With higher end burgundies, you do not usually get a choice if you are buying them from a merchant charging an honest average retail markup. But then again, for the past 20 years you would be crazy to complain about such an arrangement because you could flip even the lesser vintages for cost or more and keep the ones you want.
Otherwise, I think it is a good idea and I do it with a few wines because part of the intellectual fun of understanding wine is to see how each vintage performs over time. And if the wine is good, you can find a specific use for most vintages.
Another reason it can be a good idea is something Michael Broadbent wisely said long ago- namely that it takes a good 10 years to really see where a vintage is going. 1995 is a good example in Bordeaux. The few concerning TNs I have posted here in the past year have been met in some cases with claims the vintage was never all that great- and maybe some of you knew that. But the fact remains that in the general press hype at the time it was touted as a great vintage. Or look at how 1986s have evolved. Better still, the 2003s which some people love and some people absolutely hate.
Even in the great vintages, some wines do not fare all that well. The longer it takes a vintage to mature- the more time the wine has to either develop in all sorts of magical directions. However, many wines may spend that time going out of balance. Looking back at the past 50 years, how many great vintages in Bordeaux have been uniformly at least very good at the vast majority of the more esteemed chateaux? 1982 and 2000 are the only two that come to mind. Perhaps 2005 as well, but I have not tried enough of them to claim that. In every other good year there are notable wines that have proved quite disappointing.
The opposite is true as well. Many less regarded vintages perform to the upside with time. 1985 and 1988 were well regarded at release- but not 5 star years except for Broadbent, but these days many of my finest Bordeaux experiences are coming from those two years. I think 2002 could prove similar upside potential.
Point being, you are never as safe as you think being too selective about which vintages to buy.
More important- what are your criteria for a “great vintage”. Is it power and critical score? Or are your great vintages someone else’s 3 star vintages. Personally, I lay in my biggest stashes of years like 2008 in Bordeaux or 2001 in Burgundy. Those are the vintages that will be of greatest use to me in time. Many others think this way too, but too many collectors without the self-knowledge that comes from trying a wide array of vintages get too hung up on the big and sexy years. Those wines are great in certain settings, but not all.
But back to your main point- once one has a fair amount of experience, I think you should only do complete verticals if you really want to see the complete story of a given wine over time, and you will pay a premium in both cost and use of storage space for that privilege. BUT if you do not have that experience to fall back on, being overly selective- especially if you are paying the secondary market premium for “top vintages” during the release hype- you risk paying the same premium over the longer term as some of those wines do not turn out as you hoped- quite a few of them potentially.
For my part, I backfill every single vintage of Magdelaine I do not have as older bottles become available. Same for Schaefer Spatlese Auction and #5. Those are the three wines I really must have every year. I would add Lafite back into that mix if it were a viable option- which it no longer is. Meo is a fairly new addition to my stable, but that one will be every year as well until I stop buying new Burgundy altogether (already about there with Bordeaux as I am now 46.)