Dan McLaughlin, writing for the conservative Federalist, examines the 2016 Republican primary race in terms of military strategist John Boyd's philosophies:
Boyd’s core insight was about the interactive and disruptive nature of speed on human decision-making: success in conflict can be rapid and dramatic if one can “operate inside the OODA Loop” of the opponent. Operating inside the opponent’s OODA Loop means presenting him with a constantly shifting battlefield that keeps him off-balance and disoriented so he is unable to process information and make and implement sound decisions before the situation changes again.
So, what does this all have to do with Donald Trump? Quite a lot. Few candidates in recent political memory have been so effective at altering the reality around them in a way that crashes their opponents’ OODA Loops.
As a major-party nominee, moreover, Trump would lack the ambiguity he has deployed against Republicans, and in a two- or even three-candidate race, he could not exploit the collective action problems and Hobbesian scramble for free media that have enabled his rise. Indeed, few of the factors that have allowed Trump to trigger fear in his Republican opponents would even apply in a general election, and Clinton’s team would have plenty of time to prepare a counter to the things he has been doing so far.
That’s not to say that Trump’s celebrity and attention-grabbing power would present no opportunity to win (he would only be the nominee if he’d already figured out how to solve the low-turnout proclivities of his natural base), but ultimately, he could not deploy the same approach without major adaptations. Trump would have to prove himself flexible and open-minded enough to the dynamic general election system to attract the necessary 70 million voters. His ability to do so remains very much unproven.
I don't always read the Federalist, but this analysis made a lot of sense to me. It's a long read—and worth it.