Is economy of competitive cyclists affected by the anterior–posterior foot position on the pedal? J.R. Van Sickle Jr, M.L. Hull; Journal of Biomechanics 40 (2007) 1262–1267
When doing a bike fitting, changing cleat set-up is the keystone to a good outcome. While we have three contact points on the bike (hands, butt, feet) and all three CAN affect mechanics up and down the chain (i.e. if your left hand goes numb and so you hold it differently on the bar this can affect your shoulder position and induce a twist on the spine, which can affect how your sit bones rest on the seat....) the cleats/foot interface is always the most difficult to address, but also has the greatest gain associated with a correct adjustment. When you nail he cleat position it can often be a dramatic shift from a herky-jerky pedal stroke to a smooth elliptical rhythm.
The only way to truly read the necessary movements about the cleat is with an dynamic 3D analysis as is provided by the infrared systems (like the Retul). Video lacks the on-the-fly capability, is not taking measurements in 3 dimensions, and lacks the accuracy necessary to really make sound decisions. Prior to the Retul system being available, using the older methods of video, goniometers, plumb lines etc, we were just not able to address these small but, as we continue to find out, incredibly important factors.
So the article referenced above deals with one of the more coarse adjustments of the cleat -- the fore and aft positioning. This makes perfect sense because there is not much research out there about bike fitting, and specifically little regarding proper cleat alignment. (Incidentally the main adjustments to the cleats are the fore-aft, medial-lateral, varus-valgus wedging, and shimming for leg length under the cleat.) Marty Hull is one of the authors of this article, and if you're in interested in cycling and don't know who he is, you should. He has probably done more research into the mechanics of cycling than any other person on the planet. If you'd like to learn more about the ins and outs of cycling mechanics, you could do worse than reading his stuff.
In this study trained cyclists were tested at 90% of VO2 max with three different cleat positions: (1) standard forefoot placement (roughly under the metatarsals or "balls" of the feet), (2) midway between the rear of the calcaneous and the metatarsal heads, and (3) midway between 1 and 2.
They were tested on three separate days (with a rest day before each test day) and tested on all three cleat positions each day (in random order). They then measured how efficient each position was by way of how much oxygen was "used" (VO2) in the testing period of each cleat position. In simplistic terms, think of VO2 as the amount of oxygen that is used up (units in mL/min).
They found that there was no difference in efficiency with any of the cleat positions, so the more rear-ward positioned cleats were not an improvement in their cycling economy (more on why the rear-ward cleats might be more efficient later). So this seems to show that mounting your cleats further back on the shoe is not more efficient, but I, along with the authors, believe that there could be good reason why you would still opt for a "mid-foot" or "arch" cleat placement.
The first reason is Achilles/calf problems. When the cleats are further back on the shoe, the foot is "shortened" and so there is less of a lever arm about the ankle and so less muscular stabilization is required to keep the ankle relatively still as the main cycling muscles (quadriceps, gluteals, hamstrings) exert their force on the pedal. We do move our ankle while we pedal ("ankling") roughly 15-25 degrees, but the calves only provide about 7.5% of the total power output, so for many athletes with Achilles problems moving the cleat back can reduce the strain here and keep injuries away.
The next reason you may move cleats back is if you have foot numbness, tingling or pain issues. This may be because you have had prior foot problems (morton's neuromaor bunion, etc)
or maybe you are an ultra-endurance cyclist (the longer we ride, obviously the more pressure is exerted on our feet, but the feet swell slightly, making the shoes relatively tighter and placing more pressure on the nerves and soft-tissue of the forefoot). Many of the Race Across America (RAAM) competitors have been using mid-foot cleat placemen for years because when you push down on your pedals over 100,000 times a day for more than a week those tissues in your forefoot can get sensitive. If you aren't pressing through your forefoot, however, (and you push through the more hearty, muscled mid-foot) it cannot get sore.
You may also shift those cleats back simply if you have a hard time finding a perfect fitting shoe. Many "European" cut shoes are a hard fit for some. Poor fit can mean more pressure, and I've found that a slight shift rear-wards of the cleat can buy you some comfort.
The authors of this study were testing the theory that by moving the cleats back and having a reduction in the force requirement of the calf muscles, an improvement in economy (or efficiency) may be had. While an improvement in efficiency was not seen in this study, I think there is still hope, and the reason is motor planning. In my 14 years as a physical therapist I see people make improvements from many changes that we make to their mechanics, but those improvements don't become fully realized or don't fully coalesce until the person has mastered that new motor plan.
This process can take a few weeks or longer depending on how complex the task is. I have always believed that the pedal stroke is a lot more complicated than most people give it credit. Sure anyone can pedal a bike, but there are few people that can exert a nearly uniform force on the pedals for a majority of the pedaling cycle. In my research of pedal stroke analysis, I found early on that no one (NO ONE) can push/pull through the entire pedal stroke. Everyone, even the best pedalers cannot get their foot out of the way fast enough on the back stroke and so the put a "negative torque" on the cranks.
Then consider being able to activate the quads and hip extensors at that precise (and earliest) moment, to initiate the power stroke in the most efficient way -- the whole cycle, done well, requires a lot of coordination. Need further proof? Take 10 experienced cyclists and get them on a trainer for some one-leg pedaling drills. You will see many floundering individuals. Not withstanding this guy:
So given the complexity of pedaling a bike, with an uninterrupted block of time to adapt to the new cleat position, I think you will find that the improvement in economy would be seen. I have no hard proof of this, of course, just a hunch.
In my own personal experience I find that having my cleats back has prevented foot and heel problems when I was doing 24-hour races, helped me run better off the bike in Ironman races, and generally gave me the feeling of having very solid, consistent power output on the road bike.
It's not for everyone, but many people can benefit from this simple adjustment