Luigi’s Picture Poker (LPP) is an itch.io adaptation by Marcy4000, playable on web and on Windows, Linux, and macOS. The minigame originated in Super Mario 64 DS and New Super Mario Bros. The parent games carry PEGI 12 and E (6+) ratings respectively. LPP itself has no rating, and though simple enough for elementary grade children to follow, the gambling adjacency makes 13+ a better floor.
The rules are simple: each round you’re dealt five cards, can swap any number for random replacements, and the stronger hand between you and Luigi takes the round. Hand quality is determined by a hierarchy of combinations and card types, and the winner is awarded coins by the bet multiplied by the hand value. Because the mechanics are stripped down (no folding, calling, or raising, though you can go all in), the design emphasizes stochasticity over strategy. The only choices are which cards to swap and how much to bet. Like a slot machine, LPP is a gambling mechanism deriving great profits from “so small an investment, and with so little effort” (Addiction by Design).
LPP puts players at risk for addiction in the same way slots do: by hiding the odds, paying out unpredictably, and letting the player narrate wins as their own doing. There are only three decisions per round, and the calculations are opaque, since the player isn’t privy to the distribution of card types. Without those proportions, you can’t calculate probabilities; without probabilities, you’re left with intuition, and intuition is exactly where the addiction surface lives.
I played a few rounds alongside two friends, taking turns at the keyboard and talking through hands. What stood out wasn’t the math but the language. When someone won a risky all-in, they said “I just had a feeling” or “I’m finally on a streak.” When they lost the same bet two rounds later, the framing flipped to “of course, that’s my luck.” Luck became a stable trait one minute and a momentary current the next, depending on what the outcome demanded. The main point in Addiction by Design about machines manufacturing a feeling of being “in the zone” landed quickly: even at a free browser game, the table narrated itself into runs and slumps the RNG couldn’t care less about.
The skill story showed up too. One friend kept swapping aggressively, holding only her highest card and redrawing four. When that paid off with a three of a kind, she described it as a read on the game (“you have to commit, the safe swaps never hit”). When it didn’t pay off, the redraw itself was blamed, not her policy. This is the play the player layer in action: in poker proper, reading opponents is real, but in LPP there is no opponent. Luigi flips cards. There is nothing behind them. The skill attribution is pure narrative laid on top of a slot machine, and that narrative is what brings players back.
Compared to other chance-based games, LPP is closer to slots than to poker, and that placement matters because the same mechanic can be near harmless or genuinely predatory depending on what surrounds it. Real poker hides information behind other people, so skill (reading, bluffing, bankroll management) is doing real work even when individual hands are random. Slots hide information behind the machine and offer no skill layer. Gacha and loot box systems go further, burying odds behind opaque rates, layering in daily login pressure and limited time banners, and tying the pull to real money. The assigned Rock Paper Shotgun reading reframes live service games as “the living dead,” arguing they keep extracting from players long after the fun has drained out by making the spending loop the point. LPP doesn’t do that. No store, no battle pass, no FOMO timer, just a closed loop with fake coins, which is what makes it useful as a study object: the slot machine logic is visible because nothing else is layered over it.
Chance in games isn’t morally wrong on its own, but it becomes impermissible when real money, vulnerable audiences, or dark patterns are attached to it. Dice, shuffled decks, and random encounters are part of what makes games feel alive and removing them would flatten huge swathes of board games, roguelikes, and sports. The question is what the chance is attached to. Three things move a chance mechanic from fine to not fine. First, real money: the moment a random outcome costs cash, the design has to be honest about odds and generous about stopping points, and most monetized chance systems aren’t. Second, audience: a mechanic aimed at children carries a different duty of care than one aimed at adults who chose to walk into a casino. Third, dark patterns: hidden rates, near miss animations, and variable reward schedules tuned to maximize session length shift the design from offering chance to exploiting it. LPP passes the first test, partially passes the second (Mario branding pulls a younger audience than the mechanic deserves), and quietly fails parts of the third (opaque card distribution, satisfying win animations, an easy all-in button). It isn’t harmful in the way a gacha game is, but it is a small training ground for the same instincts.
Feature image by Gemini
great read on a bad game