Pro coaches and GM's, hockey included, are always inclined to make the "safe," by-the-book" choice, even when confronted with evidence to the contrary. They feel safer going with the big-contract name player because they feel they can't be faulted, even if the name player screws up royally, whereas if the upstart doesn't get it done, then they have to answer why they took "risk." Everybody wants risk-free, easy choices. Life doesn't work that way, of course.
In fact, that's one of the explanations for why Las Vegas is so "surprisingly" good this year. Most GM's stayed with the tried and true. the name players, and, therefore, sacrificed very talented young players (ahem, Schmidt), as well as other talented, but less "marquee" veterans. The result was that the youngsters got an immediate chance to make an impact, and the more seasoned guys had high incentive to show their former GM's that they made a mistake.
I'm near certain that Trotz, in the deep recesses of his mind, would somehow wish that Grubie screwed up, so he could just pick Holtby and not have to explain himself, or get sweaty palms by starting Grubie in the playoffs. It makes little sense, when one looks at Grubie's sterling stats, as well as the unavoidable truth that Holtby has shown conclusively that he gets deer-in-the-headlights syndrome when it's time for the crucial save in the playoffs. Playing Grubie is the better choice, and not just because he's hot now; he's been hot for months.
I guess there is life without taking risk. It's just that it's played on the golf course, not the hockey rink.