Perhaps. All I can say is I was told that too; I tried that; and it really, really didn't work out. I think I might have been more successful had I focused on my own interests more. Certainly looking at my career in my adult life, almost all my biggest successes, with maybe one exception, were when I chose what to work on instead of agreeing to work on someone else's idea.

Of course you do need to adjust this for your field. If you're working in pure math or Roman history, it's not all that hard to do your own thing. In experimental high energy physics, maybe not so much. If you need a million dollar laboratory to get started in a field, then you may not have a lot of choice in what you work on. Though even in the experimental field I'm most familiar with, observational astronomy, it still appears to me as if the most successful people did their own thing. It probably does matter than in astronomy, it's standard practice to allot telescope time based on proposals rather than ownership.

Also, of course, if you can decide early on your area and aim at the program that does that well, then that's the best of both worlds. If you know you want to work on high temperature superconductivity, you're better off in a department that does a lot of work on solid state physics and better yet superconductivity specifically rather than one that specializes in string theory or experimental high energy physics.

In any case, however, I think it's pretty important to start doing some kind of research as early as possible. My own experience is with math grad school, where it's common to just focus on taking classes for the first year or two; but it's better to also be doing research during that time if you can.