The concept of tuning your forecast to a desired accuracy is actually pretty interesting. For example, say you need a weather forecast that's 99% accurate. A meteorologist will then tell you how far into the future you can go. In this case it might be 10 or 15 minutes. It might sound silly to us but there's probably a use case for it.

I don't imagine the folks at the Max Planck institute are slouches, so I'm assuming there's a use case for scanning some literature and determining some outcome with 99% accuracy. It's probably not a very profound prediction, but again, it wouldn't surprise me if there's a perfectly reasonable use case.