(A Personal View on Opening Strategy)
by Stephen Agar
Italy is the Power on the Diplomacy board which inspires both irrational obsessional love in some, while to others the thought of being stuck on four green units for years (in both game time and real time) brings only a Venetian sinking feeling. It certainly isn't going to give you your best chance of an outright win, as the stats suggest that of all Powers, Italy is the most unlikely to win the game (R=17.7%; G=17.13%; F=16.57%; T=14.89%; A=11.89%; E=11.14%; I=10.67%). If Diplomacy was a perfectly balanced game you could expect to win 14.29% of the time, but it isn't and if you are playing Italy then you certainly won't.
A good starting point for any strategy article is a quick summary of opening frquency (by the way, most of the stats in this article are gratefully taken from The Numbers Game Nos. 17 & 18 which are available from Richard Sharp and Steve Doubleday ).
Italian Openings >1%
(A) A(Rom)-Ven; F(Nap)-ION; A(Ven)-Tyr = 22.31%
(N) A(Rom)-Apu; F(Nap)-ION; A(Ven)Std. = 15.11%
(AF) A(Rom)-Ven; F(Nap)-ION; A(Ven)-Pie = 11.64%
(N) A(Rom)-Apu; F(Nap)-ION; A(Ven)-Tri = 11.38%
(A) A(Rom)-Ven; F(Nap)-ION; A(Ven)-Tri = 6.69%
(N) A(Rom)-Apu; F(Nap)-ION; A(Ven)-Tyr = 3.06%
(F) A(Rom)-Ven; F(Nap)-TYS; A(Ven)-Pie = 2.86%
(F) A(Rom)-Apu; F(Nap)-ION; A(Ven)-Pie = 2.65%
(N) A(Rom)-Nap; F(Nap)-ION; A(Ven)Std. = 2.50%
(F) A(Rom)-Tus; F(Nap)-TYS; A(Ven)-Pie = 2.45%
(N) A(Rom)-Ven; F(Nap)-ION; A(Ven)-Apu = 1.89%
(A) A(Rom)-Ven; F(Nap)-TYS; A(Ven)-Tyr = 1.74%
(N) A(Rom)-Tus; F(Nap)-ION; A(Ven)Std. = 1.33%
(A) A(Rom)-Ven; F(Nap)-TYS; A(Ven)-Tri = 1.17%
(F) A(Rom)Std; F(Nap)-TYS; A(Ven)-Pie = 1.02%
(N) A(Rom)-Nap; F(Nap)-ION; A(Ven)-Tri = 1.02%
(A) = Anti-Austrian; (F) = Anti-French; (N) = Neutral.
Before we go into details, let's step back and consider exactly what the long term strategic goals should be. Where will the 18 centres come from? Austria, Turkey & Balkans gives a total of 13, Sevastopol, Marseilles, Spain and Portugal gets to 17, leaving Munich or Moscow for victory. If England has Portugal sown up, then you'll need two out of Munich, Moscow or Warsaw. If you aren't in a position to stab Russia then you may need Paris, Brest, London and Livepool for victory. From that it would appear that you need to control the whole Mediterranean to win the game and with luck break out into the Mid-Atlantic. Austria is often an early Italian target, but if dominating the Mediterranean is a key aim then it is clear that Turkey is the real enemy.
Italy is clearly a difficult country to play and to win you may need a little luck if you are not going to be stuck on 4 centres for the whole game. In the first year Italy has three main options: (1) attack Austria ruthlessly (37.73%); (2) head west and attack France (14.8%); (3) wait and see (36.29%). The figures in brackets are arrived at classifying the most common Italian openings into these three groups. Let's look at these options in turn.
It is difficult to estimate how often Italy really attacks Austria in 1901, because some moves which look aggressive (eg. A(Ven)-Tri) may in fact be agreed stand-offs. The Tyrolian Attack (A(Ven)-Tyr; A(Rom)-Ven) is the most aggressive anti-Austrian opening, as well as being the most popular Italian opening, the object being to have a supported attack on Trieste in Autumn 1901. If Austria orders A(Vie)-Tyr (which he does 7.39% of the time) then you end up a lot of egg on your face as neither A(Ven) or A(Rom) can move. More common is the Austrian move of F(Tri)-Ven at 23.25% which means that A(Ven)-Tyr suceeds, but it is left stranded as A(Rom) is stood-off over Venice. Putting all the figures together suggests that if you resolve to try the Tyrolian Attack there is a 70.93% chance of complete success, a 21.68% chance that A(Ven) will be isolated in Tyr and a 7.39% that neither of your armies will move at all. Looks like pretty good odds to me!
The Tyrolian Attack is most effective when combined with a Russo-Italian alliance. In 30.48% of games there will be a Russian A(Gal) after the Spring 1901 moves which can help in the quick dismemberment of Austria - with Russian support A(Tyr)-Vie is almost guaranteed to suceed. On the other hand if Austria covers Galicia and Russia does not attempt to move there (20.07% likelihood) there is no A(Vie) to interfere with your attempt to take Trieste.
The Austrian Attack (A(Ven)-Tri; A(Rom)-Ven) is played less often (6.69%) presumably because if Austria orders F(Tri)Std., F(Tri)-Ven or A(Vie)-Tri it is a disaster and there is a 56.35% chance that Austria will make one of these moves. Of course if it works it is excellent.
The attitude of Germany is very important when assessing the likelihood of success when attacking Austria. Some Germanies will make it clear from the beginning that they are willing to intervene in Tyrolia in support of Austria if necessary. If you judge Germany to be serious and to have the means to keep his promise, that could well tip the balance in favour of a Lepanto or a French Attack. It is worth remembering that if everything goes wrong and you end up with an army in Tyrolia but without prospects for early success against Austria, you can always change directions and order A(Tyr)-Mun (preferably with French support).
The problem with playing the odds and attacking Austria is that it is all too easy to get bogged down in trench warfare very early on, leaving yourself with a hard slog while Turkey grows strong. An almost inevitable consequence of attacking Austria straight away will be to give Greece to Turkey, which means that Turkey will inevitably build F(Smy). You have been warned.
As I noted above, it is relatively rare (14.83%) for Italy to make an opening which could be construed as anti-French. Even the moves which could be anti-French may not in reality be anything of the sort. Although Italy orders A(Ven)-Pie 21.79%, there are no statistics as to how often Italy tries A(Pie)-Tyr in A01. The move to Piedmont may be to disarm Austria into thinking he has an ally in Rome. Similarly the move F(Nap)-TYS (11.31%) isn't really anti-French at all if it meekly takes Tunis itself in A01 rather than convoying an army there. I would not move F(Nap)-TYS if I was just going to end up taking Tunis with the fleet, as it would be more useful to have some leverage on Greece in A01. The true French Attack is A(Ven)-Pie, F(Nap)-TYS, convoy the other army into Tun in the autumn, and order F(TYS)-GoL in S02. The trouble is France sees it coming a mile off.
If your instincts are to attack France, there is a lot to be said for not doing it until 1902. An "attack" in S01 can usually be explained away on the grounds that it isn't an attack at all, but hostile moves in A01 will undoubtedly trigger builds of A(Mar) or, worse still, F(Mar). It is better to leave France until you have built a second fleet and then attack in S02, hopefully in concert with England or Germany or maybe both. You are unlikely to make much headway until 1903, unless you can sneak a fleet into Spain via WMS. On the other hand, once the French position crumbles, it collapses very quickly. In time Italy should be able to pick up Spa, Mar and Por as well as taking possession of MAO.
The down side is that while you're fighting France, Turkish fleets may well be massing in Gre, AEG and EMS and once ION falls you're in big trouble. For that reason, unless you decide on a firm alliance with Austria and/or Russia which is guaranteed to keep Turkey pre-occupied, it may be wiser to leave France alone until the Balkan triangle is resolved.
Wait and See : Try a Lepanto?
The problem for Italy is always where will the fifth centre come from? All Italy's can take Tunis without difficulty, but will the next centre be Trieste, Marseilles, Greece, Spain or Smyrna? It is impossible for a thriving Italy to go beyond 1901 without a clear alliance and a clear strategy.
The advantages of Wait and See is that it gives you a chance to reflect on the emerging alliance structure between Austria, Turkey and Russia. At the risk of making sweeping generalisations I do not believe that Italy and Turkey are really compatible as alliance partners - control of the Mediterranean is to important for both of them for such alliances to work in the long-term. Therefore, much diplomatic effort should be put into isolating Turkey from her neighbours. If at the end of 1901 it looks as though there is a Russo-Turkish alliance on the cards, then there is much to be gained from propping up Austria while playing a Lepanto.
For those who don't know (and I'm sure most of you do), the Lepanto system is whereby Italy convoys an army into Tun in A01, builds F(Nap) and then orders F(ION)-EMS, F(Nap)-ION. The idea is to convoy A(Tun) into Turkey in A02 (usually Syria). Once the Juggernaut gets rolling Austria cannot hold out for very long and Turkey will get enough centres to fuel her expanding navy, so the Lepanto may be your only sensible option.
It is harder to determine a strategy if it appears there is a Russo-Austrian axis. My gut reaction is still not to ally with Turkey - Italy's natural enemy - but instead to try and participate in the Turkish carve-up, while priming Russia for a stab on Austria. In these circumstances it may be necessary to garrison Venice, even if it is a waste of an army.
Finding Friends : The Venice - Trieste Dilemna
This is a difficult one. Assuming you don't engage on an outright attack on the Austrian home centres, just how do you cope with the fact that Venice and Trieste border each other. It's all very difficult. Just as kids may say "show me yours and I'll show you mine" it is easy to resort to "you move out of Trieste and then I'll move out of Venice." Let's be honest - Austria has more reason to fear a stab for Trieste, which in turn borders both Vienna and Budapest, then Italy has to fear a stab for Venice. Try to demilitarise in a Spring move, but also try and leave yourself in a position to retake Venice in the Autumn if necessary. If Austria vacates Trieste in 1901 and then doesn't build in Trieste, that would be a good opportunity to vacate Venice if you can persuade Austria of your good intentions.
One popular Austrian opening is the Southern Hedghog (F(Tri)-Ven, A(Vie)-Gal, A(Bud)-Ser - 17.47%) the aim of which is to put defence first and avoid the early bath. Overall the F(Tri)-Ven move is used some 23.25% of the time. If Austria tells you he is going to play defensively, then it may well be wise to abandon thoughts of a quick attack and go for a Lepanto.
Odd though it may sound, I think the Austro-Italian alliance can work if circumstances make it necessary. Turkey needs the Mediterranean to win because by the time the Balkans is resolved Russia will either be too strong to attack or northern Russia will already be under pressure from England and/or Germany. On the other hand, with your help Austria can sort out the Balkans quickly and strike into Russia via Galicia a good couple of seasons before Turkey could probably try such a tactic. The very tension between Trieste and Venice means that once an Austro-Italian alliance is established, it can be very difficult for one to stab the other, simply because both Powers will already be paranoid in that respect.
Undoubtedly, the strongest alliance in the early stages of the game for Italy is a Russo-Italian alliance against Austria and Turkey. If the alliance holds, in the end you should be able to carve up the Austria, Turkey and the Balkans between you provided England and France don't start to create trouble for Russia and Italy respectively. Such alliances work best if you can play Austria and Turkey off against each other in the early stages. What is more, Russia doesn't need the Italian home centres to win the game, nor is Russia likely to build any more southern fleets, so Russia isn't as potentially fatal as an ally as Turkey. For Italy to prosper, it is important that someone goes under early - preferably Austria, Turkey or France, but so long as Italy is presented with some area where she can bring her forces together in strength to pick up centres it really doesn't matter.
Winning is the name of the game, so where will Italy pick up 18 centres? My best guess is Ven, Rom, Nap, Tun, Gre, Tri, Vie, Bud, Ser, Bul, Smy, Con, Mar, Spa, Ank, Rum, Sev, Por - though if you get Mos and War you wouldn't need Spa and Por. The crucial thing is to get past the stalemate line, which means getting Moscow and Warsaw at one end or Spain and Portugal at the other. You will note that this list is very similar to the likely 18 centres that Turkey would need to win. Austria on the other hand, can win by leaving Italy alone and breaking into Germany - which is possible given that she will build predominately armies.
Richard Sharp commented in his book that "in a high standard game I would put Italy's chances of winning at about zero, I'm afraid." Sadly, that may well be true. If you've recently drawn Italy in a postal game, just look upon it as a challenge!
First Published in Spring Offensive No.18