OODA Loop

The Four Stages

John Boyd’s OODA loop describes how to make decisions in rapidly changing environments:

1. Observe

Gather current information from the environment. What’s happening? What’s changed?

For research: What does the data show? What’s the current state of the problem? What feedback am I getting?

2. Orient

Analyze and synthesize observations in context. Filter information through your mental models, experience, and goals.

Critical insight: This is where bias enters. Your orientation shapes what you notice and how you interpret it.

For research: How does this fit with what I know? What mental models apply? What am I assuming?

3. Decide

Choose a course of action based on your oriented understanding.

For research: What’s the next experiment? Which parameter to adjust? Which approach to try?

4. Act

Execute the decision. Importantly, acting changes the environment, which leads to new observations.

For research: Run the experiment. Implement the change. Test the hypothesis.

Then loop back to Observe—because your action has created a new situation.

Why This Matters

Speed of Iteration

The advantage goes to whoever can cycle through OODA faster. Faster learning beats perfect planning.

Adaptation Over Prediction

You don’t need to predict the future perfectly. You need to respond quickly to what emerges.

Orientation is Key

Boyd emphasized “Orient” as the critical phase. Your mental models, cultural traditions, previous experiences shape everything downstream. Bad orientation → bad decisions even with good data.

Disrupting Others’ Loops

In competitive contexts (Boyd’s original military application), you want to cycle faster than opponents, forcing them to act on outdated orientations.

Application to Research

Iterative Modeling

  • Observe: Model fit, residuals, predictions
  • Orient: Theory, domain knowledge, similar models
  • Decide: What to change (parameters, structure, data)
  • Act: Refit the model
  • Loop back with new observations

Experiment Design

  • Observe: Pilot results, literature, preliminary data
  • Orient: Theory, constraints, feasibility
  • Decide: What to test, how to operationalize
  • Act: Run the study
  • Loop back with results

Paper Writing

  • Observe: Draft, feedback, reviewer comments
  • Orient: Goals, audience, argument structure
  • Decide: What to revise, what to cut, what to expand
  • Act: Rewrite
  • Loop back with fresh eyes

Limitations

  • Assumes environments where speed matters (not all research questions require rapid iteration)
  • “Orient” can be a black box—doesn’t explain how to improve your mental models
  • Doesn’t address when to stop iterating and commit
  • Can lead to reactive rather than strategic thinking if overused

Tensions

  • Speed vs. Depth: Fast iteration can mean shallow understanding. When to slow down?
  • Adaptation vs. Principle: Constantly adapting can lose sight of core questions. When to hold course?
  • Action Bias: OODA privileges action. Sometimes the right move is to observe longer.

Connection to My Work

The OODA loop shapes:

  • Model development: iterate quickly, don’t wait for perfect theory
  • Debugging: observe error, orient to likely causes, decide on test, act, repeat
  • Learning new methods: try, observe what breaks, orient to why, adjust
  • Research questions: observe anomalies, orient to theory, decide what to test, act

It complements PĂłlya: PĂłlya for well-defined problems, OODA for uncertain, dynamic situations where the problem itself is shifting.

Key Sources

  • Boyd, J. (1996). “The Essence of Winning and Losing” (briefing slides)
  • Osinga, F. (2007). Science, Strategy and War: The Strategic Theory of John Boyd
  • Richards, C. (2004). Certain to Win: The Strategy of John Boyd, Applied to Business