By "impractical," I am assuming you mean an aggressive / optimistic cost and or schedule to complete the project.
There are other objectives and problems with which managers must cope that the programmers, engineers, trades, laborers, and technicians do not. These other things create a perspective on the business of which others may not even be aware.
A few examples:
1) If you're selling services, you have to win the project. Estimates live on a range of possible results. Estimating on the fat side of that range is a nice, safe place to be and puts a lot of likelihood in your hands that you can achieve the scope within that fat budget and schedule. However, fat is expensive and you are likely to lose the work. Being practical may mean no work.
2) There are a host of estimating biases that a manager needs to sift through. Optimism bias, anchoring, recency, selective to name a few. This makes what is practical and impractical much less obvious.
3) As a manager, you learn quickly how amazing it is to watch the team meet a challenge. Your team screams it is going to take 3 months to finish this task; you challenge them with 1.5 months and watch the team either hit it or come very very close. This attacks this notion of Parkinson's Law or Student Syndrome, where a long runway means every inch will be used before takeoff, for no other reason than it is there. Forcing a constraint on a high performing team enables innovation and problem solving they would not have otherwise even tried. And those innovations will be used next time and that will set your team even further apart from your competitors.
It is another way of thinking, being a manager and leader.