The chaos and conniving of Multipolarity has inspired me to reflect upon what I feel are some of the major, and continual points of failure within the Imperium Offtopicum tradition that can often break a game both mechanically, but with a particular eye toward immersion. Here are some of my main complaints, in no special order.

Spheres of influence

Almost nobody seems to know what this word actually means. Rather than the zone in which a player can tangibly exercise his influence (either through hard or soft power), it's typically used as a label for people calling dibs on neutral territory. This frequently has the effect of carving up the map within the first few turns, and later arrivals become ostracized because of these invisible boundaries. What's worse, there's very little a game moderator can do without direct intervention, which custom typically prohibits.

Nebulous combat systems

The only common mechanical strategy in IOT is combat. Even if everything else is abstract and subjective, battles produce hard results. In especially refined games where armies are calculated numerically and military strength is a tangible factor, I will want as much information as I can on how the fight was determined. Irrespective of how much it helps to streamline the update, reports that provide a string of numbers and rely on implicit "trust" in the GM for their explanation deny the player the ability not only to independently interpret the results, but also make any sort of plan based on them, turning combat operations which should retain a degree of strategy into games of almost total random chance. Especially vexing is when battles lead to vast territorial changes with no mechanical explanation, sometimes culminating in the defeat of entire countries (of substantial presence) in the space of a single turn.

Ad-hoc rules

No ruleset is perfect, and GMs habitually amend them on the fly. But when the mechanics are altered radically without prior warning or consultation, someone's begging for trouble. Equally worrisome is elements with major potential repercussions that have either not been defined clearly enough, or worse, are left deliberately ambiguous. Rules can be expanded as the game develops, as they commonly are, but if the tinkering becomes meddlesome and disrupts the play with an eye toward destabilizing the power structure to which players have grown accustomed and to which they tailor their plans, they are going to feel somewhat abused. When the GM decides to violate his own rules for no clear reason, players will be justifiably outraged.

Alliance blocs

Wasn't the Cold War fun? Didn't it feel so encouraging to tow the line of either superpower or be relegated to the backwaters of international attention? There isn't a lot that can be done about this, unfortunately: ill-conceived defensive pacts spring up like weeds, and like a game of blob-tag consume as much as they can, until they collide with each other and go for the jugular. It's divisive, it's tiresome, and it typically dictates the course of the game for the rest of its life. What's worse: usually the only way to combat a bloc is with a bloc of one's own.

Letter over Spirit

I like to call this the "Great War" effect: nation A attacks nation B; nation B invokes defensive pact with nation C; nation C attacks nation A; nation A invokes its own pact with nation D; and before long every country and its second cousin is at war. This chain can be broken at the A-D pact, but only if nation D is willing to recognize nation A as an aggressor to nullify its treaty obligation. This almost never happens: either players are too naïve to realize they have an escape clause, or have a personal interest in going to war that they try to excuse as "upholding honour". Usually this can be tolerated as typical Machiavellian politics; but when the aggressor state being defended has completely forfeited any pretext of justification for its action, whatever "honour" exists in aiding it exists in name only.

OPs that never update

Knowledge is power in any diplomacy game, but rarely will players take the time to record every snippet of information themselves; nor should they be expected to. But the GM should provide a few staple data, namely the current world map, and a list of players and their countries. The OP should also be kept up-to-date so players don't have to hunt through the thread for the latest update.

Additionally, the game rules should always be reposted at the start of the game thread, even if outlined in sign-ups, so players have all the information they need in a single location.

GMs that fail to remain impartial

This is almost entirely Taniciusfox. The only thing worse than disruptive players is a GM who tries to lead the game in a specific direction. As the omnipotent and (usually) omniscient steward, one may be able to envision the many possible futures, but actively trying to provoke one through "accidental" information leaks, suggestive commentary, and abuse of NPCs/automated nations to deliberately spark conflict is dishonest and disrespectful.

Overpowered NPCs

IOT is a multiplayer game. Most of its enjoyment stems from the fact that players are interacting with each other. Non-player countries are a valuable asset, but there is a fine line between complementing and supplementing player actions. Tani has been repeatedly criticized for over-emphasizing the role of NPCs in his games; this was particularly contentious in Imperium Universalis, where they were surpassing the players within the first few turns. While NPCs certainly shouldn't remain pushovers, augmenting them too far risks belittling the players' role in the game.

Order locks

This one is hard to get around. In complex games that require numerous economic and military calculations, order locks are often the only way the GM can process commands in a timely and accurate manner. The problem, however, is that it breaks the organic nature of "real-world" diplomacy: turns are no longer periodic updates, but a set temporal mechanism to which player actions become bound. Wars are guaranteed to erupt only after the lock, frequently as surprise attacks with no possibility of retaliation until the next turn; this creates a gap in response time that favours the attacker and prevents timely international response. In this way, gameplay is less about diplomatic strategy than trying to predict and out-manoeuvre everyone else's moves.

The Chamberlain Effect

I strive for the transcendence of realpolitik as much as the next ideologue, but there are periods where pragmatism is grossly undervalued. I have been party to a number of games in which massive global conflicts would have been avoided had the world only intervened at an earlier date. Now, one can argue about whether it's "right" to break character to attack a player country based on the player and not his roleplay, but Chamberlain's "mistake" was to overestimate his opponent's commitment to reform and continue to offer second chances despite an unmistakably hostile precedent. Naturally, the Chamberlain effect most regularly results due to

People who can't roleplay

By this I'm not saying people who don't roleplay, but players that cannot maintain a distinction between themselves, and the national identity they assume in-game. Sometimes this is done on purpose, but when players try to present a myriad of different characters and régimes as distinct entities that all end up following the same predictable course of action, it ruins immersion, insults the other players, and almost always leads to spammy apologetic posts. Coincidentally, such players always tend toward some combination of arrogant boasts, aggressive play, and complete denial that they have marred their reputation in-game, to the point that everyone else will want them banned outright on grounds of decorum.