Hand Position and the Home Row

The pad of each fingertip sits on its key, not the flat underside of the finger and not the nail. That means the finger is curled, and the large knuckle at the base of it sits above and slightly behind the key rather than out in front of it. If you hold your hand this way and look at it from the side, the fingers form a shallow arch and there is visible space under the palm.

The reason for the arch is reach, not comfort. A curled finger can extend upward to the top row or fold downward to the bottom row by changing only its own curl, with the knuckle staying put. A flattened finger has already spent its extension, so reaching upward pulls the whole hand forward and reaching downward makes it slide back. That is why a flat-handed typist tends to lose their home position several times a paragraph while a curled one does not: the flat hand has no reserve of movement left in the fingers, so every reach becomes a hand movement instead of a finger movement.

The two raised marks are not there to tell you where F and J are. They are there so you can return to a known position without visual confirmation, at any moment, in the dark, mid-sentence, after a pause — and they work by touch alone, which is the entire point.

Use them as a check rather than a guide. Rest the pads of both index fingers so the bumps sit under them, then let the other three fingers of each hand fall onto the adjacent keys without placing them deliberately. If your ring finger lands cleanly on S or L, your hand geometry is right. If you have to spread or squeeze to reach, your hands are too close together or too far apart, and that misalignment will show up later as a specific class of error: letters typed one column off, consistently, in the same direction.

The habit worth building is touching down on the bumps every time your hands come back to the keyboard, before you type anything at all. It takes a fraction of a second and it removes an entire category of mistake.

The hand should hover. The forearms may rest on the desk edge or on a chair's arms, and the wrists may touch a surface between bursts of typing, but while you are actually typing the wrist is not bearing the hand's weight.

The mechanical reason is that a planted wrist becomes a pivot. With the wrist anchored, moving a finger to a distant key means rotating the whole hand around that anchor, which is a longer and less precise movement than lifting the hand slightly and placing the finger. It also means the fingers on the far side of the arc are dragged out of position while the near ones reach. This is why planted-wrist typists often report that one side of the keyboard feels harder than the other: it is, because the pivot geometry is asymmetric.

A hovering hand costs a little more effort per minute and buys back a great deal of precision. It also removes the need to think about wrist angle at all while typing, which is discussed separately in wrist angle and floating hands.

There are two different things called home position, and mixing them up causes real trouble. The first is the resting position — all eight fingers on their keys, nothing pressed. The second is the return position, which is where a finger goes back to immediately after pressing something else.

The first one is easy and everyone does it. The second is the actual skill, and it is what separates a typist who can sustain a pace from one who slows steadily through a paragraph. After a finger reaches up to type E, it comes back to D. Not roughly back, not eventually back, but back before the next unrelated key is pressed. A finger left hovering over its reach key is a finger that is not available for its own home key, and the next word that needs it starts from the wrong place.

This is exactly what the row-reach lessons train, which is why Top Row: Left Hand spends as much attention on the return as on the reach. It is also the reason the index finger reach drill exists as separate ongoing practice: the index fingers cover two columns each and are therefore the fingers most likely to settle somewhere that is nearly home rather than home.

Hands migrate. Over a long session they tend to creep inward toward the middle of the board, or drift down and forward as the arch flattens with fatigue. You will not see this happening because you are not looking at your hands, and that is the point of the following checks.

The first sign is a run of errors that are all one column in the same direction — typing S for A, or K for J, repeatedly rather than randomly. That pattern is almost never a memory failure and almost always a position failure. The second sign is a specific reach that has started to feel long: if the stretch to P or to the number row suddenly feels further than it did, the hand has usually moved rather than the key.

The fix is not to look down. It is to stop, lift both hands clear of the keyboard entirely, and set them back down on the bumps. Doing this deliberately every few minutes during a long session is cheaper than the errors it prevents, and it keeps the no-look discipline intact rather than trading it away for a glance.

The geometry above assumes a full-size keyboard and an adult hand, and both assumptions fail often enough to be worth naming.

On a laptop the keys have far less travel, so a curled finger bottoms out almost immediately and the arch matters less for reach than for stability. The bigger issue is that the screen sits directly behind the keyboard, which pulls the head down and the shoulders forward, and that changes the forearm angle before the hands are even involved. That trade is covered in typing on a laptop versus a desktop.

On a split or tented keyboard the two hands stop sharing one geometry, which removes the compromise a single board forces on both wrists. Home position is unchanged in principle but has to be found on each half independently, and the bumps become more important rather than less.

Smaller hands, including children's, cannot always reach the number row from home without the palm moving, and pretending otherwise produces a hand that has learned to shift for numbers and then never quite comes back. It is better to accept a deliberate hand movement for the number row at that size than to force a stretch. That, and the rest of what changes when the learner is a child, is in teaching kids to type.

If you have typed for a decade with two fingers, or with eight fingers in approximately the right places, the position above will feel wrong for a while — because your hands have a position already, and it is not this one.

The useful thing to know is that the discomfort is not evidence that this geometry is wrong for you. A hand that has settled into one shape resists another, and the resistance says nothing about which shape reaches further. Expect the first sessions to be worse rather than merely slower — and for the reasons behind that whole trajectory, including why its middle stretch is where most people give up, see the typing learning curve.

What helps most is limiting the change to one thing at a time. Fix the arch first and let finger assignment stay approximate; then fix the return-to-home; then worry about which finger owns which key. Trying to change all three at once produces a hand that is thinking about itself constantly, and a hand that is thinking about itself is not building any habit at all.