Browse the charts of the most popular and highest-grossing mobile games and you’ll see an odd, half-endearing snapshot of humanity in its downtime — its virtual casinos, its clashing clans, its toy worlds and endless runners, its hunting and plague and pet dress-up simulators, its ceaseless Candy Crush.

If these mobile games reflect something of our dreams and fears, the actual experiences they offer still tend toward the fast, cheap, and out of control. It’s this last part — our tenuous control — that interests me the most. We already sense the time considerations built into the majority of mobile games; we certainly know how inexpensive they are and how companies are adapting with new, often insidious models of free-to-play. We tend to focus less, though, on the basic limits of the touch-screen interface, the traditional game genres it renders unplayable, and the virtual bodies we may have lost along the way.

For example, consider: You cannot play the original Super Mario Bros. on any touch-screen device. Sure, you can emulate it, but you can’t seriously play it. A nearly 30-year-old classic has control requirements — two buttons and a directional pad — too onerous for a touch screen, and any virtual approximation will lack the precision its platforming requires. Fully controlling a simple sprite in two-dimensional space is just no longer a given.

What we have now instead is Rayman Fiesta Run. Like its predecessor, Rayman Jungle Run, and a slew of other auto-runners, its solution to touch-screen platforming is to take away basic movement and leave the player responsible for only the jumps and occasional punches. This makes some sense for a series like Rayman, with its oddly arcing, over-animated leaps and its rambunctious world that prefers madcap theatrics and just-in-time platforming to the reliable footholds of the Mushroom Kingdom. When it also mixes in inventive new mechanics, as in this year’s Legends, it charms; when it adheres solely to its rudimentary brand of punch and bounce, as in 2011’s Origins, it quickly bores.

So it is no surprise that by sticking to its basic Rayman script, Fiesta Run is a deeply unsatisfying platformer. I hesitate to even call it a platformer, because while it certainly looks like one, it plays more like a roller coaster. At most, it could be called a single-solution platformer, which is just as tedious as it sounds. There is even a new power-up (bought with in-game lums or your own dollars, if you run short) that shows you the single correct path through the level. Fiesta Run almost thrills for the first few levels, and then you realize: It has nothing to do with you. You can memorize and react, or not, but the real star of the show is the lush, quivering environment. You’re just passing through, a limbless shadow of an avatar.

Mobile games often do this. Unable to offer a range of precise movements and unite them in a body onscreen, they reduce your choices to some bare essentials and say this: Don’t worry about it — you do your job and we’ll take care of the rest! Of course, your job often consists of just touching the screen at the right time. I am reminded of the extraordinarily prescient WarioWare series, which as far back as 2003 seemed to anticipate the development of bite-size micro-games. By reducing games to mere seconds and treating a simple button press as a hilarious range of verbs — picking a nose, parking a car, pinching a chicken — it both expanded and lampooned our gaming vocabulary, all while making its meta-commentary super fun.

Many have wondered why the WarioWare series has never come to mobile devices. But in some sense, mobile gaming has already become one giant WarioWare meta-game. If mobile games are slightly longer and the humor not always as on point, at least the scene has stayed true to Wario’s true love: money. The most recent hit to crib directly from WarioWare’s playbook is a public safety warning from Metro Trains of Melbourne, Australia, called Dumb Ways to Die. It’s cute, but nothing more than a trifle, with only a handful of micro-games (dodge a bear, plug a wound, don’t launch the nukes) that grow old very quickly.

There is little else to say about Dumb Ways to Die itself except that it, too, feels like a microcosm of mobile gaming (perhaps even more so than WarioWare, since it’s actually a throwaway game). And yet I think its continued success on the charts cannot only be attributed to the Metro campaign’s cute video and catchy song from last year. I see the title there among the others and totally get the anxiety — of all the ways to die, what is more mortifying than a dumb death? But then I also think of games, in which we die all the time for dumb reasons, for small mistakes. Sure, they are really just fail states, but when you have a little guy onscreen, someone to put yourself into, you rightly call that temporary rupture a death.

The spiritual predecessor to Dumb Ways to Die is a beloved indie game that finally came to iOS this fall: Limbo. I have never understood this love one bit. Limbo is a bad joke told over and over again. Did you hear the one about the dead kid? He drowned, he fried, he got decapitated in a bear trap, he got impaled by a giant spider. Limbo has the same design philosophy as the venerable Hill Climb Racing: make really simple things unnecessarily hard to do. Its central mechanics have been dipped in molasses, its moody atmosphere is meaningless, and its dull physics puzzles are connected by no logic other than the exigencies of each isolated moment. It’s torpid, incoherent, and ultimately about nothing (not even limbo).

The best that can be said is that Limbo makes the transition to the mobile space intact, which tells you both everything about its original controls and also everything about the type of avatar a touch screen can comfortably embrace. Your attachment to little Bright Eyes may survive his many gruesome deaths, but his basic running and jumping and grabbing will likely sever the bond. These are always the stakes with bad controls — they make your avatar an unwieldy puppet rather than the locus of your presence onscreen. There are surely better ways to spend $5 than blundering through these motions. Say, a $5 footlong or a two-pack of 5-hour Energy or an overpriced birthday card. Or better yet, if you’re hankering for the humor of dead children, buy a used copy of Edward Gorey’s The Gashlycrumb Tinies. “M is for Maud who was swept out to sea. N is for Neville who died of ennui.”

On those same charts with Rayman Fiesta Run, Dumb Ways to Die , and Limbo, there’s a whole host of popular games we might place in a new genre. Let’s call it “Body Horror.” They have titles like Eye Doctor, Nose Doctor, Monster Nose Doctor, Crazy Dentist, Crazy Shave, Beard Salon, Enchanted Spa Salon, and Baby Spa & Hair Salon. The popularity of these games, whatever else they might indicate, suggests a seemingly irrepressible interest in the human body as well as its representation in virtual space (even if in a gross-cute style). We can’t seem to help wondering how we translate onto the screen. We prep characters for games that don’t exist, so after all the noses are cleaned, the beards trimmed, and the facials complete, what can we do but take a photo and share it on Facebook? There’s nowhere else to go, nothing else to do.

We have, in significant ways, lost our virtual bodies in touch-screen games, at least the ones we’d grown accustomed to moving about with more traditional controllers. The loss is not absolute, of course, but even the tiny birds and robot unicorns and countless runners we temporarily sync with offer more circumscribed avatars than we’re used to, some part of them always set to autopilot. And these are the avatars that actually work, not the fumbling puppets of so many virtual d-pads. When I think about the games most suited to the touch screen, I see the puzzle and strategy and simulation games that have me hovering unseen somewhere above or outside the gamespace. The screen I use for these games offers no tactile feedback or resistance beyond its smooth plane, but it does not need to. My interaction with would-be avatars there is often indirect anyway. I am not the angry bird; I am the one who slings.

When I do make contact with virtual objects and move them in space more directly, the experience is unique to the touch screen, but it does not evoke natural kinship with an avatar. For one, my fingers get in the way. More importantly, my fingers are seen. Contact is direct but deliberate. I am consciously and visually manipulating some body onscreen because my controller and display are one. Being reminded you are controlling a character, like being reminded you are reading a sentence, breaks the spell. It creates a self-consciousness that distances you and severs the link. In so far as you want some tether between yourself and a particular body onscreen, not the entire screen itself, you need to effectively forget your interface. And that’s pretty hard when it’s always right there in front of you.

It’s a simple but significant pleasure to just walk around as an avatar. It binds you to its body and anchors you in virtual space. As this has proven so difficult for most touch-screen games, we are seeing new approaches to embodiment. One of the more interesting has been the simple multiplication of the avatar. If controlling one character onscreen requires too much precision, what about 50 at once? That way you’ll have some to spare. Badland tries this with varying success, and the popular Where’s My Water? plays with the idea too (water is legion). Even the upcoming Super Mario 3D World is experimenting with multiple Marios at once, complicating the singularity of gaming’s most famous avatar.

Still, my most powerful game experiences over the past few years have involved a strongly felt sense of physical space and a single avatar acting within it. Sometimes I just want to be a person in a game. Not a puppet master. Not a benevolent despot. Not an unmoved mover. I want a humble body with an invisible tether. So I can know where I am. And forget how I got there.

Tevis Thompson (@tevisthompson) writes video game essays and fiction. You can read more of his work here.