html file? Google Spreadsheet really fucked me over when I tried exporting to them. Anyway, I can at least paste the top 10 here if you're curious Nigee.

Regular games:

No Average SCs Name Games played
1 12.2745 TheMasterGamer 102
2 12.1594 BigZombieDude 69
3 11.7358 dave bishop 53
4 11.5285 Rait 123
5 10.8621 rhino86 58
6 10.5171 Ivo_ivanov 350
7 10.4677 Vaftrudner 62
8 10.4127 Treefarn 63
9 10.3810 Noirin 63
10 10.3545 MadMarx 110

Gunboats

No Average SCs Name Games played
1 8.7606 V+ 71
2 8.3922 skoczek86 51
3 8.2184 Barn3tt 412
4 8.1176 EmmaGoldwoman 51
5 7.9156 Sargmacher 225
6 7.9128 SplitDiplomat 195
7 7.6491 jc 57
8 7.6316 Gary 57
9 7.5234 5nk 128
10 7.4198 IKE 81

I've got a Dropbox account if that is any good...........

I'm curious where I rank on that as well, actually. Though I probably don't have enough games yet for a fair sample size.

I made 50 played games the minimum requirement to keep a good sample size. CSteinhardt, you're even newer than my database dump :) It's from April 8.

Nigee, I have a Dropbox too, not sure how to link stuff publicly.

Or I can just get google docs working thanks to rokakoma.

https://docs.google.com/spreadsheet/ccc?key=0AuG-MFkgoLmedGxPdjFFMVhsMkhXdVVwbUpYaGNrMmc&hl=sv#gid=0

welcome :)

OK, well, if you feel like a new database dump, I'd be interested in the result then. :)

How about France being able to defe against an Italian+1 invasion. I think that really shows a weak spot in France's defensibility.

Vaft
Many thanks
Very interesting stats and shows some interesting differences between the two game formats which I thought it may do.
Not sure how old your date is though because on the GR I'm usually down at 3000th and on your gunboat figures I'm 84th so I guess there is shome mishtake or the dataset you've chosen is different
Nigee

I greatly enjoy how much thought and effort has gone into this thread.

"In fourth place in popularity is the classic Italian opening Nap hold, Rom hold, Ven hold. It results in an Italian win in 5.98% of the games. NMRing is more efficient than attacking Austria, it seems."

Maybe I'll try that next time instead of Alpine Chicken.

You guys are nuts. A snail's reproductive organs are not located in Turkey, England, or France. They're located in the snail's head, just like the card says.

Also don't forget that early edition of Trivial pursuit would predate the advent of internet diplomacy. And if my knowledge of by-mail diplomacy history is accurate, in that, both the Sealion and the Juggernaut were very uncommon alliances, which would tend to skew Turkey's defensive skills downward, and people not using the Sealion would tend to skew England's defense upward. So the question may still be wrong, but at least we can understand why they reached that conclusion.

Also I'm pretty sure when the game was designed the intent was that England would be the most defensive country, even if in practice that didn't play out like they wanted.