I have heard that in Honolulu, the polygraph was something like just all relevant questions. I guess the best bet is to "flat line" as best as possible. I was told that most polygraphs will not use just all relevant but I think with all the different types of counter-measures out there, maybe this is their way of using their own countermeasures.

Since it is not clear what objective standard (if any) state and local agencies are using to score Relevant/Irrelevant "tests," the safest option may be to not augment reactions to any questions during the "in-test" phase.

However, it may still be helpful to augment one's reaction to the number (or card) one has chosen during any "stim test" (so that later reactions to any relevant questions will appear less significant by comparison).

In addition, some R/I formats may include an announcement of the beginning and ending of the "test" as a kind of "control" stimulus. It may be helpful to produce a reaction to any such announcement.

Another possibility is that a question series that appears to consist entirely of relevant and irrelevant questions may, in fact, include "disguised controls," such as are a part of the Department of Defense Polygraph Institute's "General Question Test" (which is allegedly no longer taught at DoDPI, but variations of which may still be in use).

Finally, the behavioral countermeasures discussed in The Lie Behind the Lie Detector should be helpful in getting through a Relevant/Irrelevant polygraph screening "test."

What about Polys these days for California sheriffs and police that tend to use polysgraphs with just revelant questions, and/or revelant/irrevelant questions with no controls?

How could you use counters with these, or you just don't.