Game theory tries to solve the game. John Boyd cared about the conditions under which any game gets played. That difference is the fault line Moose and I walk in this episode.
This one is short, about fourteen minutes, and it is just the two hosts. Mark “Moose” McGrath is a Marine officer and my co-founder at AGLX. He and I sat down after I read new work from Donald Hoffman and Andrew Gallimore of the Trace Institute. As I read it, their model of perception, decision and action overlaps with Boyd’s work and runs on the mathematics of game theory.
That sent me back into Boyd’s archives and A Discourse on Winning and Losing to look for game theory. I found almost nothing. Boyd plainly understood game theory, and his work on the F-15 and F-16 trade-offs shows it. So why did he leave it out? Moose had the answer.
Moose’s answer
Moose answered with orientation. Game theory assumes the board can be represented, with fixed options and stable payoffs. Boyd assumed every representation is incomplete, temporary and open to manipulation. In Moose’s framing, you cannot optimize a matrix while your opponent dissolves it during play. He also warns that the most dangerous adversary is the brilliant one whose model has stopped matching reality, even when that model came from Princeton or a Nobel laureate.
Orientation warfare
Moose and I land on orientation warfare. Each side is the other side’s environment, and observing the game already changes the game. Orientation sits at the center of the real OODA loop, so the loop never offered a fixed menu of options to pick from. The non-linear OODA loop holds up where a fixed decision matrix breaks, because orientation keeps changing as the game changes. We also talk through destruction and creation, implicit guidance and control, Hoffman’s work, and what Boyd would attack first in a fixed decision matrix.
Which of your plans assumes the board will hold still?
Listen and watch:
YouTube:
Apple Podcasts:
Spotify:
No Way Out on YouTube:
https://youtube.com/@nowayoutpodcast
Written by Ponch Rivera with AI assistance in production. The full conversation is in the player below.


