Soloing as Italy

Use this forum to discuss Diplomacy strategy.
Forum rules
This forum is limited to topics relating to the game Diplomacy only. Other posts or topics will be relocated to the correct forum category or deleted. Please be respectful and follow our normal site rules at http://www.webdiplomacy.net/rules.php.
Message
Author
Enriador
Posts: 103
Joined: Sat Sep 30, 2017 1:15 am
Karma: 76
Contact:

Re: Soloing as Italy

#41 Post by Enriador » Fri Feb 05, 2021 8:25 pm

Oh I don’t disagree with you that the geography of the board makes it more difficult for Italy to win or that this is backed up by data. I just think your arguments for why this is the case are quite pedantic and require some very big assumptions :)
I am sorry if I sounded like that; like I said, I was trying to be thorough and cover all bases. No clue what you mean by "big assumptions" though, as any kind of theoretical debate on subjective matters (such as "is Italy hard to solo with and if so, why?") will have assumptions.

Trick is finding out whether they are close to the truth.
There is a much simpler explanation: Italy sits on the East of the stalemate line. The eastern half of the map has 5 landlocked centers and a number of key land spaces like Galicia and Armenia. However, Italy needs fleets to attack Marseilles, Spain and Portugal. Italy has difficulty winning because, more than any other country, they must split their forces between fleets and armies and between the eastern and western halves of the board.
You raise an excellent point regarding Italy's position on the eastern half of the Main Stalemate Line. Let's see your "big assumption" on saying Italy has to "split their forces" between Army vs Fleet or East vs West "more than any other country".

I will let Josh Burton - noted statistician on the Diplomatic Pouch for many years - show off some sweet maps:

SCs typically held by Italy when Italy solos: http://uk.diplom.org/pouch/Zine/F2007R/ ... solo-I.png

SCs typically held by Russia when Russia solos: http://uk.diplom.org/pouch/Zine/F2007R/ ... solo-R.png

SCs typically held by England when England solos: http://uk.diplom.org/pouch/Zine/F2007R/ ... solo-E.png

The data above confirms what you said about Italy going west and east - Marseilles, Spain and Portugal seem to be well within a soloing Italy's sphere of influence. It requires crossing the stalemate line and building fleets (also needed to take Turkish home centers!), while armies are a necessity to dive deep into Austria and Russia.

Interestingly, the data on England and Russia show similar aspects, but on different spheres. England (as most Diplomacy regulars know) cannot win with fleets alone - "there are no supply centers in the sea", a fellow hobbyist once said. To invade and hold the likes of Munich, Berlin, Paris and Marseilles, England (a naval-sided power in the early going) must cover up its continental presence with land forces. A soloing England, I suspect, has a similar ratio of armies-to-fleets compared to a soloing Italy.

Russia, on the other hand, has a whopping six supply centers on the other side of the Main Stalemate Line. Unsurprising given its status as the only power with home supply centers on either side of it, but we can safely "assume" that Russia can pull it off with a greater proportion of armies rather than fleets.

Italy has both characteristics at once, which is yet another drawback to add to Rome's long list of woes. However, little in the game of Diplomacy is, as you have put it, "simple". Beneath the clarity of its rules lies deep layers of profound mechanics that are interwoven in many different ways. Ascribing Italy's poor performance to just one (or two, or three) factor(s) is a simplistic view of the situation (pun not intended :razz: ).
Your linked source does not explain where they collected their data from or when so it is difficult to draw conclusions from it.
My linked source *does* explain where they collected their data, you just didn't follow it. Burton, in the very first lines, mentions his FRIGATE program to read through his dataset *and* that it is further introduced in his first article.

If you follow through his first article - http://uk.diplom.org/pouch/Zine/F2006M/ ... owers.html - you will see him mention Doug Massey's JDPR (which rated games from the DP Diplomacy judge, in a way similar to how Ghost Rankings used to work here). There were literally thousands of games stored there, most of which have been lost to time. Yet, thankfully, their analysis survived.

In case you wish to cross-check the findings, "the Scribe" has gathered around 600 games - http://uk.diplom.org/pouch/Zine/S1998M/Anon/1901.html - with similar records regarding who takes what in 1901, among other interesting stuff.
Now here you have not provided data and I do not see why this is necessarily the case.
I must admit I have no quantitative data for that. What I do have is a qualitative analysis by some folks you might have heard of; they are not necessarily 100% right but I suspect the truth is a little bit between what they said, what you said, and what most around the thread have been saying.

Sorry if it is a bit too long, but since we don't often see so many chipping in to discuss Italy I bet you will like the references (added emphasis in bold for those who just want to skim through it):

Baron von Powell, creator of 1900 , provides even more quantitative data and points out the Venice-Trieste dilemma as a cause of major friction and weakness (DP Zine, S2007M):
In a face-to-face game, the person who draws Italy receives condolences from the other players before the game even starts. Tournaments are designed so players won’t be "burdened" with Italy in more than one game. Player rating systems that don’t weigh a win or a draw by the Pope higher than a similar accomplishment by the President or Tsar are called into question. [...] I think the game of Diplomacy suffers if, as seems to be the case, a universal perception exists that some Great Powers are inherently stronger or weaker than others.
[...]
In an article by Melinda Holley that appeared in Diplomacy World #77 ("7x7 Gunboat Tournaments — An Analysis"), the results of twelve 7x7 Gunboat tournaments, a total of eighty-four games, are examined. Not surprisingly, the results show that Italy finished at the bottom. The Pope managed just one win out of the eighty-four games played. In comparison, France ran away from the field by winning fourteen games.
[...]
Out of 3723 conventional Diplomacy games recorded, Italy has won a paltry 221 times. Italy has also suffered 2968 losses. These numbers put Italy in last place on both accounts by a large margin. Austria-Hungary, almost everyone’s consensus pick as the next weakest Great Power, looks like a titan in comparison with 284 solos and 2899 losses. Russia, with 449 solos has lapped Italy, while France has lost 360 fewer times.
[...]
The first problem that jumps out at most people is that Italy’s SC in Venice is directly adjacent to Austria- Hungary’s SC in Trieste. No where else on the map does such a situation exist. Having SCs that touch can’t help but create tension between the two neighbors since each has to worry about the other opening the game with an attack. I think Austria-Hungary has much more to fear in this regard, but the problem does work both ways, particularly as the game progresses. Each Great Power is almost forced to have one unit pull garrison duty close to home to preclude the possibility of the other building in the Winter and attacking in the Spring without warning.

There is no doubt in my mind that this situation is annoying for both the Archduke and Pope, and contributes in some measure to each Great Power’s poor record. [...]
Steve Ray, scholar of the game (F2000M), in his (superb) article titled "Winning as Italy" exposes some of Italy's weaknesses - and singles out one in particular:
The single biggest wrecker of Italian chances to inexperienced or greedy Italian or Austrian players: those two supply centers (Venice/Trieste) adjacent to each other are a constant temptation/threat to the player across the border. Luckily, there is a simple truth that balances that: Nine times out of ten, when Italy or Austria attack each other early, both get eliminated early as well
Richard Egan, who if I am not mistaken also co-edited the zine (Vienna, #10) noticed, in particular, the contrast made with key border provinces such as the English Channel:
The adjacency of Venice and Trieste is certainly a unique situation on the gameboard. Inevitably it is a real handicap to both Italy and Austria, since from the opening moves each will be wary-of the other attempting to 'steal' their home centre, a security problem which often limits the options each is prepared to consider. France, for example, can order F(Bre) to MAO content in the knowledge that if England slips into the Channel, he can retrace his steps to minimise the damage and at least stand England out of his home centre in Brest.

By contrast, Austria and Italy have no such 'buffer', between them and the natural emphasis placed on home centres is exaggerated in Austria's instance by the fact that Trieste is that powers only coastal home centre, where fleets must be built. For one of the Weak 'Sisters to trust the other sufficiently to move out of the offending centre is a risky venture indeed - hence (the 'Balkan Gambit' title for Austria's optimistic F.(Tri)-Alb, A(Vie)-Tri, A(Bud)-Ser opening (and even this guards Trieste from Vienna, even if other versions of this particular 'Gambit' do not). Yet the alternative is for each to sit units like mother hens on each centre, hardly a constructive approach, and surely unlikely to benefit either's poor performance records.

All too often, one or other will move out at last, only to be stabbed, and it is the emphasis placed on the 'natural' unease between Italy and Austria that sponsors the lasting popularity of the 'Tyrolian Attack' opening for Italy (A(Ven)-Tyr, A(Rom)-Ven).
Charles Roburn, long-time writer at the Diplomatic Pouch Zine (S2007R), exposes a similar line of thought:
Undoubtedly, one of the main reasons is the way that Venice and Trieste border on each other. The Ven/Tri border represents a unique situation in standard Diplomacy: it is the only spot on the board where the home supply centers of two different Great Powers border directly on each other. This one fact alone is often held to be largely responsible for the generally poor overall performance of Austria and Italy relative to the other five Powers.
Allan B. Calhamer (Variations on a Theme), creator of Diplomacy. Yeah, the man himself:
"[...] Italy and Austria are considered to be the two weakest powers in Regular Diplomacy. One of the reasons for this is their mutual lack of security, because each has a home supply centre adjacent at the start of the game [...]".
I think I have already bothered you too much on this particular topic, so I won't engage any further. I am humbled by all the fascinating discussion you have been doing so far!
So the current hypothesis is that, in a high GR game, Italy will do relatively better than in the site average?
No idea, as that depends on the platform. Weaker games (such as those played by "uneducated" masses at Diplomacy apps far more popular than webDip) tend to have very weird results, which might differ from low-pot games in a place like webDiplomacy where the average player is somewhat more strategically-minded (we do have a subforum just for strategy, don't we?).

The reason is likely that people at low-level games simply play "almost" at random, as if in a gunboat game (no offense meant). That means safe countries (such as Turkey) often win far more often than in full press games of some quality, while Italy somehow manages to fare better than poor Austria (butchered by all sides as it is).

Whether that applies to webDiplomacy would be interesting to see! The Gunboat Statistics and Opening Database is great but Full Press information is most welcome.
1

AnimalsCS
Posts: 32
Joined: Sun Apr 26, 2020 7:59 pm
Karma: 51
Contact:

Re: Soloing as Italy

#42 Post by AnimalsCS » Fri Feb 05, 2021 9:28 pm

Enriador wrote:
Fri Feb 05, 2021 8:25 pm
My linked source *does* explain where they collected their data, you just didn't follow it. Burton, in the very first lines, mentions his FRIGATE program to read through his dataset *and* that it is further introduced in his first article.

If you follow through his first article - http://uk.diplom.org/pouch/Zine/F2006M/ ... owers.html - you will see him mention Doug Massey's JDPR (which rated games from the DP Diplomacy judge, in a way similar to how Ghost Rankings used to work here). There were literally thousands of games stored there, most of which have been lost to time. Yet, thankfully, their analysis survived.

In case you wish to cross-check the findings, "the Scribe" has gathered around 600 games - http://uk.diplom.org/pouch/Zine/S1998M/Anon/1901.html - with similar records regarding who takes what in 1901, among other interesting stuff.
Thank you for providing that link. The link to the article about the data collection in the article you sent before did not work – I tried it a number of times and searched the web for the author and "FRIGATE" to no avail.

Your wealth of sources is certainly helpful in any debate such as this!

Post Reply

Who is online

Users browsing this forum: No registered users and 54 guests