Recently the question of closed versus open fingers came up on the Facebook Swim Coaches group (what an incredible collection of coaching minds!), and it amazed me as to the diversity of opinion on what I thought was more or less a closed question. So I went looking to not only find out the present state of the research, but also to see what coaches and athletes were thinking. Here’s just a sampling.
Most said keep the fingers and thumb relaxed and in a natural position. The next theme involved variations on fingers tightly together so no water would ‘slip’ through. And many specified exact finger spacings or a small range of spacings. Here are some other comments from coaches.
- thumb should be at 90° with four tight fingers
- thumb anywhere but 90°
- cup your hand so that the water doesn’t ‘spill’ over the sides
- finger spacing should be with width of your fingers
- swimmers should wear finger spacing gloves to train the fingers
- (and my favourite) open fingers mean the palm of the hand will move through the water more quickly, so attention has to be paid to moving the fingers faster to catch up
The general idea is that a large number of swim sites discuss the issue but simplify the scientific study results horribly, resulting in incorrect generalizations and bad explanations. As a result, far too many professional coaches and swimmers are still confused about optimal finger spacing.
The Physics
So let’s start with a basic understanding of the problem, and then we’ll get to the studies that have been done.
Most of the problems come from a misunderstanding of how fluid dynamics work. At the simplest level, there are 3 factors affecting the drag coefficient of the hand (roughly equating to the effective surface area of the hand).
- the larger the surface area of the hand, the higher the drag coefficient. The only time this will change is a) when the fingers are tightly closed, essentially squishing the fingers and reducing the surface area, and b) when the fingers and thumb are spread, the webbing between the digits will spread apart increasing the surface area a very tiny and probably insignificant amount.
- there is a thin boundary layer of water surrounding the fingers. This layer resists movement, and acts as an extension of each finger, increasing the effective surface area of the hand. The thickness of the boundary layer depends upon a lot of factors, including speed of the fingers through the water, shape of the fingers, proximity to other fingers, etc.
- any water flowing around or between the fingers will create vortices on the other side of the hand, and this will impact the drag coefficient. However, I believe we can ignore this completely, as the effect will be negligible in a real world scenario of a turbulent swimming pool
What this means is that the larger the total surface area, the greater the pulling force. And if we leave a slight separation between fingers equal to at least the thickness of twice the boundary layer, then we can increase the effective surface area of the hand.
The Studies
There have been dozens of hand positioning studies done over the past few decades, some rigorous and precise, and some far, far less so. Of the literature I could find, four studies stand out, all using computational fluid dynamic [CFD] simulations. I should point out that all of these simulations use highly idealized situations (lack of turbulent water or cross currents, simplified modelling of digits, etc).
The Optimum Finger Spacing in Human Swimming, by Alberto Minetti and others (here).
This was the most interesting study, as the CFD simulations were the most complex and sophisticated, and involved the whole hand. Eight different finger spacings and multiple water flow scenarios were modeled.
The primary finding was that the best finger spacing was at 8 mm, with the second best at 3 mm. Interestingly, finger spacings between 3 and 8 mm produced far worse results, but I chalk this up to the highly idealized nature of the digital simulations. Spacing of 8 mm produced an effective surface area 8.8% above wide open and fully closed. This spacing turns out to be the equivalent of a relaxed hand.
The Constructal-Law Physics of Why Swimmers Must Spread Their Fingers and Toes, by Lorente and others, (here)
The CFD simulations used very simplified models (only 2 fingers represented as ideal cylinders). While far less sophisticated than Minetti paper, this study is important as it confirms that the optimal finger spacing is between 20% and 40% of the width of the finger. This turns out to be roughly 4 to 8 mm.
Swimming Propulsion Forces are Enhanced by a Small Finger Spread, by Marinho and others (here).
These CFD simulations also used far less complex models, but included 7 different angles of attack for the hand, but only three finger spacings. Not surprisingly the best angle of attack was exactly 90°, while the best finger spacing was at 3.2 mm.
Hydrodynamic analysis of different thumb positions in swimming, Marinho and others, 2009 (here)
This CFD study is interesting in that it is one of the few to analyze the role of the thumb in the pull. The results showed slightly better results with the thumb adducted (towards the hand) than abducted (away from the hand). Unfortunately, only 3 thumb positions were analyzed, and so an optimal angle wasn’t investigated.
Summary and Remaining Questions
The four studies above represent the most rigorous ones I could get hold of. All studies support optimal finger spacings roughly between 3 and 8 mm, which all correspond to a relaxed hand position as best. The study on thumbs supports a relaxed position closer to the hand. The optimal hand positions are estimated to increase the effective surface area of the hand by roughly 9%. Although it should be noted that all studies use highly idealized conditions.
Surprisingly, there are still some remaining questions that have not been adequately answered in studies. Here are two of them:
What is the optimal angle for the thumb? The most rigorous study merely found that a thumb closer to the hand is better. But I couldn’t find any thumb studies that searched for an optimal angle or range of angles.
What is the optimal hand and finger configuration for hand entry? Video analysis shows elite swimmers with a very relaxed hand position on entry, even though intuitively we might think that a more compact hand upon entry might be better.
Coach Rick - Mighty Tritons Swimming 
Hmmm…………..
Interesting info. The article didn’t say, but I’m inferring that the reason for optimal finger spacing and thumb placement is to generate more force and thus achieve more power for propulsion.
Swim teachers and coaches often fall into the trap of thinking that generating more propulsion will result in more speed and lower times. This is very similar to coaching a golfer that swinging the golf club harder will result in achieving greater distance on each shot. True, but hugely oversimplified.
The “more power” argument is oversimplified because humans are land creatures, not fish. Increasing force for sports that mostly involve running, jumping, etc. works fine because that is aligned with our design. Apply this same logic to a fish trying to become more proficient at land sports. Think of coaching a dolphin to move faster over land instead of water. He would instinctively do what he knows which would be to wiggle faster. This would not produce a successful result.
Achieving better performance in swimming is mostly accomplished by increasing the skill level to make the body more precisely act like a fish, which is to actively decrease the forces of friction and wave drag on the body.
The most successful swimmers excel at doing these three things almost automatically:
1) Aligning and balancing their bodies in the water so that they equalize the forces surrounding them as they swim. This also is a principal law of fluid dynamics.
2) Using the feeling of “catching” the water with their hands as a stimulus for engaging their core and using the most powerful muscles in the body vs just arm and leg muscles. When this is done correctly, the hands act much more like anchors than paddles. Hand and finger placement is key in this and likely varies with each swimmers “feel” for the water.
3) Recruiting the correct muscles / muscle groups accurately in terms of timing and force vs body alignment and drag reduction. This, of course, becomes increasingly more difficult as aerobic stress and muscle fatigue increase.
This is why training swimmers becomes more complex and difficult as they get older and plan to achieve higher goals.
Thanks for the excellent comment, Mike. You’re absolutely right about the holistic approach to swimming, there are many, many things that need to align in order to get the most out of a swimmer. This article, like most of mine, are dealing with specific aspects of our sport. It more or less assume that you have a decent stroke, and are looking to maximize that propulsion. When talking to younger swimmers I refer to canoeing. You want the oar to push as much water as you can, and this might mean you row more slowly than if you turned the oar 90 degrees and push nothing very quickly. I was amazed at how many coaches contacted me about this post saying they’d always told their swimmers to squeeze their hands to not let any water through (wasting energy and decreasing effective hand size), or have fingers splayed as wide as possible (again, to do that requires extra energy without any benefit).
So yes, you’re right that this is just one piece of the puzzle, and certainly not the most important one. But eventually I’ll get around to talking about all the pieces of the puzzle.
I’m inclined to think the optimal thumb position would be between 3 and 8 mm from the rest of the hand, just like optimal spacing between fingers. Same principle would suggest that there would be an effective boundary layer between the thumb and the hand. I’ve stopped teaching my students to make “ice cream scoops” with their hands. Even if they aren’t concerned about cutting .0001 seconds off their times, I find a relaxed hand position makes my swimming feel a bit more. . . relaxed. Also keeps me from getting hand cramps. I’ve also recently read that the paddle style stroke is more efficient than the sculling stroke, and I’ve never seen a boat paddle shaped like an ice cream scoop. Just my take.
Hi Mike. I’d imagine that holding the thumb that close to the fingers would be quite awkward for any appreciable amount of time. I think that’s why the thumb is treated separately. As for the hand, you’re correct in that cupping the hand drastically reduces the pulling surface (which I’m guessing is worth far more than .0001 seconds). The relaxed hand increases the surface area. And lastly, we went to flat hand sculling strokes quite a while ago, for the same reason.
I use the .0001 second as a hyperbolic with my students just to de-emphasize the speed aspect of any particular stroke detail. Most of my students have no aspirations of competing at all, let alone in an environment where fractions of a second would be a concern. With them, my focus is on preventing them from getting tired rather than going faster. So, that’s just a number I throw around for effect.
I keep my thumb relatively parallel to my hand with a gap of somewhere between 3 and 8 mm, and I don’t find it awkward or uncomfortable at all. It’s certainly more comfortable than squeezing it against the side of my hand in the old ice cream scoop style. I was quite surprised at how sore and cramped my hands got doing the old scoops! Even if it weren’t faster, I’d stick with the flat hand style now that I’ve tried it.
I like that approach of de-emphasizing the smaller details for those who don’t want to compete and just want to have fun and get fit. As far as the thumb, the research really just says that you position is wherever it’s comfortable, with the optimal spacing being twice the boundary layer. And yes, squeezing the hand does cramp things up. Try the relaxed hand with sculling and you’ll notice the same thing. More movement of water, and a more relaxed hand.
So, I have been coaching relaxed hands for as long as I can remember, but find the actual entry point a more tricky question. On my recent Senior Coach programme through Swim England, I was advised that a 40degree angle was best, but I notice that you say a 90 degree is more efficient. At 90 degrees, wouldn’t you need to ensure fingers were together at entry, relaxing during catch and pull? My swimmers “look” better with a 30-45 degree entry as they maintain the relaxed position, but I would be interested to hear your thoughts on this
Hi Sandra, I think there’s some confusion here. The 90 degree angle of attack I described in the article really refers to the pull phase and not the entry phase. 90 degrees just means the hand is perpendicular to the line of pull. That makes sense. The studies didn’t look at the angle of entry at all. In fact, that issue would be fascinating. Most experts feel the entry phase should definitely involve a relaxed hand, and result in as few air bubbles as possible. In other words, slip the hand in with little splash. A study on the impact of fingers together or slightly apart during entry would be interesting.
Thanks, that does make sense. Traditionally teachers were instructed by swim England to advocate a thumb first entry , sliding the hand in , but that just makes the catch difficult and made swimmers “snake” through the water, over reaching across the centre line. I agree with the 90 degree angle on the catch entirely. It would be interesting to see studies on the desired hand entry point/ angle. I am forever trying to minimise splash at entry point so this is a big question for my swimmers. Keep me posted if you have any thoughts or observations.
I envision a 90-degree entry as being much like using a pair of ice axes to pull yourself through the water – your hands being the business end of the axes and your forearms being the handles. Or, like the windmilling style arm movement but with the wrists bent at 90-degree angles. Would definitely be interesting to study! (But, the stroke I’m picturing is painfully awkward to imagine.)
Your response had me recall something I was looking at last week. First, in a conventional bent-elbow FR recovery, the hand enters the water before full extension, and then extends. You obviously want this small movement to have minimal resistance, and therefore the hand should be in line with the arm movement. The alternative would be to fully extend above the water and then drop down, but this puts huge loads on the shoulder for no advantage. (I’ve had swimmers try this and they always end up complaining of tight shoulders). BUT, a straight-arm recovery basically allows entry with an extended arm. Last week I was looking at some underwater videos of straight-arm FR hand entries, and the hand was slightly tilted down, I’m guessing at 20-30 degrees. I realized this enables an immediate and useful pull sequence, as opposed to a flat hand entry where time would be required to get into a useful pulling position. I can also imagine this is more efficient than a 90-degree hand entry as you’re giving up a tiny part of the initial pull sequence (moving hand from 30 degrees to 90 degrees). I have no idea what the ideal angle would be, but 0 degrees and 90 degrees intuitively seem wrong.
Yes, I can see where the differing angles would be useful on straight and bent arm recovery. I will do some work with my club and see how works best ( I suspect that, as with all things, it will be slightly different for each swimmer). It’s good to think laterally about these things and I enjoy focusing on Small details with my swimmers so I can really get to the bottom of issues and it also helps the swimmers to understand their own styles and take ownership of their progress with an understanding of how the slightest adjustment can make a difference ( good or bad). Good to talk with you and thanks for the input.