I don't understand not re-brewing, I mean you can be a good brewer and all but if you can't reproduce your beer...your getting lucky?
For me, once I started brewing, I noticed competitions. Then I noticed they had categories. Lots and lots of categories: the BJCP style list. I think, counting cider/mead variants, there are 70+ substyles of beer!
Your brewing process improves overall, even when changing the style every time you brew; plus, you get to see how certain components, and different yeast strains, can have a dramatically different flavor profile.
What does a classic Baltic Porter taste like? I have a good sense now (or at least how it differs from a Robust Porter, etc). Ive only been in the hobby for 1.5 yrs and have brewed about 35 batches (counting some aphelwine and meads), and still havent covered much ground with lagers.
In my rotation, I do a classic style (cuz Ive probably havent tried the style yet), then a clone (cuz I know what it is supposed to taste like), and usually a weird beer (maybe a Caribbean pirate beer with ancho and chocolate? banana chocolate milk stout, anyone?).
Plus, if you like the big beers (8+ months) or the long ferm time beers (Flanders Red, +14 months), seasonal beers (Oktoberfest, Christmas, etc..), you gotta plan ahead and stick those in at the appropriate times.
That rotation keeps me busy enough.
While I do agree with your repeatability point,
how do you know youve explored enough to want to repeat?
I would say your process control gets better over time.. I know my brew day notes are WAY better now than even 6 months ago....Sometime, especially with a classic style recipe, you brew it never having tasted anything like it before, and you nail it, and you know it; you know its a classic style for a reason, and its not some magical flaw.... if your notes are good (and you have proper amount of yeast at the right temperature for the right amount of time and all that), I dont know why you wouldnt be able to reproduce it.... just have to the variables controlled! And the only way to learn about the important variable, is to brew a bunch, and I dont think it takes doing the same kind over and over again to achieve that.