In my opinion? Of course it’s gambling. Where I would define gambling as wagering money on a predefined event for a predefined payout and little to no payout for the event not occurring. In regards to where they got into it with the CFTC (essentially being able to bet on elections) they pitched it as essentially the same as other forecast markets that are legal wherein you’d make the wager as a hedge against other financial hardships. Bet on the thing you don’t want to happen and either you’re happy it didn’t or you get paid if it does. Though in that way insurance is a bit like gambling.

This boils down, as far as I can tell, is the CFTC has preliminarily been told that its authority to exclude these kinds of contracts only apply to “gaming” and not “gambling” (where gaming involves being a participant in a game of stakes). So Kalshi is trying to say that essentially supersedes state laws. I doubt that will hold water in the long run, but you never know until a court rules.

Whether or not many people have estimated the odds is not what defines what a bet is, it's just something that people placing a bet are going to look at.

Derivatives exchanges are handled separately from betting because they have other tangible benefits the to financial marketplace. I'm not sure I buy the pitch that prediction markets should fall into the same carveout at face value, but why has nothing to do with how many stats are run.

> We have stats upon stats to help us predict outcomes in elections

2016 called

> Rufus Peabody, who changed sports gambling with his data driven approach

FWIW, quant sports betting trading firms have been around for a long time, just that they tend to be extremely low key (bookies ban people who win too much).