The problem for me isn't that he did the study. The problem is that he brought kids into a program he designed and was working, at least partly, under the hypothesis that weighted balls cause injury and was looking to both measure their effectiveness and measure the injury rate, which he expected, given his previously published opinions, to be high. I would guess hie disclaimer on release was nothing more than a warning that WB's could be dangerous and the participant takes the risk. However, given the facts, I don't think this is enough. If you were the parent, would it not be an important thing to know that the guy running the test believes they are inherently injurious and was putting your kid in a program for the purpose of quantifying just how dangerous they are?
To give a better example of how to study this, look at ASMI's ten year study concerning youth pitchers and the curve. They started the study with the hypothesis and belief that throwing curveballs at an early age was a significant factor in arm injury rates. It would have been unethical for them to pull in 50 kids and ask them to start throwing curveballs, especially given their sincere belief that they were harmful. Instead, they went and found subjects already throwing them and asked them to log and report their experiences. There's nothing unethical about that, but never would they have done it like Mike did.