My guess is that sloppiness is more likely to occur when it results in convenient conclusions. If the data looks good, "Hey, it must be correct! No need to double-check."
https://twitter.com/user/status/1752136421672964180 honestly it is frustrating that people choose to get most upset about *this* (bad, but mostly postdocs cutting corners under non-supervision of demanding PIs who just don't care, in ultracompetitive labs..) rather than cases of major fraud that are much more egregious & impactful
"Journal editors took a similar tack, ignoring his letters, rejecting them, or investigating on timescales bordering on the glacial." Sounds oddly familiar ...