People can find certain stats to make a player look good or bad by cherry picking or using invalid samples. Most people know to take that with a grain of salt.
Studies are not like that unless they are done by crackpots, which would never fly in the sabermetric community. Believe me when I say that if one person or group does a study and concludes something like not batting your best hitter in the 3rd slot, that study is going to be reviewed and scrutinized under a microscope. If there is any fault to the study, others will point it out. If there is any way to improve upon it, others will point it out.
Additionally, they will not be satisfied. They will continue to slice up the data and expand upon said study in many different ways. It's what these guys live for.
You seem to be under the impression that stat geeks are trying to prove traditional thinkers wrong, and that they are 'fudging' their data to do so. That is not the case at all. They're not trying to prove anyone wrong. They just want answers to their questions, which IMO, is far better than accepting what has been done for the past 100 years just because that's the way it's always been done.