It’s similar to a business principle casinos run on, since it’s still functionally random events that pay out big occasionally but most of the time is paying the business for no return. I think the betting/casino scenario is simpler though
So imagine you create a game of chance, and in this game, it pays out a jackpot of $100 when they win, and you need to decide how much to charge per play. You do some math and realize that on average, players will get a jackpot every 20 plays. This means you want to charge over $5 per play, let’s say $6. This would mean every 20 plays, you have to pay the jackpot of $100, but you made $120
Now of course, it’s random chance, you may have 2 players win in a row, so now you paid out $200 and made $12. This happens, but the thing is, there’s TONS of customers playing this game, and while there’s some random chance involved, as you get more and more customers, the actual percentage of time you pay out will look very close to the 1 win per 20 times played.
A lot of insurance works like this, thru calculate how often they’d need to pay the customer, and craft the prices accordingly so that they come out top in the end, where across all customers, they’ll make $1.20 for every $1 they pay out (made up numbers, but the point is they look to craft prices to make more than they pay out)
You’ve likely heard of insurance charging some higher risk customers more, like charging smokers more for life insurance or young men more for car insurance, this is because that “1 win per 20 plays” might be closer to “1 win per 10 plays”, ie they need to charge more because they’re more likely to have to pay these people
The people who figure out the likelihood of paying out based on known things are called actuaries, and nothing works in insurance without their estimates of how likely they are to have to pay out.
Latest Answers