I must be missing something. It wouldn't be the first time.

Sure.

Here's a way to think about it. Imagine a typical civil court, who hears lots of typical civil suits, one between a plaintiff and a defendant. Plaintiff files the suit, defendant defends. Over the course of a year, assume that about half the time the plaintiff wins, and about half the time the defendants win. Does this mean that half the plaintiffs are pursuing specious, invalid, or "made up" cases? And that half the defendants are asserting specious, invalid, or "made up" defenses?

Not at all. Because in some amount of cases - theoretically in as many as all the cases - it may be true that both sides have a decent argument that deserved to be heard in court. But sometimes those arguments lose. There is a lot of space between a case that loses and a case that was improperly brought. Some of those plaintiffs will have colorable claims that they should win, but end up losing; and some of the defendants will have colorable claims that they should win, but be unsuccessful. That's especially true because there are lots of winnowing processes before trial, so that most of the speculative "gaming the system" cases get weeded out.

It's entirely possible that every single party had a "good enough" case that they deserved their day in court, even though half of those people lost.

And that's true in immigration court as well. Some non-trivial portion of immigrants will have a legitimate, colorable claim that they should be granted asylum - but they might still lose. That's especially true because they are poor and are unlikely to be represented by private paid counsel. It's more especially true because there is a winnowing process before they even get to a hearing, so the really invalid cases get thrown out before they get to a judge. They might not prevail before the judge, but they still could have had a decent enough case that they deserved to have a judge hear their arguments.