To the particular point of Norman Cantor's credit on the book you're looking at, I too am a big Cantor fan but he also kind of had a really bad drop-off at the end of his life. The Last Knight in particular was not terribly well researched and lacked a lot of the panache that Cantor's other work had. Perhaps this accolade came from that twilight era of his career? It's hard to say. I do think that knowing that a good historian liking a particular book is a good use of appeal to authority; sadly, it sounds like that may be one of the times where that heuristic might have done you wrong.
Anyway, generally speaking it's not easy and I don't think there's a perfect solution, but here are a few guidelines...
Extraordinary claims require extraordinary proof.
This in my opinion ought to be the first maxim you hold on any subject, be it history, medicine, physics, or whatever. If someone writes a book about how King Tut had blue eyes rather than brown, that's not (I don't think, anyway) a particularly extraordinary claim so you can probably let it slide. If the author says that the ancient Greeks and Romans stole democracy from the Egyptians, that's a bit more extraordinary of a claim. That doesn't mean that you ought to dismiss it outright, but neither should it be accepted without evidence, even as a premise.
Does the subject matter conform to current opinion on the subject?
I'm not saying that if it's not mainstream, it's crap, but that should be your first question. If you're reading about, I don't know, the historicity of Jesus (note: I present this only as an example, not as a personal opinion on the subject) and the author says there's no proof, then you would behoove yourself to do a quick Google or Wikipedia search to learn that in this case, the vast majority of Biblical scholars do consider Christ to have probably been an actual person. This in and of itself shouldn't be enough for you to put the work down, but it is a question which IMO needs to be ticked off.
The flip side there is that it's entirely possible that you're reading a work and your skeptical buzzer was set off by that first question. "This guy thinks the world thought the Earth was round when Columbus sailed to the West, and he was thought of a crank primarily because of the (correct) viewpoint that the world was far, far larger than what he'd thought it was?", for instance, could be a question you ask yourself when reading a particular book. In that case, a search of the 'Net will discover that yes, this is in fact largely established history.
If it's not the popular opinion, does the historian acknowledge this? How does the historian explain this?
Bob Breyer, an Egyptologist who teaches at Long Island University and who appeared on the History Channel every now and then before the History Channel turned into Space Alien Central, has a lot of, shall we say, not commonly held opinions about his subject matter. For instance, he holds the opinion that it's possible, perhaps even the most likely scenario available, that King Tut was murdered by his vizier, who then became Pharoah. Breyer goes into this in some depth in his The Great Courses Entry on the subject. I highly recommend this series of lectures, by the way.
The way Breyer approaches this is exactly the way that a historian holding a minority opinion ought to approach it: he tells you straight out "this is not the dominant opinion by any means". He talks a little bit about why others might be led down a different path without calling them fools or marking them as part of a giant conspiracy of thought. He also acknowledges the faults in his idea (namely, the sheer lack of evidence). In this case, he uses it as an example of just how little we know about ancient Egypt, but even if that wasn't the point, the important bit there is how it was handled.
How well-sourced is the evidence? Is it available to review?
I remember around a decade ago arguing with someone about whether or not Lord Amherst gave Native Americans smallpox-infested blankets. In that case, it was very easy to win this argument: there's a letter he wrote advocating for it which has since been scanned and is available online. That's about as open and shut as you can get. Well, open and shut if I was actually right in my assertion; there's actually no evidence that he literally gave the Native Americans the blankets, just talked about how "inoculating" them was a good idea. But we did get an answer, and that answer was based on that letter.
Of course, not every document ever written is going to be available online or even by the layperson (for instance, for a historical novel I've written but not published about a particular 19th century person, I had to fly to New York City and convince the librarian of the school of journalism at Columbia University to let me read a 1930s era master's thesis on this person which was composed by talking to their friends and relatives). However, the fact that it's even available to someone means that there is at least an avenue for this historian to be proven wrong.
On the other hand, if the person's primary sources consist of unsourced "personal research", well, it's entirely possible that person is completely on the up and up but how would you ever be able to vet them?
I think the key phrase to remember here is trust but verify. Unfortunately it's not always possible to verify, but noting that the historian at least let the door open for others to verify their work is at least a good start (ideally, you might want to then take a look to see if, indeed, their work has been verified and if so, did anyone think the author was full of hot gas, but again, we don't always have the time).
How reliable is the historian?
This one is potentially dangerous, as there are a couple potential faults:
- Every historian can have a bad day. Cantor is a great example but a lot of people enjoy Stephen Ambrose's writing, too, and he produced some real stinkers as well (and he appears to have been a chronic plagiarist, but that's another matter altogether).
- A person who is an expert on one subject is not necessarily any more likely to be an expert on another subject than me or you. This isn't history, but towards the end of his life the great molecular biochemist Linus Pauling became convinced that taking massive quantities of Vitamin C was the key to living a long and healthy life. Since then this has been utterly refuted and, frankly, there was little evidence to support Pauling in the first place, but it lives on because he was such a smart guy in other fields.
That being said, "has this guy done me wrong before" is a pretty useful heuristic in my book. If you come across a weird fact you hadn't heard before in one of Barbara Tuchmann's books, I think you're probably better off accepting its veracity (again, trust but verify) than if, say, Dan Brown had some weird little snippet.
Use your resources! And ask around!
SE.History is still in its youth but by no means is it the only place on the Internet or for that matter in real life where you can ask around about this kind of thing. If you're going to school, by all means ask your history professor or teacher what they think (trust but verify them too!). If you've got a school near you, you might try emailing someone - my experience is that as long as you don't sound like a crank yourself, a lot of academics love to talk about their field of expertise (some may even say that's why they became an academic!). Depending on your particular subject matter, places like the James Randi Educational Foundation, the Skeptics' Guide to the Universe forums, or even that old standby Snopes might have either a published essay on the thing you're looking up or a helpful group of fans who can help steer you in the right direction.