I worked with someone who didn't have what I'd call a scientific mindset, and it can really be a limiting factor. They confused correlation with causation, in other words were quick to launch on to theoretical explanations without testing or proving their thinking. When solving a problem, they tried tons of different approaches until they empirically found one that worked, without always fully rationally understanding the cause of the symptom they were trying to remedy. They usually succeeded, but the cost was the time spent, and the complexity of code that resulted.

There are a lot of ways to be a programmer, but that is not one of the better ways. The programmer was productive (prolific) enough, and the business was resource-constrained enough, that their work quality was not even close to consider firing or replacing, but I'm convinced it cost the team and the company a lot of productivity and effectiveness over time.