Controller Benchmark | Get Your Free Health Score Today
Six weighted stages, stick drift, buttons, triggers, deadzone, latency, and connection stability, combine into one 0 to 1000 Controller Health Score with a letter grade in about two minutes.
Recent runs
No runs saved yet
Six stages run in sequence, each with its own point budget, together totaling 1000.
Stick drift, 5s
Rest both sticks to measure resting noise.
Buttons
Press each prompted button as it appears.
Triggers
Pull L2 and R2 fully, skippable for fight sticks.
Deadzone, 10s
Sticks stay idle while noise is sampled.
Latency, 10s
Move sticks and press buttons continuously.
Connection, 30s
Continuous movement while stability is tracked.
Grades are based on the percentage of attempted stage points earned, scaled to a 1000 point display. A skipped stage is graded against its own reduced maximum.
| Grade | Score range | What it represents |
|---|---|---|
| S | 950 to 1000 | Near perfect across every measured signal, suitable for the most competitive contexts. |
| A+ | 900 to 949 | Excellent condition with only minor variance, suitable for ranked and competitive play. |
| A | 825 to 899 | Healthy controller, one or two stages show early wear signals but plays well across genres. |
| B | 725 to 824 | Functional with noticeable issues, fine for casual play, precision genres may suffer. |
| C | 600 to 724 | Partial reliability, multiple stages show problems worth investigating individually. |
| D | 450 to 599 | Significant degradation, likely stick drift, button issues, or persistent dropouts. |
| F | Under 450 | Failing across multiple signals, replacement usually recommended. |
Each of the six stages produces a continuous sub score within its own point budget rather than a simple pass or fail, so a slightly noisy stick registers as a small deduction and a badly drifting one registers as a large one. That continuous scoring is what makes the composite meaningful, small measurement differences produce small score differences instead of everything clumping into a handful of possible outcomes.
The connection stability stage specifically scales its thresholds to your controller's actual connection type, holding a wired connection to a strict standard while allowing Bluetooth the variance it naturally has from sharing a busy radio band. This keeps the composite fair regardless of how you happen to be connected on test day.
Weighting reflects real world impact. Stick drift carries the most points because it is the single most common complaint and the one most likely to actually affect gameplay, while connection stability carries the least since a stable connection is closer to a binary condition than a graded spectrum.
Any controller your browser recognizes as a standard gamepad can run this benchmark, including these commonly tested models.
Sticks scored poorly
Run a dedicated diagnostic to see exactly what is happening, drift, deadzone, or circularity.
Run the Stick Drift TestButtons or triggers scored poorly
Check for stuck inputs, worn contacts, or a trigger that will not reach full range.
Browse the Fix guidesConnection scored poorly
A weak Bluetooth signal or a failing cable can pull this stage down significantly.
Run the Connection Stability TestWhat is the controller benchmark
The controller benchmark is a single composite test that runs six weighted stages, stick drift, buttons, triggers, deadzone, latency, and connection stability, and combines them into one 0 to 1000 Controller Health Score with a letter grade, giving an overall picture of a controller's condition in about two minutes.
How long does the controller benchmark take
About two minutes for a complete run. Sampling stages run automatically for 5 to 30 seconds each, the button and trigger stages depend on how quickly you complete the prompted inputs, and each stage opens with a short countdown.
What grade should I be aiming for
A grade of A, 825 or higher, is a realistic target for an everyday controller in good condition. A+ at 900 or higher is excellent and typical of a recently calibrated controller with little wear. S at 950 or higher represents near perfect performance across every stage.
Can I skip a stage if my controller lacks the hardware
Yes, for the trigger stage only. Fight sticks and arcade controllers without analog triggers can skip it, and the result is then scored against a reduced maximum rather than the full 1000 points, with the displayed score rescaled proportionally. Every other stage applies to any standard controller.
Why did my score change between two runs on the same controller
Some variance is normal, from measurement noise, how completely you pressed every button or pulled every trigger, and genuine environmental factors like Bluetooth interference. A swing of around 25 points run to run on the same controller in the same conditions is typical, larger swings usually point to inconsistent input or a noisy connection.
Why are the connection stability thresholds different for wired and Bluetooth
A wired connection has a dedicated data path and is held to a strict standard, while Bluetooth shares a busy radio band and naturally produces some jitter and occasional dropouts even on a healthy controller, so its thresholds are set more leniently to reflect that difference fairly.
How does the benchmark compare to running each individual test
The benchmark is faster and gives one overall composite number. Each individual tool goes deeper, showing per axis breakdowns, live visualizations, and detailed threshold tables the benchmark skips to stay quick. Run the benchmark for a fast overall health check, and run the specific individual test when you need to diagnose exactly what is wrong.
Does the benchmark store my results
Yes, locally. Your last 10 benchmark runs are saved in your browser so you can see how your controller trends over time. Nothing is uploaded to a server unless you choose to export a report yourself.
Can I test a PS5 DualSense controller with this benchmark
Yes. The DualSense works over both USB and Bluetooth and can run all six stages of the benchmark.
Can I test an Xbox controller with this benchmark
Yes. Xbox Series X and S, Xbox One, and Xbox Elite controllers can all run the full benchmark over USB, Bluetooth, or a proprietary wireless receiver.
Does this work with a Nintendo Switch Pro Controller
Yes, when the controller is connected and recognized as a standard gamepad by your browser.
Do I need to install anything to run this benchmark
No. This tool runs entirely in your browser using the Gamepad API. There is nothing to download and no account to create.
Is this benchmark the same as running each diagnostic tool separately
It reuses the same measurement techniques as the individual tools but runs shorter sampling windows to stay fast, which is why it produces one composite score rather than the full depth of detail each dedicated tool provides.
Does this work on mobile phones
Yes, on any mobile browser that supports the Gamepad API and a controller paired over Bluetooth or connected through a USB adapter.
Is my benchmark data stored anywhere other than my browser
No. Every measurement happens locally in your browser and only your last 10 runs are saved locally, nothing is sent to a server unless you choose to export a report yourself.