Stick Drift Test

The test waits 3 seconds for the sticks to settle, then reads both sticks once per display frame for 10 seconds. That's about 600 readings on a 60 Hz screen. Each reading is an x and y value from -1 to 1, where 1 is full tilt, so a reading of 0.08 shows as 8%.

• Average offset is how far the average resting position sits from center: √(x̄² + ȳ²), where x̄ and ȳ are the average x and y readings.
• Max offset is the largest distance from center in any single reading, √(x² + y²).
• Jitter is the standard deviation of that distance across all readings. High jitter with a low average means the stick flickers around center instead of leaning one way.
• Direction is the way the average position leans, such as up-left.

This test flags drift from the max offset, using its own rule rather than an industry standard. Under 3% is “No drift measured”. From 3% to 8% is “Slight drift”, which a small in-game deadzone usually hides. Over 8% is “Drift”, which is likely to move your character or camera without input. The suggested minimum deadzone is the max offset plus 2 percentage points, rounded up.

Example: the left stick rests at x = -0.04 and y = -0.03 on average. The average offset is √(0.04² + 0.03²) = 0.05, or 5%, leaning up-left. If the largest single reading is 6.2%, the verdict is Slight drift and the suggested minimum deadzone is 9% (6.2 + 2 = 8.2, rounded up).

The test also checks the triggers. If LT/L2 or RT/R2 rests above 3%, the result says the trigger isn't returning to zero. All readings come from your browser, so drivers or controller software that filter the input can make a game read the stick a little differently.

A traditional analog stick has two potentiometers, one for each axis. A small wiper slides over a resistive track as the stick moves, and the controller reads the stick's position from the resistance. When the track wears, or dust and grime get between the wiper and the track, the reading at rest stops matching true center. The game sees that offset as input.

The springs that pull the stick back to center can also loosen over time, so the stick stops returning to the same spot after you let go.

Hall effect and TMR (tunnel magnetoresistance) sticks read position with a magnet and a magnetic sensor instead of a sliding contact. With no contact rubbing on the sensor, they resist the wear behind potentiometer drift. The springs and gimbal still move, so a magnetic stick can still rest slightly off center if the mechanism doesn't return cleanly.

1. Raise the in-game deadzone. Set it to the minimum deadzone from the test, then check that the character and camera stay still with your hands off. This hides drift without repairing it, and a larger deadzone makes small aim movements less responsive.
2. Recalibrate where your platform supports it. On Windows, search for "Set up USB game controllers" and open your controller's properties. Some controllers offer a calibration option there. Steam Input also has calibration settings for supported controllers in Steam's controller settings.
3. Clean around the stick. Turn the controller off or unplug it, blow compressed air into the gap around the base of the stick, and rotate the stick in full circles. A cotton swab with a little isopropyl alcohol can loosen grime in the gap. Let it dry fully before you use the controller, then run the test again to compare.
4. Check warranty and repair options. If the controller is still covered, the manufacturer's repair or replacement service is the safest route. Opening the controller yourself may affect that coverage.
5. Replace the stick module. Replacement modules, including Hall effect versions, are sold for many popular controllers. Some need soldering and some plug in. The DualSense Edge uses stick modules you can swap without tools.

Run the test after each step to see whether the max offset dropped. The gamepad tester has a deadzone slider if you want to see how a given value looks on the live stick.