Because the kata serves as a catalog -- and also can teach rhythms and strategies, as well as collections of techniques. And can be a useful way to practice alone.
That's it in a nutshell.
This is the way I think of it. It's a cliché (but only because it's true!) that for each
movement in the kata, there are several different plausible SD
moves, depending on the interpretation of what moves the preceding and/or following movements are given. Let's say that each movement corresponds, on average, to any of 3 alternative moves. Then for any given three movement sequence in a kata, we have on average 3^3=27 different combat scenarios. High quality bunkai—the kind of stuff that Abernethy or Bill Burgar or some of the other people writing about the Pinan/Heian set and Naihanchi have published, or our own Stuart Anslow and Simon O'Neil in TKD —indicate that the way to look at forms is as a collection of short combat scenarios, possibly related by 'theme' (the way Abernethy sees the Pinans), but each involving a different attack/defense scenario. Let's say the average length of each such segment of a kata is three. That means that in a 30-movement kata, we have 10 3-movement subsequences. But since each such subsequence has 27 different interpretations in principle, it means that the kata is a summary—a catalogue, as jks put it—of
270 possible attack/defense scenarios. Let's be conservative and say that half of the combinations turn out to make no sense. That still leaves us with 135 possible attack/defense scenarios—more than enough to meet just about all of the 'habitual acts of violence' that Patrick McCarthy and others after him have identified as the statistically most typical attack initiators in street violence. So that means that each 30 move kata you've learns amounts to a collection of 135 short stories, each of which has someone attacking you in the first paragraph and
you walking away
unscathed in the story's conclusion—
them, not so much! :EG:
I know that this is an idealization. But it gives you an idea of the order of magnitude of the information contained in a single kata. I've been to Hapkido seminars of one kind or another, and the combat techniques are many, varied and brilliant; but there is so damned much to
keep track of... not easy! But HKD doesn't have kata. And once you learn a kata, it's like a portable mini-encyclopædia that you can always dip into for new insights and fighting techniques. Not surprising that Funakoshi and Motobu are said to have relied primarily on the Naihanchi set alone for their theory of combat techniques, eh?
That's what I think the great advantage of kata is in practical terms. It's not that difficult to learn a single kata or hyung, in terms of time invested. But the technical
depth of those forms is great. So, for a relatively little bit of time, a very, very efficient yield in terms of results. The original karate masters were above all else practical—that's why kata were important to them, and should be to us too.