I admit I said I'd prepare an article on the endgame, but looking back over my series thus far I noticed I've been so busy dealing with the trees, I've not said as much as I should about the forest. So I devote this article to the central aspect of my approach to long term strategy, and indeed a key aspect of any type of Diplomacy deserving of the name: the relationships you build with other powers. As such, it will be even more subjective than my other articles, and perhaps sound more like a rant, but c'est la vie. On the bright side, it has more application outside of Machiavelli than most of the other articles in the series.
Every action you take in the game has a cost. Part of this cost is made up of the obvious tactical aspects: the ducats involved, the strategic and economic value of the territory taken and lost, even the opportunity cost of not taking a different action. The other part of the cost is the effect the action has on how the other players view you. More is at stake than just their answer to the question "does their action benefit me?". Perhaps more importantly is how they would answer "What does this say about whether I want to be allied with this player?". This is impacted not only by the moves, but what you've said about how you're going to move. Lying about your moves should only be done when the gain by doing so outweighs the costs; how you evaluate those costs is one of the most important aspects of your diplomatic strategy.
One of my personal pet peeves, something that will tend to turn me against a power faster than just about anything else, is pointless lies. To lie for no good reason is to shout to the world "Never trust me! I'm a fool!". If you don't want to reveal your moves for fear of a leak, simply don't reveal them. If you plan to fight a power, and you can't gain a major advantage against them by lying, then tell them directly that you will be fighting them. Be up front about it, even tell them why if it's not critical information. You might need to work with them later. You might need to work with someone they talk to later. Most importantly, you don't want them to decide that you are so unreliable that it is better to make sure you are removed by someone more reliable.
From the very beginning of the game, you should be figuring out what powers you want in the endgame with you, and start steering events toward that end. Taking territory is important of course, but in order to win you need more than that. You need an endgame situation where players will be acting in such a way that you can win. There are several scenarios you can set up to get to a win, but most require at least one (preferably more) of the people at the end to trust you highly. I take the endgame sufficiently seriously that I'm happy to skip opportunities to take out players I want at the end, even in situations where I could easily absorb them. I also like to keep the player count high as long as possible except insofar as powers convince me that they need to be removed.
From a game theory perspective, your goal is of course to win the game. The question then is what is the mostly likely way to get there. As "carebear" proponents and opponents both often point out, in standard it's pretty easy for 3 players to band together, vanquish the rest, and take a draw. The opponents of "carebear" play point it out to mock the fact that there isn't a whole lot of game there, and that a draw isn't much of an accomplishment. The proponents point it out as a way to 'share a win' or whatever terminology they pick to try to make a substandard conclusion look better:). The important lesson from all this is not how to quickly get halfway through a game then stop- it is that closely cooperating powers that trust each other can beat a larger numbers of less coordinated powers. My approach simply takes it a step further, and continues the coordination long enough to virtually guarantee one of the players involved will win. If three powers used this approach in standard, one could say they increased their chance of winning from 1 in 7 to 1 in 3; except it usually doesn't work in standard. The existence of stalemate lines, and involvement of three powers makes it likely that they'll be stuck with a three way draw. Indeed, they may be decreasing their chances of success by prematurely removing the other powers.
In Machiavelli however, two powers is sufficient to gain a large benefit over the others. There are no stalemate lines. The geography makes a race more practicable than in standard. And one can even proceed through the whole thing honestly- you don't need to stab outside the alliance, they'll one by one turn against you as they figure out what is going on. Keeping it so that it's 'one by one' is incidentally a key important skill to continue rolling through them. If you make the decision to keep to the strategy from the beginning, and you and all the other powers are otherwise average, I'd estimate you've raised your chances to win from 1/8 to 1/3 (leaving about a 1/3 chance that some alliance of the other powers will form in time to stop you, and no reorganized alliance will bring you to victory).
So, now we come to the usual variation. You'll find that a lot of players will act on the following postulate: "If I remain true to the alliance for a while, but forsake it at the right time, I can increase my chances to win." Stated that way, it's almost a tautology, but that ignores three details. First of all, the question of the definition of 'right time'. Most people judge this badly, and in so doing actually decrease their chances of winning rather than increase them. Secondly, there is the chance that your ally has laid the proper diplomatic groundwork. That once he is stabbed, the other players will join in a "stop the leader" alliance against you, and he'll win the game in the reaction. I've done this a few times. Finally there is the chance that you'll slip, and your opponent will figure out you're going to stab. This is the most important one for me- I do a lot better at convincing people of things I believe to be true. In those times that I do stab someone, it's usually not set up far in ahead, but rather more of a decision made based on a heavy analysis of a situation when it arrives.
If you can avoid those pitfalls, then you probably can stab for a win instead of racing for a win. Just be careful. And don't let short term concerns outweigh your long term plans.
If you decide to play it straight, remember you opponent might not.
Be careful as you reach the point where he might stab for a win to take
the proper tactical steps to make a stab less than crippling. Keep those
lines of communication open. Focus on the weakening or removal of those
opponents less likely to help you if the alliance breaks up. Think through
how you'd build a "stop the leader" alliance against your partner regularly.
If you have the edge in the race, be willing to amend the plan such that
they will be willing to keep putting off the stab until you can't be stopped,
rather than sticking to it hard and fast and leaving them feeling that
they have nothing to lose by stabbing you.
When you consider all these factors, you realize that as you grow you
will trade off players perceptions of how you hold each of these strengths
for others. A very simple example: As the Pope, if you take over Florence
on the first turn by surprise, you suddenly increase your military strength
in several areas, but you have a significant impact on short term financial
and military strength, while perhaps decreasing your diplomatic strength
(depending on how you set the stage for your attack of course). Ultimately,
you probably want to be viewed as strong enough to be a good ally, but
not so strong as to force an alliance to form against you. This is tricky,
and relies on determining how your fellow players will react, how well
they rate and value each category. The features you look for in an ally,
and how you relate to the ally should be adjusted to take these factors
An important reminder here- spending on bribes is generally not as effective as spending on units. A power with 60d in savings and only a few units is at a severe disadvantage against a power without savings but with 20 units on the field. 3 purchase bribes later (or even 5 disbands) and they are still outnumbered, unless they have managed to finish off their opponent.
Of course, it's possible everyone will keep their agreements. In this case, it's a good opportunity to stockpile some resources for the future, balance your forces, and even set up some side bussinesses. Some that I've used successfully in the past to give me reasons to have a non-offensive public presence and gauge the relative hostility of the others include buying and selling assasination chits, and offering services as a bank with slightly better rates than the game bank as well as allowing investment (if it's enabled; if it's not, you can charge worse rates). Sometimes I've even made money off such ventures, but that's not the main purpose. Instead, I find out what values other players place on various commodities, how far I'm trusted, how naturally suspicious everyone is, etc. Also, when I come through on my side of public bargains, it increases my reputation in general. There is another subtle yet important effect for several of this type of schemes- it can act to level the playing field between the other powers. If you're not growing, you want a slowly developing game. So if weaker powers tend to give you more bussiness and they benefit thereby, you buy time.
The time usually comes however when you must select enemies. The geographical influences I've discussed in past articles of course play a part. The way you are trying to steer the game (who will stand and who will fall) also play a significant role here. Another consideration, is that you generally earn more loyalty (and a greater share of the spoils) helping a weak power against a strong, than teaming up with strong powers to plunder the weak. Things may move more slowly, but as long as you make sure you are keeping any strong power from developing too far, that's not necessarily a bad thing. Patience is a virtue.
Sometimes, proper selection of a power to fight will ensure peace with several other powers. A series of "stop the strong power" type alliances can work well, as long as you don't burn your bridges along the way, or get other people too used to uniting in random alliances. The more dependent they become on you to provide strategic and tactical planning the better, so long as you avoid the impression that you are dictating strategy. An exception to this is if you're working with people who can't spend enough time on the game or are newbies who aren't making good decisions. While a slow battle is OK, you sometimes have to put your foot down to avoid everything falling apart, despite the risk that the big enemy will use this to create resentment to tear the alliance apart. If you do this, it's important to continue to provide reasons for things, and keep explaining the basics. Several times I've had allies turn on me because something that was blindingly obvious to me about the strategic situation to me looked different to them, and they turned on me at the wrong time and gave an opponent the game. When players are replaced is the worst time for this sort of thing.
If you wish to e-mail feedback on this article to the author, click on the letter above. If that does not work, feel free to use the "Dear DP..." mail interface.