The city of Denton has decided for financial reasons to go with a self-funded health care plan for city employees beginning next year. It looks like a good financial bet so far — the city estimates it’ll save almost $1.5 million next year versus what it has been paying for traditional health insurance — but like all bets, it is not without risk.

Health insurance — all insurance, really — is little more than a complicated bet, with the individual policyholder in the lugubrious position of betting against himself. The insurance company is betting that most individuals covered by their policies are going to remain relatively healthy and continue paying more in premiums than they claim in benefits. The individual is betting that he will indeed become sick at some point and will receive as much or more in benefits as he has paid in premiums.

Institutions that traditionally pay a part of the individuals’ premiums — companies or governmental entities — are like participants in a crap game in which the insurance company is the shooter. They are betting on the pass line; that is, they, too, are hoping that the insured party doesn’t get sick. Part of that may be compassion for the company’s employees, but a big part of it is financial: When benefits increase too much, premiums have to go up, too.

Insurance companies can do this because they are the equivalent of the house in a gambling establishment. When too many people start hitting the slot machines at a casino, the house tightens up the slots. When a blackjack player is discovered counting cards, he or she is politely but firmly bounced from the premises, despite the fact that the ability to count cards is not cheating at all; it is simply a skill that reduces the houses chances to an unacceptable degree.

The skyrocketing cost of health care is the equivalent of a casino full of card-counting blackjack players, and that’s why health insurance is so high.

The city has decided to be a self-insurer. That is, it will pay the health benefits of its employees and save its share of insurance premiums. But it is hedging that bet by buying something called stop-loss insurance for itself. It is sort of like a bookie laying off bets when he gets extremely lopsided action on a sports event.

For a fee that’s less than a full-coverage health plan, an insurance company agrees to pay a benefit to the policy holder (the city in this case) if benefits rise beyond an agreed-upon point.

You can buy two kinds of stop-loss policies: Specific stop-loss kicks in when an individual’s benefits rise beyond a certain point; aggregate stop-loss pays off when the benefits to the entire employee group get too high.

So far, this looks like a good idea for the city and its employees. The employees will continue to pay into their health care plan as they have in the past, and the city will pay less for the stop-loss insurance than its share of health care insurance coverage would be.

The problem comes if, or when, benefits paid out reach a point at which the house — the stop-loss insurance company — no longer has the edge.

Then the rules will change again, and the city will cast about for new ways to beat the odds, or at least keep them at a survivable level.

That is what health care has become in this country: a game of chance in which both the odds and the rules are constantly changing. It’s a crapshoot, and odds against the individual player are getting longer and longer.