By definition, professionals don't play "trick moves".
Just because it looks strange, and it may trick you, or me, it doesn't mean it is meant to trick his opponent.

p4 did seem a pretty clumsy shape, it's like white played p4 first (kinda slow but not totally stupid), black played o2 knight's move which is sensible (block also possible, is it better/worse?), and then white played crude p3 push with p2 block answer bad exchange. But white then played s5 hane and black answered at s4 so white managed to sort of cut off s8 in sente. So my thinking is s5 was sente thanks to p4, because if black ignores white has the r3 tesuji. Maybe that was Kim's plan and reason for p4?

To bring this back to technical discussion, P4 is clearly a dubious move that LZ even did not consider. By playing P4 white lost about 5% of winning rate, although the significance of that cannot be quantified easily.

But for analysis I would want at least 10K rollouts for each main option (all four in this case)

To John's message: the definition for "trick move" that I objected to is the one where correct reply from the opponent will result in a loss, so one would make such a move either just for fun (in an unofficial game), or when one thinks that their chances of winning are low (the opponent is much stronger, so the chances of playing normally and win are lower than the chances that the opponent will not know the correct answer to the trick move, and get tricked).

Of course, if the game is bad and one is losing anyway, pros may try one last chance with some tricky-looking move, but that is outside of the above definition (since the game is already lost).

To Uberdude's message: I am aware of Alex Dinerstein's trick moves study; I don't think he would play any move that would lower his perceived chances to win the game though. I assume he may try some new trick move against top pros maybe, but not against someone at his own level - just my guess, I don't have any record of when/if he actually plays the trick moves he researched.

I was using my laptop which is not powerful enough to obtain 10k within reasonable amount of time. But even if we have 10k, I doubt it would pick up other locations like P4.

BTW, to use LZ, we seriously need to get a proper GPU. Last weekend, my son asked me to show him the inside of a computer after reading a computing book. I found a 12 year old computer in my garage and it actually still booted without problem. It has an old NVidia card, a card about 15 generations older than the current ones. When I compiled Leela Zero on that machine, to my surprise this very old card still managed to generate rollouts 4 times faster than my brand new laptop (using an Intel i7 CPU).