I don't know where you're getting your figures, but based on nearly 40 years of working with all of these fibers, I don't believe some of them are accurate.
Ballistic nylon, and to a lesser degree heavy basket weave polyesters are far stronger than canvas, and significantly more puncture resistant. Typical 1050d Nylon requires 175lb of pressure applied to a point (like a screwdriver head) to puncture. Canvas requires something like 10lb of pressure.
You aren't going to puncture typical canoe canvas with a screwdriver and ten pounds of pressure - not even close, and when filled with traditional filler, it is even tougher. I once removed old canvas from a canoe and decided to see what it took to puncture and tear it. So, I took a big gutter nail about 12" long and tried it. It pretty well put any fear of ripping the canvas on a canoe to bed in normal use unless I hit something extremely sharp - especially considering that the canvas on a canoe is not suspended out in space, but is backed up everywhere by wood. The focus of such an impact seems more likely damage to the planking than the canvas.
If you do a tongue tear test of similar weights of nylon, polyester (Dacron is polyester, by the way) and cotton canvas, the nylon will win, the cotton will be next and the polyester will come in last. This is due to the tearing properties of the three and their ability (or not) to stretch and for the fibers to give a bit and help reinforce their neighbors. Where the nylon and cotton will give a bit and spread the strain over multiple yarns, the polyester yarns tend to take the stress one by one and break. This is called explosive tearing and is why I can take a hunk of heavy polyester and easily tear it in half. This does not happen with nylon or cotton.
Unlike skin-on-frame constructions, I think you also have to take the non-suspended nature into account which will limit the type of impact the cloth will typically be subjected to. Should you slam into a screwdriver with a wood canvas canoe, you will likely have more serious problems to worry about than the canvas.
Nylon and polyester also have significantly longer lifespans if appropriately UV protected, and I figured that would reduce the necessity of regular recovering. What I've found so far seems to indicate that with canvas you should be prepared to recover every 10-15 years, but that it may last significantly longer.
This would depend on what conditions the various fibers (and canoes) are subjected to. Cotton canvas does not self-destruct after 10-15 years all by itself, and we have certainly seen plenty of canoes where it has lasted far longer than that. Yes, it is the only one of the three where mold and mildew can actually eat the fibers, rather than just sitting there on top of them, but with mildewcide treated canvas these days, and the ability the boats seem to have to dry out between uses, this is less of a factor. You also should probably consider that the modern wood/canvas canoe usually leads a fairly pampered life. After going through all the work to restore one, or spending the cash to buy a new one, few people are going to beat it down a rocky stream, or store it out in the back yard on saw horses next to the Grumman.
In terms of durability of the raw fabrics, for UV the polyester is best, cotton will be next and nylon will be the worst. For abrasion, the cotton is best, and it's a toss-up between the synthetics, depending on the weave. Cotton's somewhat fuzzy nature, because it is made from short fibers spun together, builds in a little bit of cushion which helps protect it. The synthetics with their continuous extruded fibers don't have this. Rock climbers found this out 40 years ago when they found that their new-fangled nylon assault packs were getting quickly abraded to death being dragged up the rocks. They quickly switched back to their old canvas packs. The good news about this was that it helped spawn the development of cordura nylon, which resists abrasion much better because it has a fuzzy surface. I suppose you could take samples of all three and test them with a belt sander to see which is toughest if you wear through your filler, but I haven't tried it. My suspicion is that the cotton will probably win.
With solvent-based (like traditional) fillers, you will probably get a much better filler-to-canvas bond with cotton due to its drastically higher rate of absorption. In order to get a truly smooth surface with no weave texture for a classy paint job on the synthetics you would probably need an epoxy-based filler. Polyesters usually aren't very absorbent (like just a few percent of their weight) and nylon isn't a lot better. Epoxy is likely to stick much better to these than a traditional filler will.
I suspect the best synthetic system might be something like heavy Oceanus polyester canvas sailcloth and an epoxy mixture to fill it, though it would rely on being able to stretch the Oceanus enough to get it tight (which may or may not happen) and using the right filler mixture. Most of the easily sanded epoxy mixes are too soft, and the ones that are similar in hardness to canoe filler are a real bear to sand smooth.
On the other hand, I have no problem with sticking with real canvas and traditional filler because we know it works and because it's what was there originally.