The comment from @ChildrenOfTheQuorn is interesting, and does not reflect my person experience, or any data analysis I've seen. I worked in LMP for a few years, before transitioning to legal profession.
For the cases I've been involved with the the overwhelming trend is that the accused has a series of accusations against them, often going back many years. In many cases the accused also had a number of non-sexual offenses, this could be common assault, driving, burglary, etc. Their peers knew of their "reputation".
Regardless of the legal situation, as with other aspects of life, people make and will continue to make assumptions (if we hear someone's speaking voice, then we typically infer a lot within minutes, often incorrectly, their gender, age, ethnicity, region of origin, education level, etc.)
Rape is hugely underreported.
Weiser, D.A. (2017). "Confronting Myths About Sexual Assault, Family Relations, 66, 46–60" is a paper based on data from England, Wales, and several other countries. The number of false reports is in the low single figures. Of course, this percentage is inflated since it only looks at the ratio in respect to reports (and most rape is not reported).
The data clearly shows you should believe the victim. There is a small chance you can be wrong. False accusations are a very very serious issue, and they will happen.