If you think one color “can’t win” (or “can’t lose”), you need a different strategy! The sides are “nearly equal”.
Like virtually all multi-player games, this one is a psychological game as well as a competition on the board. Perceptions count for a lot. If you are in the lead but not perceived to be in the lead, you have an advantage. You want to “control” the game without appearing to. Give other players a better reason to attack someone other than you. You don't want to get in a situation where someone says "well, you didn't leave me with a choice".
To do well you need to know what new units are coming, and when; and where your opponents score their points. Think of your forces as a whole, not as separate nations. Maximize your entire score, not the score of each individual nation. One of your nations can divert an enemy, or “take it on the chin for the cause”, if this will sufficiently improve the score of another.
If you deprive someone of the chance of winning, be sure it’s too late in the game for him to retaliate, or that he’s too weak to harm you.
Something you do in a round could have an effect several rounds in the future. Every little thing you do is important in some way! And there’s plenty of time to recover if you have a bad start.
The question of “who is in the lead” is sometimes unclear, and relates to expected scoring at particular times. The Romans score a lot of points early in the game, the question is, did they score as many as expected? Yellow can have the most points and be in fourth place!
Just because you can take something or kill someone doesn’t mean it’s the best move. Weakening one color can help another of your opponents too much. Sometimes it's important to keep an "enemy" around (whether a color or a nation) because it can help you against someone else later on. Force preservation can be as important as scoring points. Just because you can make a 2-1 attack doesn't mean you should do so.
Points are important, but position is just as important, because position strongly influences who will score most in the future. So you might choose, for example, to keep some raiders peacefully at sea in order to be in better position in the next round. Your armies don't NEED to DO anything as long as they're scoring points (and breeding more armies, usually).
Leaders are far more effective in attack than in defense; they are especially good for attacking difficult terrain.
Better odds mean less death for you. Preserve your forces whenever possible. Three to two is poor attacking odds when defenders are in difficult terrain, as is two to one.
Red and blue have more control over the course of the game during mid-game than yellow and green; red and blue can get very high (or low) scores, green and yellow rarely get very high scores. So green and yellow want to avoid someone (usually red or blue) getting way ahead during midgame. Expect the Romans to max or nearly max their R3 points. The Roman difference comes in position, Limes points, and points scored by opponents by R5.
Don't forget, when running a big invasion, to leave yourself in a defensible position. When it's a Major Invasion, be sure to attack with every army (if you attack at all) in the first half: don't waste them "holding territory" that you'll be able to occupy in the second half.
Advice about multi-player conflict games in general: Never make a threat you’re unwilling to carry through; always honor your deals (never break a deal). If you're inexperienced, don't make any long-term deals. Simple deals help you and your “ally”, and usually harm the other two players; e.g., red and blue often agree on a demarcation between the Saxons and the Angles. Simple deals (= "common sense") often work best.
Copyright Lew Pulsipher 2006
Fantasy Flight Games