Because all of the other theoretical models that attempt to explain the observed redshifts fail to match observations.

> Maybe... just maybe... that's because it isn't actually expanding, but there's some other, unknown effects at play?

Nope. Already been tried. Doesn't work.

> It's about time we think outside the box.

Cosmologists have been trying out of the box ideas for decades. All of them failed. That's why we have the models we have now: they're the only ones that survived that process.

> I get it, both make the calculations work out, but they're basically "we have no clue what this actually is, here's our best guess" kind of variable.

For dark matter, you are correct: we don't know what it is, and the term is basically just another way of saying "whatever we need to add to our model to make it match observations works basically like ordinary matter does in the equations".

But that is not true of expansion itself. All of the alternative models that have been proposed to eliminate the need for dark matter still have an expanding universe. The expansion itself is a much more solid conclusion than dark matter is.

Another way of looking at it, to add to my previous post: dark matter and MOND are thinking outside the box. They cover both possibilities for how to expand on our current theories: either (a) there's more "stuff" out there than our current theories of particle physics know about; or (b) there's more aspects to gravity than our current theory of gravity knows about. Consider whatever "unknown effects" you like: they will end up coming down to one of those two possibilities.

Thinking "inside the box" would be something like: it's far more likely that either the data is wrong or we haven't calculated the predictions of our current theories correctly than that our current theories, which have tons of experimental confirmation, are wrong. Historically, most of the time in science, when there's been a discrepancy between theories and data, that is how the discrepancy has ended up being resolved: either we've figured out something was wrong with the data, or we've figured out that something was wrong with how we calculated the predictions of our current theories.

The reason why cosmologists are driven to consider models like dark matter and MOND is that they have checked and double checked and triple checked both the data and our predictions from our current theories, and the mismatch hasn't gone away. So they are driven to consider "out of the box" ideas, and, as above, dark matter and MOND cover the possibilities.

Where do you get the idea that it "makes the calculations work out"? You got it backwards. The calculations showed that the universe must expand. Georges Lemaitre found that out. People were skeptical at first and said the calculations don't apply, but when the cosmic microwave background was discovered in the sixties, all but the most stubborn hardliners were convinced that the universe must expand and that there must have been a big bang, i.e. a singularity in the finite past.

The microwave background is basically a picture of the universe when it was a baby. There were no galaxies, only hot gas that just cooled down enough to turn from plasma to neutral gas which later clumped into galaxy clusters, galaxies and so forth. Because the speed of light is finite, we can literally see how the universe looked liked billions of years ago. And it looked hot, because it was compressed, and it looked young, because structures hadn't had the time to form yet. The gas composition also shows only light elements, because heavy elements need to be forged in supernovae. Young stars have lots of heavy elements, far away, old stars only consist of hydrogen and a bit of helium.

The tiny, tiny irregularities that we see in the cosmic microwave background match beautifully what we know about thermodynamics, statistical physics, quantum mechanics, general relativity and electrodynamics to a very high degree. It's marvelous, really.

> the very idea that the universe is expanding makes about as much sense as dark matter

The universe does not care about what makes sense to us. It just is. And the expansion is an observational fact.

On a side note, I find it a bit presumptuous to assume that a layperson knows better than legions of professional cosmologists. These are very smart people who work full time for the better parts of their lives on these problems while you clearly haven't put in the time to understand the fundamentals. They wouldn't come to the conclusions they come to if they didn't think they had merit. You can be curious about it, you can have questions about it (please do!), you don't have to understand it, but please trust experts on their opinion and show some humility and some respect. I'm sure you wouldn't appreciate a physicist who tried to tell you how to design your programs either. (We're known to write horrible software, just look at ROOT.) Imagine some amateur telling you "for-loops don't make sense to me, why don't you use GOTO". No offense, and sorry to say it like this, but it really bugs me a bit.