Claydol seems to be a non-carbon-based lifeform so it's plausible it could have that kind of longevity whereas Nintetails is more animalistic in terms of its biology thus it would probably have a normal lifespan.

Exactly this. Ninetales is the exception, not the rule - as far as I know, it's one of just three organic Pokemon that mention any kind of lifespan (Absol, at 100 years and Wartortle at 10,000 years being the other). I could change them to accommodate this, but it just looked weird to me (and it isn't the first time the Pokedex has put out weird or just plain reality-defying information - see Magcargo's/Camerupt's entries, Wailord's weight making it less dense than air, etc.).

For Pokemon that are either inorganic and have some sort of longevity mentioned or Pokemon who have to be a certain age (usually elder gods or some such), they get the numbers from the lore.

Maybe you should assess Pokémon that aren't based on real things on a case-by-case basis? For example, Latias' PokéDex entries repeatedly mention that it has down (fine feathers), but it's definitely not a bird, so we can assume it's most similar to theropod dinosaurs.

This ended up being pretty simplistic, but I went with the premise 'it has wings, it has down, it's avian.' The Eon duo are definitely more like dinosaurs in that they don't appear to have a beak and have ears, but dinosaur lifespans are awfully hard to come by.

As for other inorganic Pokemon, I did give this a go, but gave up after a while. Certain inorganic Pokemon look vaguely like organic creatures - Golem resembling a reptile, for example - but this is so unhelpfully obscure that putting any sort of math to it is beyond my capabilities. For now, they get the generic lifespan based on weight.

So, I just took a look at that sheet.

Imagine picking Tepig as a ten-year-old, having it evolve all the way into Emboar, and then it just dies of old age while you're studying at college.

I also find it weird that Grotle apparently has a shorter lifespan than its pre-evolution and its evolved form, but I like the idea of Swampert living for almost 80 years, though. I would have thought 40 years would be more likely, but this is interesting.

Most of the starters end up being surprisingly long-lived, but there is the odd and highly unfortunate exception every now and then. Generally, it seems to happen with animals that already weigh a lot, like pigs - they can't scale up as well as tiny axolotls or snapping turtles can. Speaking of snapping turtles, Grotle suffers from the unfortunate transition of Turtwig being based on a wood turtle to Grotle being based on a snapping turtle, and the scale gets tipped backwards accordingly (but shoots up again because of Torterra's weight).

Evolution families in general are a bit of a weird subject for this spreadsheet - I interpret them as how long each stage would have to live if left unevolved, but what happens when they do evolve? Does the lifespan reset since they're effectively changing all of their DNA into a new species and possibly re-elongating their telomeres, thus effectively reversing aging? Absolutely no idea.

I've since moved on to making a Pokemon 'family tree' - given what each Pokemon is based off of, what its nearest genetic neighbors should be. It's still mostly incomplete (want to include Egg Groups and Types to see if there's any patterns, minus the obvious Grass-type dominance in the Plantae kingdom), but I think I'll put it here when it's a little more fleshed-out. Let me know if you find anything else weird on this spreadsheet in the meantime - every correction just brings it closer to being accurate!