"War is also interrupted (or moderated), and thus made even more a gamble, by: the superiority of defense over offense; imperfect knowledge of the situation; and the element of chance."
- Clausewitz, "On War"
War theorists can pinpoint when the nature of war changed. It was the American Civil War. Prior to this conflict, war basically revolved around getting as many troops to the scene of battle as possible and then the best general won. This one battle almost always decided the conflict, if the defeat was complete enough. The Battle of Waterloo was a great case in point. War could very much be seen as a skilled conflict between commanders in the field and their armies, with chance the usual contributing factor. However, only a short time later, in the American Civil War, the factors and parameters of war had been utterly recast. War became, and still is, a battle between economies. The advent of improved travel and technology meant that armies could be replaced again and again. Battlefield expertise, whilst still important, became secondary to how well a country could produce and distribute soldiers and weapons. This is why the contribution of the U.S. and Russia was so critical in World War II. The Americans outproduced Germany and the Russians absorbed so much of Germany's resources on the Eastern Front.
So, what about poker?
I think the analogy is clear and important. For most of the time, and it is still true below the very highest levels, the skill of the
combatants is by far the most telling factor; think of Napoleon versus Wellington. Having and maintaining a clear edge in terms of skill is critical. But at the top of the poker pyramid, how much skill difference will there really be? Occasionally, there will be a "provider," but the differences in playing skill become tiny. Key factors become meta-game factors, such as emotional control - and bankroll, the poker equivalent of bullets and soldiers. All things being equal, the player with the biggest bankroll has by far the best chance of prospering. Everyone gets unlucky and runs bad. But if you can keep on putting bullets in your gun without starting to sweat about where the next bullet is going to come from, you will have a big edge over your shorter-bankrolled opponents who will need to run fairly well before they run fairly bad, as the alternative could cripple them before they even get started.
But the sophisticated high-level pro also uses his bankroll as a direct weapon. The two most clear examples of this occur in no-limit hold'em and pot-limit Omaha. It used to be the case in no-limit hold'em cash games that massive overbets simply did not happen much; the value-bet was the norm. After the arrival of Internet players like Spirit Rock, everything changed. The common ploy of these types of players was to make hugely out-of-proportion bets, normally on the river or the turn. This asked the poker question: "I have you beat, or I have nothing; want to gamble?" The answer then became one based on not just hand values, but your propensity to gamble with what might be a much higher proportion of your bankroll than it was for Spirit Rock. You can create a similar effect in pot-limit hold'em, especially shorthanded. By in effect raising nearly every hand you play a small amount preflop, you create a game essentially three times the size. As long as you do not take this to excess, the importance of post-flop play and the parity of starting-hand values means that you are not giving away a lot of equity by doing so, either. But you will have shot variance through the roof and created a real battle of the bankrolls.
"Superiority of numbers is the most common element in victory. It thus follows that as many troops as possible should be brought into the engagement at the decisive point."
- Clausewitz, "On War" 
David has played poker all over the UK for the better part of a decade. Originally a tournament player, now focused on cash play and almost entirely on the Internet for the last three years, David makes a healthy second income playing a wide range of games.
