exactly.. my somewhat lame examples were mainly to showcase my point that the first O in OODA is often the least trained of the four "steps" and perhaps the most powerful!
Doggieman - I think I see a disconnect here.
Remember that OODA is an ongoing process, not just one loop for the event. In the description above;
the
Observation was the loitering person panhandling.
Orientation was "this guy is asking for something to get closer to attack".
The
Decision made is to move laterally and
The
Act is to implement the decision to move.
=== Reset OODA Loop ===
Now we repeat the loop as his response to your action develops. Between the time you act and observe his reaction, your brain processes
1 or more hypothectical responses (he'll advance, he'll play meek and apologetic) and layout temporary solutions. When you observe the reaction, if it matches any that your brain has already "mapped out" the decision is to respond with the appropriate action quickly. Thus, the secondary OODA is observe his reaction, orient it in the context of the situation, decide on the "best fit" of your options and act on the decision. The the loop is reset yet again.
Assuming that you saw the loiterer and when he approached, you stepped to the side, this gives approximately 1/2 to 3/4 of a second for your brain to "think ahead" of the loiterer's next action. For someone alert and with some training, this is enough time to put together some reaction pairs (
if he does "X", I'll do "Y") or multiple pairs from which to select.
Because your prediction of his next move is filtered through
your orientation filters most people will judge what his next likely move will be, based on their knowledge, culture, experiences, etc. This is okay as long as this orientation is similar to his. But if your orientation is wrong (e.g. assuming criminals won't get violent in an area with lots of people) then he does something unexpected, thus getting "inside" your OODA loop and possibly you lose.
I should probably not use the term "reset" as the OODA loop is supposed to be running continuously. But I think it's important to note that every action on either party's part does require you to re-run the loop and re-evaluate the circumstances.
Also - this happens very fast and training is an enormous help here. With training, the brain can recognize clues and automatically match them up with a response because "I know this clue and I know what to do". Little thought needs to occur to match a response to an action.