I agree that "total body water" is the number one impact to daily weight. My criticism of weighing as a means of determining hydration is the inaccuracy of the measurement itself in addition to other variables (primarily "gut solids") that can't be accurately accounted for. This makes a weight measurement too indirect to be useful IMO.
I also agree with you that chronic dehydration can adjust the sense of thirst, but the proposed weight method does not help with this as a chronically dehydrated person would just have a lower base weight. All observations are being judged on the delta from current normal for that person, so is no better than thirst that is based on current normal hydration levels for that person.
I hadn't heard those skin scratch and tugor tests before. Interesting. Although I reckon I'm currently pretty dehydrated and I passed both! ;)

How about I restate my opinion as follows: Body weight, especially when not taken in the same circumstances each time, has too much noise in the signal to give reliable guidance, except when dehydration is significant enough to be determined just as well, if not better, by simpler methods.

Now THAT is a set of caveats I can get behind. :-)

How so? Do you mean the task of weighing often is unreasonable? Inaccurate? Something else?

My point is simply that having more "data points" gives me a clearer picture of how much my weight typically ranges and where I'm trending. Given that weighing often is a quick and easy task for me, I don't see what else would be unreasonable about it. A little ocd, maybe, but that's not the same thing as being weight obsessed. The numbers don't bother me. If I'm trending up, I know I need to buckle down on my nutrition a bit. That's about all there is to it. I don't track calorie or fluid intake.

To the point about hydration ... as an example, say I'm in the range of 163-168 for a few days, maintaining relatively normal training and nutrition, then wake up one morning at 160. Well then I suspect I might not have hydrated enough the previous day. It's pretty simple, in my opinion. Too simple?

By all means, if there's a legit physiological or medical concern that I'm overlooking please enlighten me. Hope I'm not coming across as argumentative here. I see it as interesting discussion. I'm relatively new to this, and I'm still learning how my body reacts to nutrition, training and recovery.

I think I already covered it really, and I'm not sure how I can clarify further.
If you are mostly concerned with whether you're gaining or losing weight over time, as opposed to hydration, sure weigh yourself regularly. But once a day is regular enough. More is unlikely to help much with accuracy. A random spray of data points is not what you want. Far better to take a single measurement each day before breakfast. This will generally give you the most consistent and comparable value. It doesn't matter what your max and min is each day. What matters is a consistency to avoid unnecessary noise and confusion in the data. From what you've described it seems you're trying to measure everything in the hope you'll then be able to spot outliers. Not a wise approach in my view. The only way that would provide some extra info is if you're also recording detailed circumstantial data each time you take a weight and then doing some really detailed analysis afterwards. Seems utterly pointless even if you were doing that but you'd learn a little about what influences your daily variation perhaps - but why bother?
It's not good practice in experimentation to just measure everything as is and figure out what to do with the data later. The priority is to control or understand as many variables as possible, to remove error and noise so that whatever unexplained variation remains corresponds to the phenomenon of interest. If you want to know how your body weight is trending over the long term, (i.e. days and weeks), there is no real value in measuring how much you gain when you eat or drink, and lose when you use the WC, or sweat, or breath. It may satisfy idle curiosity, or an obsession, but it doesn't help answer the original question in my view.

On average, there is a 24% chance that more than 0.04 inches of total precipitation will fall in Tampa throughout the day on March 26, all of which can be expected to be rain.

Should I use this data or the current weather radar to decide if I need an umbrella tomorrow?
------------

Edit: Scratch that point, probably not the best analogy. You are right that over time, the average is going to show the trend. I guess it's just interesting to see, in real time, "hey, I lost 2 pounds on that run. Need to make sure I hydrate." Or, "fck I'm up to 170. No beers for me until the weekend." Or whatever. Maybe it takes the place of calorie tracking. I did that for a couple months early on. It works, but much more of a pain than hopping on a scale.

That analogy has nothing to do with what I'm saying. I'm not talking about probability versus what occurs.
I'm saying real time weigh-ins don't tell you if you've lost far or muscle, they tell you about the contents of your stomach and bowels and how hydrated you are, kinda. Fat and muscle don't change perceptibly on an hour to hour basis, and barely on a day to day basis, given the accuracy of domestic balances. Thus jumping on a balance multiple times a day is is not useful. It's either just muddying your data, if you're using it for long term trends, or has you chasing short term illusory targets and compensations, if as it sounds, you are reading every loss or gain as something that might warrant a response.