Skip to test
Adaptive20 min read

One-Handed Typing: How to Reach 40 WPM With One Hand

Most advice about typing with one hand starts from the assumption that you are starting over. Learn a new layout, expect months of frustration, accept that you will be slow forever.

The research says something quite different, and it has said so since the early nineties. If you already touch type with two hands, the skill required to type with one is largely already in your head — it is just mapped to the wrong hand. Given a way to unlock it, people beat hunt-and-peck speeds in an afternoon and reach genuinely useful speeds in under ten hours.

This article covers what that research found, the four practical approaches available today, which one fits which situation, and how to start. It is written for a wider audience than the phrase one-handed typing usually implies: people recovering from surgery or a broken arm, people with permanent limb difference or hemiparesis, people managing repetitive strain or pain on one side, and the very large number of people who simply need to type while their other hand is occupied.

The finding that changes the calculation

In the early 1990s Edgar Matias, Scott MacKenzie and Bill Buxton published a series of studies on a technique they called Half-QWERTY. The idea is simple: you type with one hand on a normal keyboard, and when you hold down a modifier — typically the space bar — the other half of the keyboard mirrors onto the keys under your hand.

So the F key, struck normally, gives you F. Held with the space bar, the same physical key gives you J — the key your other index finger would have hit from its home position. The mapping is a mirror image about the center of the keyboard, which means every letter sits under the finger that already knows it. The original paper, Half-QWERTY: a one-handed keyboard facilitating skill transfer from QWERTY, sets out the design; the follow-up Typing with one hand using your two-handed skills reports the training results, and Bill Buxton hosts a copy of the original as well.

That design choice is the whole trick. It is not asking you to learn where letters are. It is asking your working hand to perform movements your other hand already performs, in mirror image, which turns out to be a far smaller ask than learning a layout from scratch.

What the numbers actually were

The follow-up study, One-handed touch-typing on a QWERTY keyboard, ran ten right-handed touch typists through ten practice sessions, no more than one per day, with three subjects continuing to extended testing. Their mean two-handed speed was 58 words per minute.

One-handed speed, from the Half-QWERTY studies
3–4 hrs
to beat hunt-and-peck speeds

Faster than most people expect

8–9 hrs
to pass 50% of two-handed speed

About a weekend of practice

23.8–42.8
WPM after ten sessions

41 to 73 percent of two-handed speed

~60
WPM by one subject with extended practice

Around 68% of two-handed

Matias, MacKenzie and Buxton. Ten right-handed touch typists with a mean two-handed speed of 58 WPM, across ten sessions, with three subjects continuing to extended testing.

Three things in that table deserve emphasis. The first is how early the crossover comes: three to four hours to beat hunt-and-peck. If you have broken your wrist and are facing six weeks in a cast, that is an afternoon against six weeks of pecking.

The second is the ceiling, which is genuinely high. Forty-two words per minute is above the average typing speed for adults using two hands, a distribution laid out in what counts as an average typing speed — and comfortably clears the requirement in most job postings that state one, as typing speed for jobs sets out. One hand, ten sessions, and you are typing faster than the average two-handed typist.

The third is the spread. Forty-one to seventy-three percent is a wide band across ten people, which is normal for motor skill research and worth knowing before you measure yourself against a single number. Where you land depends on your starting speed, how automatic your two-handed typing already was, and how much you practise.

The shape of the learning curve, as a share of two-handed speed
02040608012469102040
Half your two-handed speed

Practice hours

hr 1: 12 %, hr 2: 20 %, hr 4: 31 %, hr 6: 41 %, hr 9: 52 %, hr 10: 55 %, hr 20: 62 %, hr 40: 68 %.

Indicative curve reconstructed from the reported milestones — hunt-and-peck surpassed at 3 to 4 hours, 50 percent at 8 to 9 hours, 41 to 73 percent at session 10, and roughly 68 percent by session 40 for extended subjects. Points between milestones are interpolated, not measured.

Notice the shape. It is steep early and flattens hard, which is the standard form of a motor learning curve and the same pattern that shows up in ordinary two-handed practice — the reasons are covered in why typing speed stops improving. Almost all of the practical benefit arrives in the first ten hours. Chasing the last few percent takes far longer than getting to usable.

Why mirroring works: you already have the skill

The mechanism here is worth understanding, because it explains which approaches are fast to learn and which are not.

Touch typing is a motor sequence skill. Your hands do not look up where the letter T lives; they execute a rehearsed movement, and skilled typists overlap those movements rather than making them one at a time — the pattern the Aalto study of 136 million keystrokes found most predictive of speed. Those movements are stored as patterns, and the patterns are specific — which is why motor skill transfers only partially between related tasks, and why practicing prose does very little for typing code or numbers.

Mirroring is unusual because it lands on the one kind of transfer that does work well. The movement your right index finger makes for J is the mirror of the movement your left index finger makes for F, and the motor system handles mirror-image transformations comparatively gracefully. So the mirrored mapping is not asking for new learning so much as a re-pointing of learning you already have.

That also predicts, correctly, who benefits most. If you already touch type, this is fast. If you have never touch typed — if you hunt and peck with two hands — there is no stored skill to mirror, and you would be better served learning a purpose-built one-handed layout or, honestly, learning the ten-finger method first if both hands are available to you.

The four approaches, compared

There are four realistic ways to type with one hand, and they suit genuinely different situations. The most common mistake is picking on the basis of what sounds most serious rather than what matches your circumstances.

One-handed typing approaches
One-handed typing approaches
ApproachLearning costSpeed ceilingBest for
Plain QWERTY, one handNoneLow — long reachesVery short-term, or occasional use
Mirrored / half-QWERTY3–10 hoursHigh — 40+ WPM documentedExisting touch typists, any duration
One-handed DvorakWeeksHighPermanent change, or no prior touch typing
Chorded keyboardWeeks to monthsHigh, with specialist hardwarePermanent, committed users

Learning times for the mirrored approach are from the Half-QWERTY studies; the others are described qualitatively because comparable controlled data is not available for them.

The row people default to is the first one, and it is the worst of the four for anything beyond a few days. Typing QWERTY with one hand means one hand covering a layout designed for two, so the reaches are long, the home position is meaningless, and you end up looking at the keyboard — which caps you around the same place hunt-and-peck caps everyone, for the reasons catalogued in the mistakes that quietly cap your speed.

One-handed Dvorak, and how to turn it on

August Dvorak's name is usually attached to the two-handed layout, but one-handed left and right variants have existed for decades, and they take the opposite design approach to mirroring: rather than reusing your two-handed skill, they rearrange the alphabet so that one hand can reach everything with minimal travel from a central resting position.

The practical surprise is that you do not need to buy or install anything to try it. Apple's own documentation for using Dvorak keyboard layouts on Mac lists Dvorak, Dvorak - Left-Handed, Dvorak - Right-Handed and Dvorak - QWERTY as options you add under System Settings, then Keyboard, then Text Input, then Edit. Windows ships equivalents. Background on the layout family generally is on the Dvorak layout overview.

The honest caveat is the same one that applies to layout switching in general, and this site has been consistently sceptical about it. Learning a new layout costs weeks of reduced speed, every keyboard shortcut you have memorized moves, and the evidence that alternative layouts beat QWERTY on speed for two-handed typists is weak — the full arithmetic is in the honest math on Dvorak, Colemak and mechanical keyboards.

What makes the one-handed case different is that the comparison is not against your two-handed QWERTY speed. It is against typing QWERTY with one hand, which is genuinely bad. Against that baseline a purpose-built layout has a real advantage, and if the change is permanent the weeks of learning amortize over years.

Getting a mirrored layout today

Half-QWERTY was commercialized in the nineties and the dedicated hardware is long gone, but the technique is a software mapping and several implementations exist. The general approach is described under one-hand typing, and community and charity resources such as AbilityNet's factsheet on single-handed keyboard use list current options for both Windows and macOS.

What you are looking for is a remapping tool that supports a held modifier producing a mirrored layer. On Windows, general-purpose remapping software handles this; on macOS, keyboard customization tools do the same. The important feature is that the mirror is on a held key rather than a toggle, because holding is what preserves the rhythm of typing rather than turning every hand-switch into a mode change.

One practical note from the research design worth carrying over: the space bar is the conventional modifier because your thumb is free and it is the one key you can hold without disturbing your finger positions. If you are configuring this yourself, resist the temptation to put the mirror on a key your fingers need.

The modifier problem, and the setting that solves it

Typing letters one-handed is the part people think about. Modifier combinations are the part that actually causes daily friction, and it has a simple fix that a surprising number of people never turn on.

Sticky Keys is built into Windows, macOS and Linux. It changes modifier behavior so that pressing Shift and then a letter produces a capital, rather than requiring both at once. Every two-key combination becomes two sequential single presses, which makes Shift, Control, Alt and Command combinations trivially reachable with one hand.

This matters more than it sounds. Capitals and punctuation are roughly a fifth of ordinary writing, and shifted characters are a far larger share of anything technical — measured at around one character in eight in real source code. Without Sticky Keys, every one of those is a contortion. With it, they are ordinary keystrokes.

It sits alongside the rest of the built-in computer accessibility features every major operating system ships and most users never open: filter keys for tremor, key repeat delays, on-screen keyboards, and dwell control. The broader landscape of assistive technology extends well beyond that, and the W3C's Web Accessibility Initiative is the reference point for how software is supposed to accommodate all of it.

For characters that are not on any key — accented letters, symbols, currency marks — the platform methods are collected in how to type special characters, and most of them are one-handed friendly once Sticky Keys is on.

Chorded keyboards and specialist hardware

At the committed end sits dedicated hardware. Chorded keyboards produce characters from combinations of a small number of keys pressed together, so one hand can address a full character set without moving. Purpose-built one-handed boards exist too, including the long-running Maltron single-handed models. Specialist boards also tend to drop the keys nobody presses, which is a running theme in the history of the keys nobody uses.

These are real options with real users, and they carry a real cost: specialist hardware, a learning curve measured in weeks to months, and a skill that does not transfer to any other keyboard you encounter. That last point matters more than people expect, because it means every borrowed laptop and every public terminal becomes a problem.

Chording is also the mechanism behind the fastest text input any human performs. Stenographers reach 300 words per minute by pressing whole syllables at once rather than letters in sequence, which is worked through in how stenographers type 300 WPM. That is an existence proof that chording works, not a suggestion that it is an easy path.

Where dictation genuinely helps, and where it does not

Speech recognition is the obvious suggestion, and it deserves a more careful answer than either the enthusiasts or the sceptics give it.

Modern speech recognition is genuinely good, and for composing continuous prose that you have already thought through it is faster than one-handed typing by a wide margin. If your work is drafting documents or replying to email, it will carry a substantial share of the load.

It falls down in specific and predictable places. Editing is the big one — moving a cursor, fixing a word in the middle of a sentence, and restructuring a paragraph are all clumsy by voice. Names, technical terms, passwords, code and anything with unusual spelling are error-prone. And you cannot dictate in a shared office, on a train, or anywhere the content is private. The general case, including the study that measured speech at three times typing speed on mobile, is in why dictation never replaced the keyboard.

The sensible position is that dictation and one-handed typing are complements. Dictate the prose, type the edits, and stop trying to make either one do the whole job.

Choosing, based on your actual situation

The right approach depends far more on duration and starting point than on anything else.

Temporary against permanent

Temporary (weeks to months)

Best first move
Sticky Keys, then mirrored layout
Layout change
Mirrored — reuses existing skill
Hardware
Not worth buying
Dictation
Useful stopgap
Time to usable
3–4 hours
Who to involve
Nobody, usually

Permanent or long-term

Best first move
Sticky Keys, then evaluate both
Layout change
One-handed Dvorak may pay back
Hardware
Worth evaluating properly
Dictation
Useful permanent complement
Time to usable
Weeks, for a higher ceiling
Who to involve
An occupational therapist

Practical guidance rather than research findings. The learning-time figures for the mirrored approach are from the Half-QWERTY studies; the rest reflects the trade-offs described above.

That last row is the one I want to be careful about. This article is about typing technique, not about your hands. If your situation involves pain, weakness, numbness, a neurological condition or recovery from injury, an occupational therapist can assess what you should and should not be doing in a way that no article can, and workplace accommodation processes generally expect that input anyway.

Nothing here is medical advice, and the general point that popular belief about keyboards and injury has outrun the evidence — in both directions — is set out in what the research on typing and carpal tunnel actually shows. If typing hurts, the answer is an assessment, not a layout.

A plan for the first ten hours

This assumes you already touch type and are using a mirrored layout, which is the situation the research covers directly.

Ten hours, spread over ten sessions
  1. 01

    Sessions 1–2 — find the mirror, not the speed

    Type slowly and deliberately, focusing entirely on the modifier press and release. The goal is that holding the space bar and striking a key stops being two decisions and becomes one. Expect to be slower than pecking. You will pass that point in the next session or two.

  2. 02

    Sessions 3–4 — the crossover

    This is where the research subjects passed hunt-and-peck speeds. Type ordinary prose rather than drills, because the mirroring benefit comes from real letter sequences your hand already knows. Resist looking at the keyboard, which is what breaks the transfer you are relying on.

  3. 03

    Sessions 5–8 — build the rhythm

    Work below your maximum on continuous text. The thing you are training now is the transition between mirrored and unmirrored keys mid-word, which is where all the hesitation lives. Words that alternate hands in two-handed typing are the hard ones — and English is full of them, which is the point.

  4. 04

    Sessions 9–10 — measure and consolidate

    Take a timed run of three minutes or longer and record it. Expect somewhere around half your two-handed speed. Then keep using it for real work rather than practice, because the remaining gains come from use rather than drills.

Session structure follows the research design — no more than one session per day, which is also what motor consolidation research would predict is optimal.

One session per day matters and is not an arbitrary detail. Motor learning consolidates between sessions rather than during them, so ten sessions across ten days beats ten hours across a weekend. The general principle, and why frequency beats duration, is in a practice routine that actually moves the number.

What to expect, honestly

A few expectations worth setting, because over-promising here would be easy and unhelpful.

  • You will not reach your two-handed speed. The best documented outcome across extended practice was around 68 percent, and half is a more typical landing point. Half of a good two-handed speed is still a perfectly workable typing speed.
  • Your accuracy will suffer before it recovers, and net speed is what counts — the arithmetic is in what WPM, accuracy and consistency actually measure. Slow down until the errors stop rather than pushing through them, and expect a further dip if anyone is watching you learn — an effect with its own mechanism.
  • Numbers and symbols are disproportionately awkward one-handed, and they are the characters ordinary practice never drills anyway. Give them their own sessions.
  • Fatigue is real. One hand is doing the work of two, and the sensible response is shorter sessions with breaks rather than pushing through discomfort.
  • If you are typing one-handed by choice rather than necessity — holding a child, eating, on a call — you do not need any of the layouts. Sticky Keys and a bit of tolerance for slowness will cover it.

Measuring yourself

If you want a number, take it properly. A timed test at three minutes or longer gives a figure worth trusting, where a thirty-second run flatters everyone and flatters a new skill most of all. Record accuracy alongside speed, and the tutorial explains what each figure means.

Compare against your own previous one-handed results rather than against general benchmarks, which assume two hands. And do not read too much into a single run — day-to-day variation is larger than most people credit and larger still while a skill is new, for reasons covered in why the same typist scores differently from one day to the next.

There is no widely used one-handed typing benchmark, and I would treat any site claiming a definitive one with suspicion — confident numbers with nothing behind them are endemic in this subject, as the typing myths that refuse to die documents. The Half-QWERTY figures on this page are the best-documented numbers I could find, they come from ten subjects in a laboratory, and they are a reasonable expectation rather than a standard. In keeping with our editorial standards, where the evidence is thin I would rather say so than invent a benchmark.

The short version

If you touch type and you need to type with one hand, you are not starting from zero. The movements are already in your head, mapped to the hand you are not using, and a mirrored layout re-points them at the hand you are.

Turn on Sticky Keys first, because it takes a minute and removes the largest daily friction. Set up a mirrored layer second. Expect to beat hunt-and-peck within an afternoon and to reach around half your two-handed speed within ten sessions, which for most people means somewhere in the thirties — above the average two-handed typist.

And if the situation is permanent, take the time to evaluate a purpose-built layout properly, involve an occupational therapist, and treat dictation as a genuine part of the toolkit rather than a consolation prize. The gap between typing badly with one hand and typing well with one hand is about ten hours of work, which is a remarkably small price for something you may use every day for years.

Frequently asked questions

How fast can you type with one hand?

Faster than most people expect. In the Half-QWERTY research, touch typists with a mean two-handed speed of 58 words per minute reached between 23.8 and 42.8 words per minute one-handed after ten practice sessions — 41 to 73 percent of their two-handed speed. With extended practice, individual subjects reached around 60 words per minute.

How long does it take to learn one-handed typing?

If you already touch type, less time than you would think. The same research found subjects exceeded hunt-and-peck speeds after roughly three to four hours of practice and passed 50 percent of their two-handed speed after about eight to nine hours. That is a weekend of focused work to reach a usable speed, not a months-long project.

What is the best keyboard layout for one hand?

It depends on whether you already touch type. If you do, a mirrored or half-QWERTY approach is usually fastest to learn because it reuses the motor memory in your working hand. If you do not, or if the change is permanent, a one-handed Dvorak layout is worth considering because it was designed for the task rather than adapted to it.

Does macOS have a one-handed keyboard layout built in?

Yes. Apple's documentation lists Dvorak, Dvorak - Left-Handed, Dvorak - Right-Handed and Dvorak - QWERTY as options, added under System Settings, then Keyboard, then Text Input, then Edit. No extra software is needed to try it.

How do I press Shift or Ctrl shortcuts with one hand?

Turn on Sticky Keys, which is built into Windows, macOS and Linux. It lets you press a modifier and then the letter in sequence rather than holding both at once, which converts every two-key combination into two single presses. It is the single highest-value setting change for one-handed use and takes about a minute to enable.

Should I just use dictation instead?

Use both. Dictation is genuinely good for composing prose you have already thought through and poor at editing, code, passwords, names and anything you cannot say aloud in your current room. Treating it as one tool among several works considerably better than treating it as a replacement for the keyboard.

Is one-handed typing worth learning for a temporary injury?

If you expect to be one-handed for more than a couple of weeks, almost certainly. Three to four hours of practice to beat hunt-and-peck is a small investment against several weeks of frustration, and the mirrored approach means you are borrowing an existing skill rather than building a new one from scratch.

Can you touch type properly with one hand?

Yes, in the meaningful sense: your fingers can find keys by feel with your eyes on the screen. That is what the research subjects were doing, and it is why their speeds climbed so quickly. What you cannot do is reproduce two-handed speed, because half the available fingers are doing all of the work.

Put it into practice

Take a quick typing test and see your WPM, accuracy and consistency.

Start a test