Quote from Samsara:
I get what you're saying and why you're saying it, but I think there's a bit of nuance in the idea that there are no right levels. Certainly there's a right and wrong way to approach his method.
Fisher's trading began from the premise that the intraday opening high-low of an instrument determines its close at a statistically significant level of confidence. Care must be put into choosing the right "domicile market".
The theory implies that you're casting a relevant net on the behavior of those traders who move the instrument, and that the net should land around when big players take their positions (relevant to your preferred holding period). You mentioned gaps and above-average volume as possible boundaries. To me, for the indices, I'm using the time between two big cyclical reports. But I'm mostly using it to support my other method, by providing a lens to intraday behavior so I can fine tune when I jump in.
The levels can vary, and they're not "objective" in the sense that they always work for all people at all times. But it also doesn't mean that the method is completely relativistic.
It wouldn't make sense to choose the last 47 minutes of the day as an OR in determining what the next morning's open, for example. For other commodities it makes no sense to me to arbitrarily choose, say, the first five days of every month. Why would it? Do funds or big traders change their models like clockwork based on day 1 of every month? Intraday (to me) seems to accommodate more flexible ORs; longer term seems like it requires some data analysis.
That's why I think it's fair for people to wonder what the right levels are. "Right" implying logical -- consistent with the theory and not chosen out of a hat, while still accommodating the right volatility and holding period for the individual trader.