"Input lag" on a wireless mouse isn't one single delay — it's the sum of several small ones stacked between your hand moving and the cursor moving on screen. Knowing which stage adds the most delay lets you actually target the fix instead of guessing.
Polling rate is how often the mouse sends a position update to the PC, measured in Hertz (Hz). At 125Hz, the mouse reports its position every 8 milliseconds; at 500Hz, every 2ms; at 1000Hz, every 1ms; and newer high-polling mice running at 4000Hz or 8000Hz report every 0.25ms or 0.125ms respectively. A higher polling rate doesn't make the sensor more accurate, it just reduces the maximum time your movement can sit unreported before the next update goes out.
The practical effect diminishes quickly past a point: the jump from 125Hz to 500Hz removes up to 6ms of worst-case reporting delay, which is very noticeable, while the jump from 1000Hz to 4000Hz removes at most 0.75ms, which is at or below what most people can perceive. Polling rate also has a CPU-usage cost, since each report is an interrupt the system has to process, so pushing it far higher than your monitor's refresh rate can actually add overhead without a matching benefit.
Wireless gaming mice generally use one of two connection types, and they are not equivalent for latency. A proprietary 2.4GHz RF connection through a USB dongle uses a low-latency protocol built specifically for continuous, high-frequency polling, and modern implementations achieve response times matching or very close to a wired connection. Bluetooth, by contrast, is a general-purpose protocol with a minimum connection interval around 7.5ms even in its lowest-latency modes, and real-world Bluetooth mice frequently negotiate longer intervals to save battery, adding extra, inconsistent delay.
If a mouse offers both modes, using the 2.4GHz dongle instead of Bluetooth is usually the single largest latency improvement available, often cutting multiple milliseconds of delay versus Bluetooth's connection-interval overhead. This is a protocol-level limitation, not something firmware settings can fully compensate for, so it's the first thing to check before adjusting anything else.
A 2.4GHz dongle mouse shares spectrum with Wi-Fi routers, wireless keyboards, and other 2.4GHz peripherals, and packet collisions force the mouse to retransmit, which shows up as intermittent stutter or dropped movement rather than constant lag. Plugging the dongle into a USB extension cable to bring it within a foot or two of the mouse, and physically separating it from Wi-Fi routers or USB 3.0 devices (which are a known source of 2.4GHz-band noise), reduces retransmissions.
Wireless mice also commonly throttle their polling rate or sensor sampling in low-power/idle states to extend battery life, which reintroduces latency until a movement wakes the sensor back to full rate — checking the manufacturer's power or performance settings for a "high performance" mode addresses this. Separately, every mechanical switch has debounce delay: firmware waits a few milliseconds after detecting a click before confirming it, to filter out the mechanical bounce of the switch's metal contacts, and this delay (typically single-digit milliseconds) is independent of polling rate or wireless protocol entirely.