The Tower Stacker timing guide

Written and play-tested by the Super Massive Games team · Updated August 2026

A stacking game looks like the simplest thing on the site. A block slides across the screen, you press once, it lands. Anything hanging over the edge of the floor below gets cut away, and the next block is the width of what survived. There is one input and one decision, repeated. And yet nearly everyone hits a wall at a similar height, gets frustrated in the same way, and blames their reflexes.

Reflexes are rarely the problem. What limits a stacking run is a specific, learnable relationship between how wrong your average drop is and how much room you have left. Once you can see that relationship you stop playing floor by floor and start playing the whole tower, which is a very different experience. Everything below applies directly to Tower Stacker, and to any timing game where a mistake shrinks the target for your next attempt.

Your average error is the score. Your worst error is just the ending.

Ask a player why their run ended and they will describe the last drop: "I was way too late on that one." True, but not the explanation. By the time a single bad press ends a run, the platform was already narrow enough for one bad press to end it, and that narrowness was built out of the previous twenty drops.

Think of it as a budget. You start with a full-width platform, and every drop spends some of it. A drop that is a hair off spends a hair. A drop that is well off spends a lot. The run continues while there is budget left, and the size of your budget at any floor is the starting width minus the sum of every mistake you have made so far. That sum is your average error multiplied by the number of floors — which is exactly why average is the number that decides how high you get.

Two players make this concrete. One is streaky: mostly perfect, with an ugly drop every fifth floor. The other is unspectacular: never perfect, always a little off, but always the same little. Watch them both and you would say the first one is better. Count the width they have left at floor twenty and the second one is usually well ahead, because a single ugly drop can spend more platform than ten small ones.

This changes what you should be trying to do while playing. The instinct after a good drop is to try for another perfect one, and the instinct after a bad drop is to press harder and faster to make up for it. Both instincts increase variance, and variance is the thing that kills towers. The target is boring repeatability: press the same way every time, accept that "same" includes being slightly off, and let the consistency do the work.

Look at the edge below, not at the block in the air

Here is the single change that improves most players' runs immediately, and it is about where your eyes go rather than what your hand does.

The natural thing is to watch the travelling block. It's the moving object, it's the thing you're controlling, and your eyes are built to follow motion. The problem is that watching it puts your gaze in motion too, and a moving eye is a poor judge of position. You end up trying to answer "is this block above the right place yet" while the reference point and the target are both sliding through your visual field.

Instead, park your eyes on the near edge of the block below — the one the travelling block is approaching. It does not move. It is a fixed line on a still part of the screen. Now the question is no longer "where is the block" but "has the block's edge reached this line yet", which is a much simpler thing for your visual system to answer, and you can answer it with your eyes still.

A few practical notes on making this work:

• Pick the leading edge, always the same one. If the block is travelling right, the edge that matters is its left edge meeting the left edge of the platform. Choosing whichever edge looks convenient on a given floor means you are effectively learning two timings.
• Do not follow the block back. When it reverses direction, your eyes want to go with it. Let them stay on the line and let the block leave and return.
• Use peripheral vision deliberately. You can see the approaching block perfectly well without looking at it. Peripheral vision is bad at detail and good at motion, which is precisely the division of labour you want here.

The first two or three runs after switching will feel worse, because you are overriding a habit. Give it a full session before deciding.

Press early, because you are always late

Every press is late. Not because you're slow, but because the chain from "the edges line up on screen" to "the game registers your input" takes real time: the frame has to be shown by your display, your eye and hand have to do their part, and the key or tap has to be read. That total is small, but the block keeps moving during all of it, so a press that felt perfectly timed lands slightly past the mark.

The fix is not to try harder. It is to aim at a point slightly before the alignment and let the delay carry you onto it. Most players find this within a handful of floors once they know to look for it: you pick a spot a little short of the line, press when the edge reaches that, and the block lands where you actually wanted it.

Two warnings. First, the amount of lead depends on your screen and your input device, so a friend's advice about exactly how early to press is worth nothing on your hardware — find yours. Second, the lead has to scale with speed. The faster the block travels, the more distance it covers during that same fixed delay, so the same amount of "early" in time becomes a bigger error in pixels as the tower climbs.

Small trims compound, and that is the whole game

The rule that a drop's overhang is cut away is not just a penalty. It is a feedback loop, and it is the reason stacking runs end the way they do.

Each trim narrows the platform. A narrower platform means that on the next floor, the same timing error in milliseconds produces a bigger error relative to the space you have. So the difficulty of floor twelve is not set by the game; it is set by how well you played floors one through eleven. Two players on the same floor number can be facing completely different problems.

That has a couple of consequences worth internalising:

• Early floors are the ones that matter most. They are the easiest to get right and they set the width you carry for the rest of the run. Sloppiness at the bottom because "it's still wide" is the most common way to cap a run at a mediocre height.
• A run that survives a big trim is not saved. Surviving a bad drop feels like relief, but you are now playing a harder game than you were a second ago. That is the moment to slow your breathing and get deliberately conservative, not the moment to relax.
• Recovery is impossible in one direction. Width only ever goes down. There is no drop that gives you space back, which is why every decision in a stacking game is about spending as little as possible rather than gaining anything.

Acceleration changes the task, not just the difficulty

As the tower rises, the block travels faster. It is tempting to file that under "gets harder", but something more interesting happens: at a certain speed the game stops being a reaction task and becomes a rhythm task, and the technique that got you there stops working.

At low speed, there is enough time between seeing the alignment and needing to press that you can make a decision. You watch, you judge, you commit. At high speed, the block crosses the useful window faster than a considered decision takes, so watching-then-deciding is guaranteed to be too late. What works instead is pre-commitment: you settle into the block's back-and-forth cycle, feel where the beat is, and put your press on the beat rather than on the sight.

Players who make this transition describe it as ignoring the screen slightly — still looking at the reference edge, but pressing to an internal count rather than to a visual event. Players who never make it describe the fast floors as "impossible", which they are, if you insist on reacting to them.

Practically, the handover comes gradually. There is a middle range where you are half-reacting and half-rhythmic, and that range is where most runs die because neither method is being used properly. If you can notice that band on your own tower and consciously switch to counting before you reach it, you will pass through it much more often.

A short practice plan

1. Five runs, eyes on the edge below, no score goal. The only success condition is that you never once tracked the moving block. Expect the heights to be bad. Ignore them.
2. Five runs, lead calibration. Deliberately press a touch early on every floor and notice which direction you are landing off. Adjust once, not every floor.
3. Five runs, bottom-heavy discipline. Treat the first eight floors as the whole exercise; if any of them trims visibly, restart immediately rather than continuing.
4. Five runs for height. Now play normally and see where the ceiling has moved to.

Between runs, resist the temptation to change anything mid-session. As with any timing skill, a change you make three floors into a run gets evaluated against three floors of data, which is not enough to evaluate anything.

Questions players ask