Spot the Setup: What Even Counts as Gambling?

Before we can teach students how to spot risky products, we need a definition of gambling that does not depend on what the product calls itself. A casino game and a $4 trading card pack do not look the same, but they can produce the same kind of risk for the person buying in. We need a single, structural test that works on both.

The test the research literature settles on has three parts. If all three are present in a single activity, the activity belongs somewhere on the gambling spectrum. This framework comes from Mark Griffiths' foundational work on adolescent gambling (Griffiths, 1995) and is the one Drummond and Sauer apply when they argue that loot boxes are "psychologically akin to gambling" (Drummond & Sauer, 2018).

The three conditions

Stake. The person gives up something of value and could lose it. Stake is not just cash. It can be:

• Cash or a gift card balance paid to the operator.
• In-game currency that was bought with cash (Robux, V-Bucks, gems, gold).
• A tradable digital item with real-world market value (a CS2 skin, an NFT, a sports trading card).
• Time spent earning in-game currency that could have been spent another way.

The operative question is, "What does this person walk away with less of if the result goes against them?"

Chance. The result is uncertain in a way the person cannot fully control. Skill can still matter. A daily fantasy sports lineup, for example, is partly skill. But chance is present whenever the person cannot know which card, skin, item, matchup result, or payout they will get. If the person could guarantee the outcome with effort, it is not chance.

Prize. The possible result is something the person actually wants and that has value to them. Prize is not just cash either. Status, a rare cosmetic, a "chase" card, a higher rank, a spot on a leaderboard, or a tradable item can all function as prizes. The prize does not have to be exchangeable for money outside the platform; it just has to be something the person was willing to pay or play for.

Why structural matters more than marketing

The test deliberately ignores legal status, age limits, licensing, and branding. Those are important policy questions, but they answer a different question: should this be regulated. The framework answers a more basic one: is this person, right now, in a structurally gambling-like activity?

This matters because operators have strong incentives to call gambling-like products something else. "Game", "skill", "social", "free", "no cash payout", and "for entertainment only" are common framings. None of them changes whether the product has stake, chance, and prize. As Drummond and Sauer put it, the structural similarity is what produces the risk pattern, not the label.

Two micro-examples to anchor it

Movie ticket. Stake (the ticket price), no chance (you get the movie you bought), no variable prize. Not gambling. The framework cleanly excludes the everyday transaction.

$5 lottery scratch-off. Stake ($5), chance (random outcome the player cannot influence), prize (cash payout the player wants). All three present. Clearly gambling. The framework cleanly includes the obvious case.

The interesting work happens between these two examples. That is where the rest of the lesson lives.

How to phrase this with a class

Two sentences are usually enough to introduce the test:

Then have students try it on something they know, before you use it on something they do not.