There's a bit of circularity here, since I acknowledge that it is possible to think about belief in such a way that it would not be voluntary. But I voluntarily choose to think about belief as voluntary, namely as having a definition that implies that it is voluntary, because I think that the consequences of thinking about it this way (both epistemically and instrumentally) are better than the consequences of thinking about it in such a way that it would be involuntary.

The reason both are possible is that saying that someone believes something is a vague generalization. It does not have rigid borders. It normally includes both voluntary and involuntary aspects, and we normally expect these things to go together. But when we consider edge cases, there are different places where we could draw the line, and common sense does not sufficiently determine the matter. Consequently we have to choose. I choose to draw it by saying belief is what you voluntarily treat as a fact. I think that this corresponds better to common usage than alternative definitions, even if it has a few odd edge cases; they are much less odd than the ones that follow from definitions implying that it is involuntary.

"Voluntary" is not a confused term; it means "because I wanted it." Feeling like I made a choice would often be a consequence, but not always, since in some cases I would want something so much that I can't conceive of wanting anything different. When I say that belief is voluntary, I mean that people have beliefs because they want to have them.

What I mean by "belief": treating something as a fact, namely in all that ways that one is able to do so. So if you have an involuntary expectation of something, I do not count that as a belief unless you choose to act as if the thing will actually happen; if you choose to act as if it will not, then I say that you feel an inclination to have that belief, but choose not to have it.

I understand why you're giving the definition you suggest, and I agree that expectations are involved in understanding the meaning of any statement (we've had that discussion before.) Nonetheless, you cannot understand the idea of "this will correspond with my expectation" unless you already feel you understand "this will happen." So at least the idea of correspondence with reality has to come before the idea of fulfilled expectations, even if we cannot fully cash out the idea of correspondence with reality without talking about our expectations.

I agree that people's concrete beliefs involve many involuntary things, and I agree that theoretically you could define belief to refer to some of these things. But I do not think this corresponds best with common usage, I don't think it gives us the best understanding of what is going on, and I don't think it has the best practical consequences.