Your mouse keeps freezing because the input signal is being interrupted somewhere between the sensor and the operating system, most often by power management putting the USB port or wireless radio to sleep, by 2.4 GHz interference, or by a failing cable or receiver connection. A freeze that lasts under a second is usually a power or polling issue; a freeze that requires replugging points to a hardware or driver fault.

That one-paragraph answer covers perhaps 90 percent of cases. The other 10 percent is where most thin explainers stop, and where the actual diagnosis lives. A mouse that stalls for 200 milliseconds during a spreadsheet edit and a mouse that dies for two seconds every few minutes are different failures wearing the same symptom.

How a mouse signal actually travels

Every mouse, wired or wireless, converts physical motion into a stream of reports. An optical sensor images the surface thousands of times per second and computes a delta between frames. The microcontroller packages that delta into a HID report, typically 4 to 8 bytes, and sends it over USB or a radio link.

Wired mice send reports over USB interrupt transfers. The host controller polls the device at a fixed interval, the bInterval, which is what people mean by polling rate. A 1,000 Hz mouse is asked for data every 1 millisecond. A 125 Hz mouse every 8 milliseconds. The mouse cannot push data; it can only answer when polled.

Wireless mice add two hops. The mouse transmits to a receiver or over Bluetooth, and the receiver then behaves like a wired mouse to the host. Each hop adds latency and each hop can fail independently. That is why a wireless freeze can originate in the mouse, the radio environment, the receiver, the USB port, or the driver, and why swapping the mouse sometimes fixes nothing.

The operating system then stacks the report into a cursor movement. If any layer drops or delays a report, the cursor stops. A single dropped report at 1,000 Hz is invisible. A run of dropped reports lasting 100 milliseconds reads as a stutter. A USB port suspended by the power manager reads as a full freeze until something wakes it.

Plain language: the mouse is a polite device that only speaks when spoken to, so anything that interrupts the conversation looks like a frozen cursor.

The time course of a typical freeze

Freeze duration is the single most useful diagnostic clue, because different mechanisms fail on different clocks.

Duration Typical cause What you observe
Under 50 ms Dropped reports, 2.4 GHz retransmission Micro-stutter, barely visible
100–500 ms Polling hiccup, USB bandwidth contention Noticeable skip during fast movement
1–3 seconds USB selective suspend, radio reconnect Cursor stops, then resumes without input
3–10 seconds Driver reset, Bluetooth renegotiation Cursor returns, sometimes with a jump
Until replug Cable fault, receiver crash, port failure Device dead until physically reset
Random, worsening Failing cable or sensor, interference Frequency increases over weeks

The table reconciles the thresholds people argue about online. A "freeze" in forum posts can mean anything from a 30 ms hitch to a dead device, so advice rarely transfers between users. Measure duration before you accept anyone's fix.

Where the stall physically happens

USB power management is the most common culprit on laptops and the most under-diagnosed. Windows enables USB selective suspend by default, and the operating system can suspend a port that appears idle. A mouse that reports no movement for a few seconds looks idle even though it is perfectly functional. When you move it again, the port must resume, and that resume takes time.

The default suspend timeout for USB root hubs is often set to allow suspension after idle periods measured in seconds, and the resume latency depends on the host controller. On some chipsets resume costs 20 milliseconds. On others it costs over a second. This is why the same mouse freezes on one laptop and not another, and why the freeze often happens on the first movement after a pause.

Wireless adds its own power management. Most 2.4 GHz mice drop to a low-power reporting state after a few seconds of stillness. The first movement triggers a wake and a burst of buffered reports. If the wake handshake is slow, you feel a lag before the cursor catches up. Bluetooth Low Energy mice are worse here because connection interval negotiation can stretch the wake to 100 milliseconds or more.

Radio interference is the second physical layer. The 2.4 GHz band is shared with Wi-Fi, Bluetooth, cordless phones, microwave leakage, and USB 3.0 ports. USB 3.0 is a documented noise source near 2.4 GHz, which is why a mouse receiver plugged next to an external SSD can stutter while the same receiver on a front port is fine. Interference does not usually kill the link; it forces retransmissions, and retransmissions produce exactly the 20 to 200 millisecond stutters in the table above.

Cables fail in a characteristic way. The conductors inside a mouse cable flex thousands of times near the strain relief. Strands break progressively, so the cable works at some angles and fails at others. The classic signature is a freeze that resolves when you wiggle the cable or move the mouse to a different position. Cable faults are the leading cause of "works until I move it" freezes.

Sensor and surface problems are less common but real. A mouse on glass, high-gloss, or a patterned surface can lose tracking and report zero movement even though the LED is on. This looks like a freeze but is actually a sensor dropout, and it usually happens only during fast motion or on part of the desk.

What the numbers say about polling, latency, and power

Polling rate matters less than people assume for freeze diagnosis but more than they assume for stutter. At 125 Hz the host asks for data every 8 milliseconds. If two consecutive polls are missed, you lose 16 milliseconds, which is at the edge of perception. At 1,000 Hz two missed polls cost 2 milliseconds, which is invisible. This is why switching a stuttering mouse from 125 Hz to 1,000 Hz often makes small hitches disappear without fixing the underlying cause.

Report rate also interacts with CPU load. A 1,000 Hz mouse generates 1,000 interrupts per second. On a heavily loaded system, especially with USB polling shared across devices, the interrupt can be delayed. Professional caveat: raising polling rate can make a marginal system worse, not better, because it multiplies the interrupt load.

USB bandwidth is rarely the bottleneck for a single mouse, but it becomes one when many devices share a controller. A keyboard, mouse, audio interface, and external drive on the same root hub compete for periodic transfer slots. Each interrupt transfer reserves bandwidth whether or not the device has data. A saturated controller delays mouse polls, and the delay shows up as irregular stutter rather than a clean freeze.

Wireless battery voltage matters at the edges. Most 2.4 GHz mice run on 1.5 V cells or 3.7 V lithium packs. As voltage falls near the cutoff, transmit power drops before the low-battery indicator triggers, and range shrinks. A mouse that freezes only at the far end of the desk may simply be transmitting at reduced power. Fresh cells typically restore range immediately, which is a fast and cheap test.

The bulleted list of red-flag warnings

A freeze that requires unplugging and replugging the device points to a hardware or driver fault, not a settings issue. Freezes that worsen over weeks usually mean a degrading cable, a failing receiver, or a dying sensor. A mouse that freezes only during fast movement is more likely a sensor or surface problem than a USB problem. Freezes that coincide with plugging in a USB 3.0 drive suggest radio interference, not a broken mouse. A cursor that freezes while the mouse LED stays lit and the buttons still click is a report-path failure, not a power failure. Freezes on a desktop but not a laptop, or vice versa, isolate the cause to that machine's USB controller or power settings.

Edge cases that break the usual advice

Laptop touchpad coexistence is a genuine edge case. Some laptops keep the touchpad active while an external mouse is connected, and the two input stacks can conflict during driver enumeration. The result is a brief freeze exactly when the external mouse is detected or re-detected. Disabling the touchpad in device settings often eliminates it.

KVM switches and USB hubs introduce their own enumeration delays. A KVM re-enumerates devices when you switch, which can take seconds and looks like a freeze on the inactive machine. Unpowered hubs split current across ports; a mouse that draws near the limit can brown out during a burst, producing a reset. Powered hubs fix this, and it is the single most common fix for freezes on multi-device desks.

Virtual machines and remote desktop sessions add another layer. The guest OS sees a virtual USB controller, and the host may throttle or batch input reports. Freezes in a VM usually track host CPU load, not the mouse itself.

High-precision mode and raw input settings can also produce apparent freezes. Some games and applications apply smoothing or acceleration that visibly lags the cursor during direction changes. This is software behavior, not hardware failure, and it disappears outside the application.

Practical actions, in order of cost

Start with duration measurement. Note whether the freeze clears on its own, on movement, or only on replug. That single observation eliminates most of the list above.

Then work through these steps, cheapest first.

Disable USB selective suspend for the affected root hub in the power plan advanced settings. This is free and fixes most laptop freezes within one reboot.

Move the receiver or cable to a different USB port, preferably one directly on the machine and away from USB 3.0 ports and external drives. Front-panel ports often share a controller with other devices; a rear port may not.

Replace the battery or charge the mouse fully, then test at normal desk distance. A range test at 3 feet and at 10 feet separates a power problem from an interference problem.

Swap the cable if the mouse is wired and the freeze responds to wiggling. A replacement cable costs USD 10 to USD 15 and is the fastest way to confirm a strain-relief failure.

Test the mouse on a second computer for an hour. If it freezes there too, the fault is in the mouse. If it does not, the fault is in the first machine's port, driver, or power configuration.

Update or roll back the USB host controller driver and the mouse driver, in that order. Driver rollback matters because a recent update is a common freeze trigger, and newer is not always better.

If the device is under two years old and freezes persist across ports and machines, treat it as a warranty case. Sensors and radio modules rarely fail gracefully.

Reconciling the thresholds: a freeze under 50 ms is normal retransmission and needs no action. A freeze of 1 to 3 seconds that clears on movement is almost always power management. A freeze that needs a replug is hardware. Apply fixes in that order rather than replacing hardware first.

What this diagnosis cannot tell you

No amount of desk testing can distinguish a marginal radio module from a marginal antenna without a spectrum analyzer, and almost no home user has one. Interference from a neighbor's network or a building's Wi-Fi density can change week to week, so a mouse that froze last month may be fine now for reasons outside your control.

Firmware bugs in receivers are also effectively opaque. Vendors rarely publish receiver firmware changelogs, and a receiver that crashes every few hours may simply be running buggy code that no setting can fix.

Finally, operating system scheduling is not fully observable from user space. A freeze caused by a delayed interrupt on a busy system may never reproduce under test conditions, and the honest answer is that some stutters are environmental and intermittent. Measure duration, isolate by machine, and replace hardware only after the free steps have failed.