In my other job (the one that pays), I always caution students about using history.
The first issue is selection bias, defined as choosing your case on what you are trying to prove (the dependent variable). Given over 5000 years of recorded history, it is almost always possible to find a historical case or cases that will make your point. This is why I caution them about when anyone ever starts an argument with the term, “History shows . . .” For every case that students present, I can and do present a counter-example. This has been evident in the justifications by the Russians for their invasion and the Ukrainians for their sovereignty.
The second issue is interpretation. Even if there is agreement on which case, what that case means is subject to the analyst’s worldview. For example, when I was still a graduate student researching my dissertation, one of my cases was the Cuban Missile Crisis (or the Caribbean Crisis in Soviet/Russian journals). In 2000, I found over 130 peer-reviewed journal articles in the political science literature. As a friend in the history department told me – history told the first time is glorification, the second time it is told is revisionism, and the third time you are starting to get to the truth.
Although this is a problem in the social sciences, at least there are criteria and tests that are applied to fix both problems.
For a writer of historical fiction (my third career), it presents a challenge – how do I write an entertaining story with fictional characters and still present a true account for what happened? This is why caveats and history notes become so important.
And remember that as long as we have done our research, we are probably no better or worse than historians.