Mechanical keyboard communities have strong opinions about which switch type is "best for typing speed." The honest answer is more boring than the marketing: switch type matters, but far less than people assume, and it matters in a specific, narrow way.
What actually varies between switch types
The meaningful physical differences between linear, tactile, and clicky switches come down to a small set of measurable properties:
- Actuation force - how much pressure is needed to register a keypress, typically ranging from roughly 35g on light switches to 60g+ on heavy ones.
- Actuation distance - how far the key needs to travel before it registers, which on many switches is shorter than the full travel distance to the bottom of the keystroke.
- Tactile bump (or lack of it) - whether there's a physical feedback point at actuation (tactile/clicky) or a smooth, uninterrupted press (linear).
- Reset distance - how far a key needs to travel back up before it can be pressed again, which matters more for rapid double-taps than for sustained typing.
The actual effect on speed
Lower actuation force and shorter actuation distance can, in principle, reduce the physical work per keystroke, which matters at the margins for very high-volume, sustained typing. But the effect size here is small compared to the variables that actually separate fast and slow typists - technique, chunking ability, and error recovery (the things we cover in our WPM plateau post) dwarf switch-level differences. Controlled comparisons of experienced typists on different switch types tend to show speed differences within normal test-to-test variance, not a consistent, meaningful edge for one switch category over another.
Where switch choice does show a more consistent effect is fatigue over long sessions, not peak speed - lighter actuation force can reduce finger strain during hours of continuous typing, which indirectly helps sustain speed over a full workday rather than boosting your ceiling on a 60-second test.
Where switches genuinely matter more than "typing feel"
- Actuation distance for competitive/gaming use - very short actuation switches genuinely do register keypresses sooner, which has a real (if small) effect in contexts where fractions of a keystroke's travel time matter, like rapid key-repeat scenarios. For sustained prose typing, this effect is much smaller than it is for gaming inputs.
- N-key rollover and anti-ghosting - this is a real, measurable factor, but it's a keyboard/controller feature, not a switch-type feature. A keyboard that drops or misregisters simultaneous keypresses during fast typing will cost you real accuracy regardless of what switches are installed - this matters more than switch choice and is worth checking on any keyboard you're considering.
- Consistency of actuation point - cheap switches with inconsistent actuation force between individual keys (a common issue on low-quality boards) can genuinely hurt accuracy, because your fingers can't build reliable muscle memory for how hard to press.
The honest recommendation
If you're choosing a keyboard for typing speed specifically: prioritize good rollover/anti-ghosting and consistent switch quality across the board over chasing a specific switch's actuation-force spec. Pick a switch feel (linear, tactile, or clicky) based on comfort and personal preference for long sessions, not because a spec sheet claims it's faster - the speed difference between reasonable switch choices is smaller than the difference technique makes, and much smaller than what a bad, low-quality keyboard costs you in accuracy.
A cheap test before you spend on switches
If you're on the fence about upgrading hardware purely for speed, run the same test format at the same time of day on your current keyboard for a few sessions to establish a real baseline, then borrow or test a different switch type before buying. Most people overestimate how much a hardware change will move their number and underestimate how much day-to-day variance (fatigue, warm-up, even the specific passage) already accounts for. If your baseline variance is already ±5 WPM between sessions on the same keyboard, a switch upgrade would need to clear that noise floor before you could even credit it with a real improvement - which, per the research above, it usually won't.