Everything is a guess if you want to take it to the extremes but estimating based on experience and historical patterns is far more justifiable. It should be observable that gains are far more likely to occur after halvening then before.
Of course but people massively over-estimate the value of apparent patterns and the likelihood of them being repeated. Take the current situation as a perfect example ... in all previous halvings, the price has tended to go down approaching the halving and only to rise significantly quite some time afterwards. However, this time we have had a very significant rise (including a new ATH) beforehand. I didn't see anyone predicting a new ATH in early 2024 six months ago.
We've only had three halvings and three data points is just not enough to give future predictions much certainty.