I would like to pose this question. How does someone review speakers? The new pair of speakers is always setup in a slightly different location to the original pair. If we were to hold to the "one variable at a time" philosophy then the new speakers should be put in EXACTLY the same place as the pair that just left. We all know this doesn't work and fully accept that the reviewer had to reposition the new speakers to get good sound in his/her room. Yet, when Roy mentioned that he repositions the speakers when he gets a new amp to achieve best sound everyone (well not quite everyone) freaked out.
I don't think there is a one size fits all protocol and the approach taken will depend on the product under review. And while I believe in keeping variables to a minimum, as
@ddk suggests, this isn't a controlled scientific experiment, it's an audio review.
Consider amps and speakers - they have a co-dependent relationship, with some pairings working really well together and others not so well and some combinations in-between. For each pairing there is likely an optimal setup within a given room. That setup may have a relatively broad or narrow range of position options.
What do you do when you bring an amp and pair of speakers into a room? You hopefully work out their best setup. Let's say you have a set of speakers to review or an amplifier to review. I would bring the review product into my audio room and proceed to setup the pair for the best positioning I could find. That's what an end-user would normally do. Do the same for the reference where one product stays the same (amp or speakers.) I don't think that is confusing or somehow deceptive to the reader if each pair has its own position. Just briefly explain what you've done and describe what you hear.
If you are reviewing turntables then, imo, there is a co-dependency between arm and cartridge, but not between cartridge and turntable. As long as an arm fits correctly on a table (P2S distance, etc.) and all its adjustments are allowed then there is less co-dependency between arm and table than between arm and cartridge. Maybe some odd resonance issue might come into play with some arm and table combo - but that can be described in the review if it happens. Otherwise, there should be only one variable and that is the turntable itself. If you're reviewing a turntable that comes with an arm, ask the manufacturer to borrow an extra arm to mount on your reference.
Phono stages? Pick one cartridge that is simpatico with the review product and the reference. Pre-amps? Each pre-amp is the sole variable. Don't mess with trying different cables. Only if one is XLR only and the other is RCA only might you need different cables. I keep an instance of each with otherwise the same cable just for that situation.
Reviews are about one person's experience with the review product. Ideally they include a comparison to a reference product. Ideally they keep variables to a minimum. There is no universal protocol. The reviewer just needs to describe well and succinctly what he did. If the reviewer is reasonable with the process and the 'mechanical' portion of his description, then the most important part (imo) is his sonic description. A comparison is important but secondary.