I had the impression that wars mostly break out when each side thinks they have the upper hand. While there are some instances of suicidal attacks by societies

I had the impression that wars mostly break out when each side thinks they have the upper hand. While there are some instances of suicidal attacks by societies that have been backed in to a corner, I would expect that in most cases, the obvious loser of a potential conflict will try to find some less costly way out. I think I first encountered this idea in an essay about Lewis Fry Richardson's research, but I can't find the citation. His idea was that by analyzing the theory of conflict, he hoped to make prediction of the eventual winner more reliable, reducing the occurrence of mis-matched estimates leading to war.