Piano World Home Page Forums Our Most Popular Forums Piano Teachers Forum The correct way of pressing a key ?

The physics part is a bit difficult to understand; worse still, there is no conclusion.

But I know if I were to use my fingertips, I'll have to trim my nails. I can't do simple household chores with my nails at reasonable length. Even picking up a piece of paper would be difficult.

If you note the shape of your hand as your arm hangs loosely at your side, it forms a natural, gentle curve. What I’ve read and my own experience tell me that playing the keys with this natural curve enables me to play with the least amount of tension. I think that’s why the B major and E major scales are easier to play than C major, because they allow for this relaxed, natural curve. Chopin always started his students with the B major scale rather than C, no doubt for this reason.

The shape of your hand is going to have to change depending on what you’re playing, of course. But I think trying to retain and use the natural curve, which will have you playing more on the pads than on the tips, is probably a good idea.

Beginners tend to play in one of three ways:
1. with the finger more or less rigid from the knuckle
2. with the finger forming a straight line out from the knuckle to the first joint and then the second joint as straight (or beyond straight!) as much as possible (with the fingertip section almost bending back at up to a 20 degree angle)
3. with the fingers falling in a naturally curved position

method 3 will give the best results by a country mile

In an effort to get some beginners to create this third position teachers encourage students to think about playing on the tips of their fingers. In actual fact, once one has progressed very far at all one would be hindered by an adherence to playing specifically and uniquely on the fingertips. But if one starts from playing with the entire length of that third joint one is limited in one's ability to modulate the touch (let alone working very hard to press down each key).

Just as many students try to play with method 2 and with method 1, and maybe to a teacher who was purely concerned with playing on the finger tip this would be no problem, method 2 is at least as bad as method 1 in terms of restricting tonal control, and worse in terms of creating tension.

The more a student is inclined to play with method 1 (above) the less flexibility in tone will be possible. And in my experience students who practice on instruments with less resistance tend to develop a method 1 style of playing. Method 2 is typical of tense students (of course, this is not a one-way causal street - students who play this way will, by definition, experience tension - of the colloquial bad kind).

At the end of the day the shape of the hand is key to the production of tone, not the placement of the finger tip. But again, the injunction given by many teachers that students play on the tips of their fingers is a short-hand way of seeking to develop method 3 as compared to method 1. And playing with the entire length of the joint depressing the key is rarely of any positive value at all when in the first stages of learning.

You asked the correct way of pressing a key, and I guess we're remiss in not telling you that the best way is to press it down. When you press side to side or upwards, you don't get any sound.

All kidding aside, the focus on the finger tip avoids the entire issue which is that we play piano with a whole body mechanism, consisting of torso, arms, wrists, hands and fingers, and legs for balance and pedaling. Your teacher should be working with you on each of these elements, both separately and as a whole.

How can playing on the tips lead to a balanced natural curve? You most easily learn the natural curve by sensing what the return force of the key does to the neutral starting point. Then feeling how a tiny amount of added squeezing from each joint restores the equilibrium that was taken away as the key pushed back and caused that natural position to deform. You need to perceive what that force does and then perceive how to cancel it out. How can coiling the finger substantially past this point of maximum balance help to achieve it? I don't think that's the most direct path. When working with beginners I use a lot of different exercises. However, for collapsing tips I always get them to start with a completely flat finger that contacts a depressed key along its whole length. When they squeeze inward very slowly, I've rarely encountered anyone who wasn't able to get the finger into a naturally rounded position. Sometimes they need to work on only squeezing in with the end joint first, but almost everyone goes on to find how to balance without collapsing, when using this approach. I get them to gradually go further and further until they can feel that any additional curling is serving to weaken the contact and balance, rather than add anything positive. Just about everyone perceives this point without having come anywhere remotely near to the very tip of the finger. It's not surprising because any action from the final joint works increasingly across the key rather than into it, the further you continue.


In general I'd actually go so far as to say that those who I have seen attempting to play on the very tips of their fingers tend to suffer a lot more collapses than those who remember how to just lightly squeeze an initially flattish finger into balance. Another aspect here is that a collapsed flatter finger is easily corrected. Simply feel that inward action from the last joint onwards up to the knuckle. That small inward motion is a very simple fix. Students always seem to find it harder to correct collapses that occur when they have very curled fingers. You basically have to reset and start again. Far easier to fall short of the right muscular activity and then add a little more- than to go way beyond the right level and have to start looking to go backwards, I think. I's say that this would be a big danger with seeking to start with curled up fingers but then hope to return to a smaller curve. It doesn't give you any sensitivity to the right level of activity or teach you how to feel what you need to counter the return force of the key. So if you get accustomed to this, you're basically hoping to leap backwards without knowing exactly how to get to your destination or precisely where it even lies. If you learn what the muscular activity is specifically for, you get a lot of feedback along a very direct route.

PS. Playing with a totally flat finger does not either cause or necessarily imply great tension. I play this way for specific purposes with substantially less effort and tension than when I coil a finger. In the case of of inexperienced pianists, it may suggest that they are using opposing muscles to force the finger straight. That is not good. However, when the finger naturally straightens out as the product of action that is all at the knuckle, there is no cause to speak of notable or undesirable tensions.

Of course it does. It is not the the natural position of the fingers. If you let your arms dangle by your sides in as relaxed a manner as possible, the fingers form a natural curvature. Any attempt to straighten them beyond this point induces stiffness, not only in the fingers themselves, but also in the hand.

Surely, playing in C major is also very cumbersome if the fingers are opened beyond the natural curvature, as the thumbs are forced away from the keyboard? Or are you advocating a high knuckle position, which seems to be the only way to keep the fingers relatively straight, in front of the black keys, and with the thumbs on the keys?

The logic does not follow. When you put a force on something either it moves or something responds to cancel that force. So if a joint collapses (in the procedure of action from the knuckles- not due to active straightening forces within the joints themselves,as I specified) that is due to LACK of muscle tension. Conversely, when a finger keeps a curved shape while the key exerts a force against it, muscular tensions are precisely what cancel out that force and keep it from deforming in response to that force. Flat fingers can be associated with poor muscular use and extreme tensions. However, they do not require them to occur. It's all down to what else is going on and how the forces add up.

This can be done either with near vertical fingers on near horizontal ones, by the way. Either is possible without much effort and both have their uses.

Try putting a totally limp hand on the keys and slowly let your wrist drop. Even without going far enough to depress any keys at all, the return force of the keys should straighten your fingers. If it does not, they are not in a relaxed state. Fingers can straighten without an ounce of tension. Of course, there's some tension in playing with a flat finger, no matter how well you organise it. But the same applies with any motion. Why a flat finger should necessarily involve any worse or more uncomfortable ones than anything else, I don't see any reason. If you do it the right way, you could put up a highly justifiable argument that it serves to minimise tensions- scarcely requiring any except in the strong, capable lumbricals that operate the knuckles. I think it's just a case of leaping to false conclusions because flattened fingers are often the product of ill-focussed tensions all over the place. But that's not the only thing that results in them.

The truth about piano technique is that there is NO "CORRECT" way at all!

Many virtuosos will display virtuosity by having a curved finger technique. Others will display viruosity with a flat finger technique ex: Horowitz.


There is no correct way of pressing a key. The fact that every individual is different and has slightly different hand mechanisms is the truth. Everybodies hand and wrist is slightly different but all have common charactistics - 4 fingers and 1 thumb.

Not true. Horowitz played with an arch. Watch some of his videos. He also played with a very low wrist position, so as to be able to play on the pads of his fingers as often as possible.

I wouldn't say either description is totally wrong. His most normal playing is not as flat as some people think, but if you look at the black key study he really is totally flat most of the time in the fingers of the right hand. It shows that there are plenty of adaptations that are worth making in the right context. However, while no way is a single correct truth, there are many things that might be worth employing in context that would could form a very poor foundation from which to base things overall. While I would never rule out either entirely, I don't believe that either extreme use of the tips or complete flatness is a good way to start. I think there are many different approaches that are valuable, but I do regard the level of balance and comfort that comes from using marginally curved pads as the primary point of reference- from which countless adaptations could be made. It's a good middle point from which you can either flatten more or curl up more, as the necessity arises. If you start with ease of balance as the number one goal, you can go on to whatever you want. Also, the student is not forced in one direction. It leaves them better equipped to make their own adapations if they start in the middle.

Of course. But the statement was that Horowizt played flat fingered, but most of the time, he has a curve, however slight, in his fingers.

Yeah, absolutely. I'd certainly agree that it was more normal for him to do that. Even when very slight, you can usually see that there is a sense of action in the final joint. He doesn't actually let it truly collapse very often.

I found this post enlightening in regards to the apparent disagreements and for me at least it gave food for thought.

As I understand, there is an aim to prevent a habit which will create a problem - namely the straight-out fingers. The idea is that if you start with a curve, eventually the other shapes will grow out of it, but if you allow a start with the rigidly straight fingers, nothing can (probably) grow out them. I also imagine that the first music will have a span of 5 fingers and few if any black keys, which has fingers more curved anyway. Then as the span spreads out and black keys are introduced, the hand will start having more shapes and be less curved. An observant teacher would watch for that.

I could also see a problem if a teacher (all teachers not being created equal) was not aware of this, and rigidly imposed the fingertip-down, round hand, in all circumstances because of being taught that herself without understanding it, that this could prevent more natural use of the hand. Of if a student follows too religiously, straining to maintain that curve even when the music doesn't call for it. I bet that this has happened lots of times. It would also explain some of the debate.

One thing about the human body - We make minute adjustments all the time. If you tried to make all the correct motions in stirring coffee, you would be relatively stiff. But if you allow yourself to respond to the weight of the spoon in your hand, and the resistance of the liquid, your coffee stirring will be tension-free. Likewise, I imagine, if you are reaching naturally for the keys it would be a shame to destroy this by religious adherence to the curved hand. Wouldn't that break a connection that we have with ourselves?

It would seem that a good teacher would observe a student carefully and adjust her advice according to what the student is doing rather than having one rigid rule - but maybe a "rule of thumb".

I imagine that various teachers would have different ways of approaching this, since there is not really "one single shape" or use of the fingers and hand. But they have to start somewhere.

KS, the problem I see watching my colleagues' students and others is that the fingers are too arched. They are playing too much on the finger tips. There is a time and place for using 'high fingers' and it is most often in Baroque and Classical repertoire, where we need a very articulated sound.

I am not blaming the teachers for this, as it is what most of us were taught. I was certainly taught this, and I believe if you look in many primers of very methods, including John Thompson, you'll see a hand diagram with a suggestion that it be shaped as if it were holding a tennis ball. In the USA and probably Canada as well, we've been suffering from this nonsense for at least a century.

Better for the teacher to help the student press the key and lift off the key with a high pointing wrist, so the student gets a feel for touch and release, not stab, jab, and jerk. A while back, someone put up a really nice link which illustrated this motion very well.

Yeah, I think it would make as much sense to argue that all students should start with flat fingers to protect them from overcurling, as it would to suggest that they use only the tips to protect them from straight fingers. Either could be beneficial in some cases but also harmful in others. A student who learns to feel what role every joint can play- balancing into the equivalent of the neutral position (but with small activities to counter the response force from the key) can either choose to curl up more than others or less. They can base it on the feedback of the sensations, not on the idea that a certain shape is always better. I've learned a lot from the exercises that Alan Fraser uses in his book and adapted various ones in slightly different ways for students. I don't personally say to use any particular shape- as long as they can get the feeling of each joint contributing something positive rather than doubling back (including the knuckle above all). I just show them how to form a comfortable balance on each finger. This leads to some small curves and some vastly more pronounced arches. However, when a student slowly explores the full range (by gradually moving from a totally flattened finger to the very tip, in one ultra-slow but continuous movement) I don't find that they come out feeling the tip was either the most comfortable or the most stable point to contact with. In some cases I've been really surprised by just how quickly what had previously looked like very weak, tiny hands start to find comfortably supportive shapes. If the student can feel what the purpose of the muscular action is and what achieves balance, it's a lot better than simply aiming for a certain shape because the teacher said so. When you evolve into the shape from scratch through a slow, continously balanced action, it's a lot easier than having a wild stab in the dark at what might be supportive shape- before then stabbing it into the key to see if it will collapse or not

Yes, well the point is that a good teacher seeks to have the student achieve good tonal control and relaxed approach to the keyboard at all times, and a lot of this conversation is based on a very adult way of approaching thinking about the body. A fixation on solely playing on the finger tips is patently absurd, but when students 'naturally' play with rigid digits (which is the norm for children from families without an instrumental culture - by which I mean, no one in the family has a clue how to play any musical instruments) teachers asking students to create a relaxed curved finger shape where the tip (as compared to the entire first two joints) of the finger is the point of contact with the key is an eminently sensible approach. You only need to listen to the difference between children who play with rigid fingers and those who play using the tip of the finger.

DISCLAIMER: I've realised that some participants in this conversation restrict the term 'tip' to that part of the finger that touches the key when the last joint of the finger is at a 90 degree angle to the keyboard. Take that angle back to 60 degrees at most, often closer to 30 degrees, and this part of the finger 'tip' is that which I would call the 'tip'. Playing with the joint at right angles to the key is not appropriate when pressing the key in the context of any musical flow.

The entire first two joints? What on earth makes you think that anyone might have suggested that, when referring to utilisng the pad? Personally I'd still regard 60 degrees as being the tip (and certainly not very close to 'relaxed' or a 'natural' angle), but I'd consider 30 degrees to count more for the pad. It's a little different for different hands maybe (not mention a little vague, with regard to exactly where the line would be imagined to extend along the finger) but those 30 degrees make all the difference for mine. Between those lies a totally different sensation of contact and balance. I'm probably closer to 20, for my regular position.

PS. technically the last joint is more tense if you keep some curve- not more relaxed. Forcing it down too much from the arm can make the collapse more probable. However releasing it doesn't make the joint stay rounded. A light inward grip does. No matter what you relax, a student who cannot perceive that tiny grip will usually collapse unless on the very tip. It's certainly more comfortable to keep that supporting, but that support really does come from muscular activity in the joint- not relaxation. I've found it far easier to deal with collapses since I understood this- in fact, excessive slack in these joints had been a big hole in my own playing until I learned of this less than two years ago. When a joint collapses either it is overrelaxed or it is being subjected to too much force elsewhere in the mechanism. The end product feels more relaxed overall, but it is not created by relaxation in the localised joint. I think it's likely that the rigid finger you described is the result of failure to feel how to employ action properly in the joint. When everything is felt to draws inwards (to equalize the force that makes for a collapse) it creates a tremendous opportunity to begin to relax elsewhere.

I wasn't thinking anyone was referring to that - I was thinking of the students who present having been practicing on a dodgy keyboard (or who believe they need to play with rigid digits - and there's LOADS of them about).

Bringing this thread WAYYYY back to the start, I think the Cashley might have asked too much question, and got things pointed in the wrong direction.

I play regularly on both a digital and a grand, and my fingers don't change at all. Why should they?

So was the original question about the difference, or about finger position?

One big difference I see between MY digital and MY grand is that the power curves are very different. It's much easier for me to get a more controlled dynamic range on the digital because the power curve seems longer--different levels of volume are farther apart in pressure. On the real piano, things happen in a smaller range of pressures, and I don't have enough control to get what I want; my playing on the grand can be uneven in volume from finger to finger, note to note, and the dynamics erratic and out of control, but on the digital it's easier for me to control these things. That is, the real piano is "twitchy" and shows up all of my faults.

I wonder if that's what prompted the original question?