IMO:
Like "seeing the curve", sanding is very much an experiential and existential thing, and in a lot of ways it cannot be taught but can only be learned by doing. I've found, and others have alluded to the fact that there are no hard rules, only guidelines. It's great to share these, and some of the tips here will make more sense to people who have more experience and a better context to think "ah, good tip, I'll try that when it is needed", or "ah-ha! that's why I get in a rage when X happens and I just can't fix it." Then they go try it and it works like a charm, or use that input somehow with whatever individual instance of whatever species of wood or other material, with whatever abrasive, using whatever machine (or not), running at whatever speed, moving in whatever vector in relation to the work and it's sub-parts (grain orientation, hardness variability fromone band to the next, band width, etc.), using whatever lubricant, in whatever light, with whatever dust control, with whatever PPE that impedes their vision, using whatever pressure they decide at that moment.
The inexperienced can be told all those whatevers, and the 50 others I didn't mention, before they start. They might get overwhelmed, but they might follow all of it or some of it, and learn "whatever" they learn from that one experience with all those whatevers. They might make the mistake of forming a mental rule they have to discard later, or it might apply almost all the time. They will surely gain some level of enlightenment if they are paying any attention at all, but they will not become experts even if the results are miraculously perfect (they got very lucky, most likely).
The experienced woodworker will already know many of those whatevers by having experienced them ins some context at least once before, even if they have never been in that exact context -- like with walnut or poplar, for example, nor ever used a 1" RO sander, but have three 5" sanders in the shop, etc. The experienced woodworker will sometimes still be surprised and baffled that after fifty-X years of sanding thousands of pieces of hundreds of different raw materials with hundreds of abrasives and lubricants... surprised that they just now found a piece of maple that only tears out in one quadrant while the visible grain is consistent with the first quadrant but still doesn't even agree with what its own sanding behavior is!
The only thing the experienced person has that the novice does not is, well, experience. This instructs them where to start or what is the thing I should change that will probably make the difference I want, and that's all. With a harder wood, for example, they will know that they should start with Whatevers Combination 347, where that infuriating soft fuzzy poplar should be Combo Zero: sand it with fire. The experienced woodworker knows what I mean there instantly, and a beginner can hear it, heed it, and still not truly know what it means until they sand a piece of poplar that just laughs at the idea of becoming smooth in a completely different way than a freak-of-nature piece of hard maple with a Rockwell hardness in the diamond range ("it shouldn't be this damn hard, even for hard maple!").
An example I will come back to: A worn piece of 100 is about as aggressive as a new piece of 150, but there are times I would choose the worn 100 over the fresh 150, and I cannot put it into exact words or give you a hard rule, nor suggest that the same rule applies to someone else's experience and current context. Sometimes I would just try both and quickly figure out which one works better for what I am trying to do, knowing that all those other "whatevers" are still in play, including the edge cases where a certain material just isn't behaving like almost every other instance of that material has in the past. Perhaps a more experienced person also has an ability to pick up the effects of making the switch between the tired 100 and the perky 150 much faster than a novice, or be able to see some other new thing emerge that would override the otherwise better choice (like conforming to the shape, which I will harp on quite a bit, since it is a fresh-is-best dogma-killer).
Having tips from others is great, don't get me wrong. But there is no substitute for experience and no true shortcuts to gaining it. Advice from others is just a shortcut to the proper starting point, but gaining the actual experience is still something that just needs the effort. "Transferrable skills/knowledge", or Context, builds on existing context exponentially until one "knows everything". After that, the only variables left are, eh, well, still all of them. Muscle memory and "feel" are a big thing here. These are experience too, but not of the cerebral variety. Pressure is a perfect example. What is "gentle pressure"? The only thing oen can know for sure is that it is less than "firm" pressure. The rest must be learned through experience, and yes, always in the midst of all the other whatevers.
The experienced can react faster and get to the things they should try right here and right now when the usual stuff isn't working or they are trying a new thing they just heard. The inexperience might never get to one of the "correct" combinations at all, but the ones who persevere will come back and ask, keep trying, and so forth. It is not terribly difficult, you just have to put in the hours.
As for some of the things I have learned, none of which are earth-shattering but might help someone just starting out:
I don't disfavor worn paper as much as a lot of others here, but I certainly respect and agree with their opinion as well, and fresh paper cuts faster than worn paper without a doubt. As I said, I have found that sometimes a worn piece of 100 is "better" than a fresh piece of 150 in certain circumstance, and this applies across the grit-size spectrum. Is it slower? Usually. Except for the times when that new 150 has a piece of 120-ish sand in it and makes a nice scratch the very first time it is used until it wears down that first little bit. The worn 120 you just left has given you a 135-ish finish. The new piece of 150 (grits are average sizes) and the 120 grain in it actually takes you backwards. I typically hate those first few swipes with ANY abrasive, aside from the very first grit. They are often a step backward, a little too aggressive. Does worn paper generally conform to a profile or curve better? Absolutely! Add that to the list of whatevers - the paper and its current condition matter a great deal too. That said, there is a time in every abrasive's life when it's just over.
When hand sanding, folding the paper into thirds keeps it folded up neatly and behaving far better than a piece folded in half and if one is meticulous it presents a more durable sharp edge at the fold. The half-folded piece lacks friction between the two sections, but the one folded into thirds stays put because the abrasive keeps the interior sheet from slipping around. It also makes the flat surface less flexible, which is useful sometimes. If that's not desired, just skip the fold altogether and use a single sheet.
You can skip grits. Really, you can. You can skip five grades if you want to. But it doesn't save you all that much labor and generally costs more time and abrasive in the long run if you skip more than one grade and that depends on which one it is. The "Dont skip grits" advice is a dead horse, but there are also costs for all that task-switching too, and that fresh piece of the next grit may surprise you by leaving a worse surface than the previous grade's worn paper did. I already mentioned that above; this is a slightly different context. With that said, I haven't seen a whole lot of benefit from doing a round of 120 between 100 and 150. Same goes for using 180 between 150 and 220. Experience is once again the best teacher here. If I only had one (of 100, 120, and 150, that is), I would use 120. This personal axiom is fluid throughout the range; that is to say that if I could only choose 150, 180, or 220, I would pick 180. Worn 180, to me, is the same as 220. Taking this all the way full-circle, my experience tells me that *i* can skip a grade, pretty much anywhere along the spectrum. That soft rule does break down a bit with the lower, larger grades, however. There is always going to be a significant difference between 32/36 and 80. 60-grit is not as skippable between 32/36 and 80-grit the same way that 180 is a viable skip on the way from 150 to 220. One of the firmest rules to me is that if you have a scratch at a certain grade, you simply are not done with that grade, but that's academic.
You can stop sanding and increasing the grit when it's done. I can't tell you when that is. I can say that I have had very good results stopping at 400, 600, or 1000 wet-or-dry paper lubricated with oil, depending on the polish I am after.
Clogged paper is the worst. Clean it often if it can be cleaned at all or chuck it and figure out how to change things up to get less clogging if possible. No-brainer, but sometimes it is hard to toss a piece of paper you've only used for five minutes. Always value your time, that makes it easier. I have a hard time with this one. I wish I knew a trick for cleaning cloddeg paper. The rubber things help, but aren't worth the effort for inexpensive disposable papers.
Hard blocks work much better on end-grain than soft ones do, but they also wear out the paper super fast. A loose rule of thumb I have come to is that you should always use a backing that is harder than the hardest part of material (grain is almost always variable within the piece) that you are sanding. This is difficult to achieve when sanding curves, especially 3D ones where you need a soft backing just to conform to the work. In that case, a gentler pressure will do better than firm pressure. Gentle pressure will push more evenly on uneven hardnesses within a material and tend to want to dig out the softer stuff less dramatically while still eating at the hard parts. Grit size selection helps here too. Think of the grit appropriately as tiny shovels or saw teeth; a small shovel only digs a small hole. Once again, the combination of worn paper (for the conformity, AND the shallower shovels) on a hard block might work better than fresh, stiff, non-conforming paper on a soft block. The soft block will not prevent fresh paper from creating a ton of little folds as it wraps around a curve, and those will scratch the hell out of the piece and unevenly.
There is no shame in hand-sanding, it just hurts. Sometimes it is all you have available. The more experienced woodworkers might think all the way ahead when conceiving a piece and realize that hand sanding, a LOT of it, is the only way to be able to finish it, and decide that the fire needs a bit of fuel instead. On the other hand, finishing off with hand sanding is almost never a mistake.
Lubrication is your friend, heat is the enemy. Someone else mentioned that, I echo it here. If the paper is warming up, it is either time to change grits or ease up on the pressure and/or speed. Generally speaking, like all the rest.
Abrasives are a cutting tool just like a saw or drill bit. "Let the tool do the work." Usually, gentle pressure is "always" better, but there is a sweet spot and it can vary with all of the whatevers.
Lots of woods are toxic and/or carcinogenic. Some can cause allergies with sustained exposure. I'm looking at you, WALNUT. If it smells rank or it makes you a bit nauseated after a while, you should probably stop and do some research if you don't already know it is harmful. If you do already know and you got that far, you're just bringing it on yourself. I find that it is easier to just stop breathing while I work. When that is not possible, I just avoid that kind of wood and even fine walnut is not worth the cost of working with it, but that's just my opinion. Sometimes it's pretty, but it is always just so RANK. All wood is toxic/sickening to some degree, though, and dust particulate in the lungs is bad even when it is chemically inert, so the rules of respiratory PPE are always in effect anyway. It is sad to have to use dust control for health reasons when working with cherry, though.
Patience is a virtue. Two or three extra minutes at each grit, not just the final one, will produce remarkably better results and almost paradoxically it will not take notably longer. It takes a lot longer to sand out scratches than to just get through a grade. I don't like sanding any more than anyone else seems to, but I really don't like being disappointed in my work. That, I cannot stand, and I am a tough critic that never excludes himself. When you think you are done with a stage, take the time to give it a bit more love before moving on. If you are the type that enjoys the overall woodworking process, learn how to love, or at least not despise sanding. It's hard to love, but hating it ruins the whole vibe, and considering it a necessary evil still gives it that ruinous power. This easier said than done and involves some self-trickery. I find that these kinds of mundane tasks pair with music rather well; it is a distraction from the drudgery and improves my mood, but doesn't distract so much as to cause a hazard. I second the notion that sanding is a make-or-break process just like everything else leading up to it and it must be done so try not to waste energy actively hating it. And, it is the penultimate riskiest process, since it builds on nearly all of previous effort.
The very best way to sand is to not have to sand at all. In a perfect scenario, which never happens but still informs reality, you would only sand your cutting edges, not your materials. Sanding will not [efficiently] remove chatter marks, tear-out or similar injuries to the wood, those are most efficiently tooled away. Unless you just like sanding for hours and actually re-shaping the piece with sandpaper. If so, you should not want any of this advice anyway, toss your chisels and sand-shape the finish right from the start. And have fun with that!