Court opinions often get facts wrong.

From the Volokh Conspiracy:

In my Academic Legal Writing book, I caution students who are writing law review articles against relying on court opinions’ factual assertions about social science evidence, or even about past cases. Always read, quote, and cite the original source, I tell them (though I realize, of course, that many lawyers don’t have the luxury of taking the research time to do that). Don’t let the intermediate source’s errors become your errors.

The problem goes beyond the injustice of a particlar case. The courts are given too much credit for getting facts right. That's simply not true for social science "facts" which are often the result of sloppy of biased scholarship.   In the comments "Hans" says:

You can bet that this false statistic will now be cited, despite its obvious falsity, by countless other court decisions, by hired expert witnesses in domestic violence cases, and by lobbyists for domestic-violence advocacy groups.

False statistics often get invested with credibility by court decisions that cite them or create them.

For example, Lenore Weitzman admitted that she erred in claiming that men’s living standard rises by more than 40 percent after a divorce while women’s falls by more than 70 percent. (These figures were mathematically impossible, even if no child support or alimony were ever paid. But they were made in her 1985 book The Divorce Revolution. She later admitted the error in a 1996 AP story (See Associated Press, “Study Goofed on Gap in Post-Divorce Standard of Living,” MANCHESTER UNION LEADER, May 17, 1996).

But before she admitted the error, it got cited in hundreds of state court decisions, as well as an opinion by Justice Brennan and Justice O’Connor.

As a result, this false statistic continues to be included in family-law textbooks, and continues to be parroted by courts. (On average, spouses of both sexes experience a decrease in standard of living immediately after a divorce because of the added expense of maintaining two households rather than one. Men DO NOT fare better than women in divorce. If anything, men fare slightly worse on average, and some end up broke as a result of a divorce).

And part of the problem is the lack of scientific education in the Legal community.

To me, this is really interesting from a legal informatics point of view mainly because it’s now largely accepted that courts have been citing to non-legal resources (or ‘extrinsic aids’) with greater and greater frequency...but information literacy education for law students and legal professionals is pretty much non-existent. Which leads to things like the warm embrace of Wikipedia articles in both court opinions and law review articles to a level that would never fly in other scholarly writing, and blissful ignorance of the simple concept that if something is on Wikipedia, it also almost definitely exists in an actual reputable source.