It's a means of weaseling out of responsibility for one's actions and a dereliction of one's duty as a scientist. Whenever we (and I use "we" here to mean anyone, not just scientists) interpret data and their results, we necessarily inject our own judgements into it (not the least of which is the answer to the question "what data to I use?"). When someone asks our advice, it is our duty as an expert to take ownership and responsibility for that advice, not cower behind platitudes.
For example, if someone were to ask me my opinion of the minimum wage, I have a duty to give them a proper, well-considered answer. I need to weigh the various outcomes, models, data, etc., and offer a reasoned opinion. I will reject models like monopsony because I find their reasoning and empirical support suspect. Are there data that lend support to that model? Yes, but I do not follow that data because I judge it to be poor and unreasonable. If I am wrong, so be it. I will be wrong on my own and accept it, rather than these weasel words of "I just follow the data" or "the data changed."
Why is all this important? Why is taking responsibility important? Because I, just like everyone else, am a flawed human being. We are made in God's image, but we are not God. My judgements will be based on numerous things and I may be practicing motivated reasoning. By taking responsibility, it forces me to honestly consider why I was wrong, and how to prevent it in the future. And it forces me to consider criticism of my opinions rather than glibly dismiss them by saying "I am just following the data!"
Galileo once said: "I do not feel obliged to believe the same God who has endowed us with sense, reason, and intellect has intended us to forgo their use." When we simply "follow the data," that is what we do: we forgo the use to sense, reason, and intellect. We act viciously, rather than virtuously. We insult not only ourselves but our fellow humans with our vicious behavior.