There are a zillion different substances classified as supplements. It's reductionist and intellectually dishonest to lump them all together. Some are effectively just placebos, or even harmful in high doses. Others appear to have beneficial effects. In many cases those effects are subtle, and might only be noticeable to people who push their bodies to the limit. Unfortunately, this is almost impossible to study through high-quality randomized controlled trials given the numerous confounding factors.

Let's look at it empirically. Among elite athletes, how many take zero supplements? I can't prove that they're getting any benefit but the correlation between performance and the use of certain supplements is so strong that I suspect there's something real going on. (And by supplements I mean real supplements, not PEDS which are an entirely separate category.)