|SMS Scharnhorst: "Battle of the Falkland Islands," Panu Hakkarainen|
These general propositions vary according to technological and geographic considerations. On land, combatants can often force combat on an unwilling foe through surprise, because of difficulties associated with mobility, and because of the critical nature of some landmarks (although the determination of "critical" depends on operational, strategic, and political considerations). Concerns about the relative strength of offense and defense also shape expectations. Moreover, asymmetric expectations exist within organizations and alliances, as leaders often have incentive to deceive followers on the probabilities of victory and defeat.
Historically, naval forces have been more capable of avoiding unwanted conflict. Anti-access technologies give the defender a major advantage that both sides acknowledge, meaning that fleets can avoid conflict by remaining behind their "walls." Major engagements happen through surprise (forcing the enemy to fight against his will), disagreement about the basic conditions of battle (possessing greater knowledge of the capabilities and numbers of deployed forces, as well as atmospheric factors), or both. Nelson's victories tended to involve elements of both; the audacity to force combat in unexpected situations and in unexpected ways, combined with great confidence in his own capabilities. Togo won Tsushima through complete confidence in the qualitative superiority of his own ships to those of the Russians. Jutland happened because both sides misjudged the size and position of the enemy. When faced with superior enemy numbers, the German, Austrian, and Russian navies simply refused to engage their foes.
Even following the development of strike-capable naval aviation, anti-access technologies (land-based air) could provide sanctuary for a weaker force. In the Mediterranean, the Regia Marina and the Royal Navy avoided unfavorable combat situations so effectively that the combat power of the former was in large part intact when it surrendered to the latter in 1943. After a series of relatively symmetric battles (most of which pitted relatively equal forces that both had a chance for victory) in 1942, the USN and the IJN avoided each other for most of 1943 until the advance of the former forced the latter to engage under risky, adverse conditions. Of course, by the end of the war (in both theaters) Allied strike superiority was so great that the remaining major Axis units were destroyed at anchor; you can only run for so long.
Why discuss this today? We have an interesting electoral situation in that both Republican and Democratic partisans seem virtually certain of victory tonight. It will surprise no one to find that I'm with the Democrats on this one, both as a partisan and on in assessment of the electoral situation. In a similar situation eight years ago, I was optimistic about the chances of John Kerry, who on objective measures certainly had less of a chance at winning the popular vote than does Mitt Romney. The electoral vote is a different story; I think that Mitt has a very tough hill to climb. In any case, an election isn't quite like a battle, because of course it happens at a specified time whether the antagonists want it or not. Nevertheless, asymmetric belief plays an important role in how elections play out, because such beliefs affect the behavior of party actors (how many resources to devote to a particular race or particular state), of activists (whether to commit themselves to various difficult and unpleasant tasks in support of a candidate), and of voters (whether to drag themselves to the polls). Just as in battle, leaders have incentive to deceive followers as to true prospects, whether in order to maximize chance of victory or to prepare the post-electoral space. Both leaders and followers are subject to motivated bias, the belief that the necessary is possible (or even likely). If there's a difference between now and eight years ago, it's that it seems that the institutional apparatus of the right (the major media organs in particular, but also the GOP itself) have gone more "all in" on an a favorable interpretation of the available data, but it's hard for me to say how valid that impression is.
In any case, make sure to vote and have a safe election night!