I'd still rather they be assisting with the making of a new game like with BOTW, but I think I've said more than enough in the past about how I feel about last gen ports on Switch so I'll refrain from another spiel and try to look on the bright side, that at least the game might have a chance to reach an audience now that it never did on Wii/3DS.
Well, this involves a major misunderstanding on how much resources are needed for a new game and for a port or a remaster. That said - ports from a HD-console like WiiU need much less resources as a HD-remaster like this. Let's get into detail a bit.
A port needs only a small technical team for a short time. Most of the work is done by the compiler here, the technical team has to fix some issues that arise from the porting and test it. As games need only very small technical staff, this can be done aside a bigger project. Actually with a port to a new machine the programmers can learn the ins and outs of the new machine, so I'd argue that a port actually HELPS getting a new game done faster and better (as they learn better about the machine, they can make the software better performing).
A port like Mario Kart 8 also includes a bit more content. But for this you need what, a level-designer to create the new levels for the Battle Mode? And a bit of designers for the additional characters? So a few weeks worth of work for maybe five people? Compared to years for a team of hundreds of people for a new game. I guess that fits nicely into downtime. This is, if a project is completed and the new project is in it initial phase, you don't need all designers yet, as there is not yet this kind of work to do. Even if not in the downtime, this takes minor resources.
Now a remaster like Xenoblade is more work. You need to create more detailed textures for HD, I don't know if they possibly also remade some models, most likely character models (as they are seen most often) with all the needed animation. And probably some more to be done. So that needs a team of ten, twenty, maybe thirty for a few months. If you make them work for a year, you would get a small scale game (say Kirby Clash as we seen in the direct). For a game in Xenoblade dimension you need hundreds of people working multiple years. So from a resource standpoint you get say ten remasters of the Xenoblade size for one new game. And it is a proven game. I think these resources are well invested, if it is a game like Xenoblade.
So ports are not cutting into new game development, maybe even helping the development of new games by faster acquiring skills. A remaster uses resources that could also be invested in a small scale game. And that is something to consider. Do you want a Kirby Clash vs. a Xenoblade remaster? And that answer can be differently for different sorts of games internally pitched and the back-catalog possibly to remaster. But for a new project the size of Xenoblade or BOTW you need a much bigger team for a much longer time.
And also: a game needs very different roles to a different amount. You just can't use a programmer to create a character model or make sound effects. There are to be downtimes for certain employees at times, as their work isn't needed currently in the big game project. To use such personnel in small side projects like ports, remasters or even stuff like NES remix seems like a valid use of their resources, so they still improve upon their skills and you get a game to release. Because the reality is: most gamers don't care if the game is an old game or a new game. They want something to play which is fun.