If you aggregate loads of data from a large number of crappy or failed or pointless experiments, then you're going to end up with noise. It won't be a nirvana packed with stunning revelations about how things work.

Comments