Epic Games Store Rocket League has no aim. It has two deadzones and a broken default
Every other lab here measures where you point. There is nothing to point at in this game — the whole ceiling is two analog sticks, and the game is unusual in putting its deadzones on the settings screen as numbers you can read. So this page does something it cannot do anywhere else: it takes a flick off your controller and tells you which value in your own settings decided what came out — a flip or a double jump, a diagonal or a flat side dodge.
And if you play on keyboard, none of that is your problem — your steering is on or off, so your precision is measured in milliseconds instead. There is an arm of this page for each.
Type your two values in from Settings › Controls, then flick the left stick as if you were dodging. The blue cross is your Deadzone drawn to scale: it is cut per axis, so inside the horizontal band your forward input counts as zero and a flick there goes flat sideways. The orange diamond is your Dodge Deadzone: a flick that ends inside it at the second jump is a double jump, not a flip.
Connect a controller, press a button so the browser can see it, then push the left stick as if you were dodging and let it go.
Everything above this is about a stick, and none of it applies to you. A and D give full lock or nothing, so the only way to make a shallow turn is to hold the key briefly and let go. Your steering precision is your tap-duration precision, and unlike almost everything else players argue about, that is a thing you can actually measure.
Pick a hold length, then tap. Five or six attempts is enough for the numbers to mean something.
If you are 40 ms long every single time, that is a habit and you can simply decide to release earlier — the number moves the same session. If you are 40 ms out in both directions at random, that is the repeatability of your hand, and no amount of deciding changes it. The panel above tells you which one you have, because the advice is opposite.
With no shallow turn available for free, a keyboard player reaches for the handbrake to scrub angle instead — which on the shipped layout is the same key as boost. That default hurts keyboard players specifically and much harder than it hurts anyone on a pad.
A controller player’s equivalent of this skill is holding a partial deflection, which is a completely different motor task with a completely different failure mode. If you switch, expect to be bad again for a fortnight; that is normal and it is not the settings.
In a shooter, stick drift shows up as a view that slowly turns and everyone recognises it. In Rocket League it shows up as a car that pulls to one side, and players blame their own hands for years. Leave the stick alone for a moment and this reads how far from centre it actually sits — then says whether your Deadzone setting is already covering it.
A stick wears out over months, not minutes, so the number below is close to meaningless on its own — a 3% offset is fine on a new pad and the start of the end on a four-year-old one. What is worth having is the same number with two dates on it. Keep this one and the page will compare against it whenever you come back. It stays in this browser and goes nowhere.
Let the stick settle, then keep the reading. Next visit this becomes a comparison.
If the number is already past your deadzone, the method for confirming it properly — and for telling a worn stick apart from a dirty one — is in how to test controller stick drift.
Leave the stick alone
Take your hand off the controller for a second or two and this fills in.
Rocket League applies the steering deadzone to each axis separately, which is why the picture above draws it as a box rather than a circle. A small diagonal push clears a box sooner than a circle of the same size, so tiny diagonal inputs register when tiny straight ones do not.
A bigger deadzone hides drift and takes fine steering with it. If the offset is small enough that your current setting covers it, the honest answer is to leave the setting alone and stop thinking about it.
Most bind advice is taste dressed up as fact. Two entries in this table are not: on keyboard the game puts boost and powerslide on the same key, and on controller it ships the most important air control with no button at all.
| Control | Keyboard default | Controller default | Status | Why it matters |
|---|---|---|---|---|
| Boost / Powerslide | Both on Left Shift | Boost on Circle/B, powerslide on Square/X | Conflict | On keyboard the shipped layout puts two controls you use together on one key. This is a defect in the default, not a preference — one of them has to move before anything else is worth tuning. |
| Air roll left / right | Q and E | Not bound at all | Unbound | Directional air roll is the control that separates players the moment they start playing in the air, and on a controller it ships with no button. Most players put air roll left on L1/LB. |
| Powerslide / free air roll | Left Shift | Square / X | Conflict | Same button, two unrelated jobs, decided by whether the wheels are touching the ground. A habit of holding powerslide through a jump is what makes the car tumble. |
| Jump | Right mouse | Cross / A | Fine | The keyboard default is right mouse; a lot of players move it to Space so the mouse is free for air roll. Either is fine — this one is genuinely preference. |
| Ball cam | Space | Triangle / Y | Fine | A camera, not a control. Leave it on and toggle off only to dribble. |
These are competitive conventions rather than facts, and the page says so. The one exception is camera shake, which is not a preference.
| Setting | Common range | What it trades |
|---|---|---|
| Field of view | 110 | The maximum. Almost universal above bronze; anything lower hides the car in front of you. |
| Distance | 270 – 280 | How far the camera sits behind the car. Further out trades car detail for field awareness. |
| Height | 100 – 110 | Raising it shows more of the pitch and less of the ball at close range. |
| Angle | −3 to −5 | A slight downward tilt. Steeper values help with ground play and hurt aerials. |
| Stiffness | 0.35 – 0.50 | How hard the camera is glued to the car. Lower is smoother and slower to follow. |
| Swivel speed | 5 – 8 | Only affects manual camera look on the right stick. |
| Camera shake | Off | The one entry on this list that is not a preference. |
The controls, this lab, and the pages built for one job each. Every figure is counted from what the page holds.
This page reads the stick against your own settings. These read the hardware underneath it.
It is how far the stick has to be pushed at the moment of your second jump for that jump to become a dodge (a flip) instead of a double jump. Below it you jump; at or above it you flip in the direction the stick points. Raise it if you flip when you meant to double jump; lower it if flips come out as jumps.
Usually because of your steering Deadzone, not your Dodge Deadzone. Rocket League cuts the deadzone per axis, so if the forward part of a mostly sideways flick is under your Deadzone it becomes zero and the dodge goes flat sideways. A forward part under a tenth of the sideways part is also snapped straight. The lab on this page reads the flick off your own controller and tells you which of the two it was and the Deadzone that would have kept the diagonal.
There is no correct value, only a trade. Lower it and light flicks become dodges, but so does a double jump with the stick slightly off centre, which is where accidental flips in the air come from. Raise it and double jumps are safe, but every dodge needs a deliberate push. Most players sit between 0.50 and 0.80. Measure your own flicks here first, then decide which of the two you are losing more often.
Neither: it is applied per axis, which players call a cross deadzone. Each axis is cut on its own, so the dead centre is a square, and along each axis there is a band where the other axis is thrown away. That band is why a flick that is mostly sideways with a little forward comes out as a flat side dodge, and it is why the picture on this page draws the deadzone as a cross.
Yes, and several professionals do, but the trade is real and worth naming: keyboard steering is binary. A controller stick gives you every angle between straight and full lock; A and D give you full lock or nothing. Keyboard players compensate with taps and with air roll on the mouse, which is why their bind layouts look nothing like a controller player’s.
By holding the key briefly and letting go — there is no other way, because the input has no middle. That makes your steering precision the same thing as your tap-duration precision, so it can be measured directly. The keyboard panel on this page gives you a target hold length, times what you actually produced, and separates a consistent bias, which you can correct on purpose, from random scatter, which is the limit of your hand.
Leave the stick alone and read how far from centre it rests. The catch is that a single reading barely means anything: 3% is fine on a new pad and a warning on an old one. Wear is only visible as a change, so this page stores your reading in your browser with the date on it and compares against it the next time you visit. Nothing is uploaded.
On keyboard, move boost or powerslide — the default puts both on Left Shift. On controller, bind air roll left to L1 or LB. Those two are fixes for shipped problems; everything after them is taste.
No. Nothing in the browser can see the game. You type your two deadzone values in from your own settings screen and the page measures your controller against them.
I answer from this site only — the 28 browser tests, the controls library for 46 games, the sensitivity converters and the two games. I can tell you which page to open, how a tool works and what its result means.
I will not guess. If something is not on the site, I will say so.
Answers come from this site's pages.