Imagine yourself alone on the plains of Africa and facing a lion or two. Your odds are not good. Even a group of people on the plains of Africa facing a few lions with no ability to throw stones looks like a losing proposition.
However a group of people in an outward facing circle facing lions and each with a pile of stones to throw with deadly accuracy, now this is something that even lions would think twice about.
The thing about defence is you don't have to move, the stones are prepared before hand. The predator has to come to you and is therefore easier to hit, preferably in the head. Sticks and stones will break their bones but names will never hurt them.
You wouldn't even need to be strong enough for lions to start with; every step up the pecking order would be an advantage.
I don't envy paleontologists and archaeologists trying to make a case for "why" an adaptation emerged when it has so many plausible advantages, especially when cultural developments might have happened too fast or at too great a distance in the past for us to have an accurate idea of their timing.
See Table 1 https://www.apa.org/pubs/journals/releases/amp-606581.pdf
The Cohen's d value here is difference in means between two populations, normalized by the pooled standard deviation. What d = 2 means in concrete terms is that an average man can throw faster and further than ~98% of women.