Earlier quoted context omitted.
For most people, poker will probably more likely lead into a gambling addiction (or at least a habit of regularly flushing money down the train at the tables) than an inquiry into probability and statistics. Most people are probably just not inquisitive enough to dig into the maths behind poker.
I think you need a teacher to put the game in context. Let's put aside for a moment the optics of a school teacher having their class play Poker (even though obviously they wouldn't be using money). I can imagine a lesson plan going something like: 1. Have kids play some individual games with each other. 2. Stop the games, and go over some of the actual math concepts behind Poker. At this point, the kids are engaged…
Sure, if the "kids" at hand are university students, you could use poker. But I find it extremely dangerous to teach kids poker theory, who as theorized by the Dunning-Kruger effect might start believing themselves to be really good at it, and then proceed to spend a lot of time and money on a pursuit that might not lead them to happiness. You can teach statistics without making any new gambling addicts in this world.
I know the majority of the students wouldn't be affected and the effect could be even positive, but for some the drive to gamble is so strong that they really go off if that predisposition is nurtured. Although perhaps they are ticking time bombs either way, so no amount of protection will save them. But yeah, you got to be careful teaching kids positive feedback loops on things that might have serious drawbacks.