I've posted already about how swell is formed and propagates across the oceans, but it is based on swell created by local conditions. But in an RPG, we're generally not concerned with the size of the swell created where the PCs are and how far it goes, we're more concerned with the size of the swell generated elsewhere and now arriving where the PCs are. The system I posted earlier is fine for this, if you're willing as DM to generate weather for the entire world, work out the swell generated in dozens of separate weather systems, and then figure out how much it decays in the distance it has to cover to get to where the PCs are. If you want to get really technical, you'd have to figure out how long it takes that swell to reach the PCs' area from its source weather system, not to mention all the lesser swells generated by that weather system (it stands to reason if it takes 50 hours to fully develop the waves, that there would be less developed waves created beforehand, which would also travel until they hit a shore).
No, the best thing to do here, from a game feasibility standpoint, is to work out some sort of statistical stand-in that would take into account all the directions a swell could come from, the likelihood of strong winds somewhere in those directions, and how far the swell generated by those winds would have to come. Roll some dice to see a generalized picture of the world's weather and go from there. A simple table wouldn't meet my standards here; the Hawaiian Islands, surrounded by thousands of miles of ocean in all directions, should not get the same results as Galveston, Texas, which only has any kind of long fetch distance to the southeast, and even that gets buffered by Florida, Cuba, and Mexico. I'll have to figure out some sort of equation, or a table complex enough to take everything into account, but not so complex that it can't be used on the fly if need be.
I'll probably have to dredge up my old statistics textbooks, too. If I can't figure out a decent way to math this problem out the smart way, I'm going to have to brute force it, and generate a world's worth of weather for several hundred days, and then figure out what kind of swells that produces in different places. I'm not entirely sure I want to take on that kind of number crunching if I don't have to.