In the November 2006 issue of CrossFit Journal, Mark Rippetoe published, “A New, Rather Long Analysis of the Deadlift.” He concluded this breakthrough article by identifying three criteria for a correct deadlift starting position:

  • The back must be locked in extension.
  • The bar must be touching the shins with the feet flat on the floor.
  • The shoulders must be out in front of the bar so that the shoulder blades are directly above the bar.

Rippetoe expands on these criteria in the second edition of Starting Strength and has refined his analysis in the Basic Barbell Certification course that he teaches through CrossFit. I’ve used Rippetoe’s pulling model to improve my own deadlift and teach clients how to deadlift. The 58-year-old client I described in a previous article recently increased his deadlift from 340 lbs to 355 lbs using this model, and my 132-lb friend, Cathy, pulled 175 lbs after a couple of workouts following this model as well.

Rippetoe and I have recently been analyzing deadlifts by powerlifters such as Andy Bolton, Ed Coan, and Brad Gillingham. Without exception, these videos confirm his claim that the bar does not leave the floor on heavy pulls until the shoulder blades are over the bar with the bar over the middle of the foot. I recently spoke with Rippetoe about some specific points of his pulling model and how he has refined his analysis.

MK: In Starting Strength, you discuss the importance of the quadriceps in the beginning of the deadlift to produce knee extension. You state, “The quadriceps must participate in the deadlift properly in order for the movement to be safe and efficient.” In this regard, you have recommended weightlifting shoes to better utilize the quads. Powerlifters are nearly always found deadlifting in some kind of flat shoe because it reduces the range of motion and makes it easier to get on the heels and recruit the posterior chain. From a powerlifting perspective, would the benefit of extra quad recruitment with weightlifting shoes outweigh the extra inch of the range of motion (ROM) on the pull and diminished posterior chain recruitment?

MR: I have deadlifted heavy both ways. In my later lifting career, I switched over to weightlifting shoes and pulled my biggest deadlift—633 lbs—on two separate occasions. The shoes I wore were the old, Adidas weightlifting shoes that had a little heel (about 5/8-inch elevation from the ball of the foot to the heel).

When powerlifters think about getting back on the heels, they’re putting the bar over the middle of the foot because when you start pushing on the ground with a heavy bar, that’s where it’s going to be in balance. If it’s behind the middle of the foot, you’re going to be off balance backward. If it’s in front of the middle of the foot, you’re going to be off balance forward. So we don’t really want to be on the heels. We want the bar directly over the middle of the foot. A better cue would be to push the middle of the foot into the floor. I like to think about the contact path directly under the bar and push it into the floor. That keeps me in better balance than doing it any other way.

If I have a little heel under my shoe, that places my knee in a position that is a little bit forward, maybe three to five degrees more forward than it would be in flat shoes. This increases the quadriceps’ effectiveness off the floor because it increases the knee angle enough that it allows the quadriceps to open the knee up over a longer range of motion. Thus, the quadriceps has more work to do. It doesn’t increase the knee angle enough that posterior chain recruitment is diminished at all. The only way to do this would be to have enough of the heel so the lift would produce a knee angle so acute that it actually shortened the hamstrings from the distal end. We’re using the hamstrings from the proximal end to extend the hip. That’s the action of the posterior chain on the hip. If the knee angle is made acute enough that you actually put slack from the distal end into the hamstring you’re attempting to use from the proximal end, you diminish the effectiveness of the use of the posterior chain.

If the heel is low enough, there is no compromise between posterior chain recruitment and quadriceps recruitment. You only get that when the heel is too high. The heel on the shoe, while taking a tiny bit of tension off the hamstrings and making it a little bit easier to get into that extended lumbar position, also produces enough increase in the knee angle so that you get a little bit more work out of the quadriceps off the floor. I think it’s a reasonable compromise between that and the extra distance that the bar would have to be pulled. And the extra distance would be the height of the shoe directly over the middle of the foot, not at the end of the heel because you lock the bar out over the middle of the foot as well. That’s the extra distance that you’re pulling.

I also think wrestling shoes almost make it harder to get into the correct position because of the increased tension that the straighter knee places on the hamstrings. It’s harder to get the lumbar spine into an extended position against that hamstring tension. The pelvis is in a war for control between the muscles of the low back and the muscles of the hamstrings. If you’re going to pull the bar off the ground and not eventually hurt yourself, the lumbar spine needs to be in extension. That means the back muscles have to win the war.

In fact, the reason some deadlifters and most Olympic lifters rock forward onto their toes before they pull is because this takes the tension off the hamstrings and allows the low back to get into an arched position more easily. But it also means that the bar has to travel back to the mid-foot during the pull. This explains the “hook” in the bottom of the bar path in a clean and snatch, which is light enough compared to a deadlift that the bar can actually leave the ground in this position. They can be done this way, but a heavy deadlift won’t leave the ground very far in front of the mid-foot.

The bottom line is that I don’t want less muscle in the movement. I want as much muscle in the movement as I can get. I need them all because I’m going to use them all. Whether I’m at my one rep max or on my third attempt, I need as much stuff helping me out as possible.

MK: When you mention a compromise with the extra distance that one pulls the bar with weightlifting shoes, one could likewise say that flat shoes create compromises regarding quad recruitment and proper lumbar position.

MR: Yes, you could. There are compromises on both sides that have to be appreciated before you can make an informed decision about it. I did it both ways, and I like a little bit of heel better. I understand that many people are doing sumo deadlifts right now and that might change the equation a little bit. I never did sumo. Sumo is used because it produces a more vertical back angle, and the wide stance reduces the effective length of the femur. There are mechanical differences between conventional and sumo that might favor a flat shoe. I don’t pull sumo though so I don’t know. I don’t know anybody who does pull sumo and doesn’t wear flat shoes.

Now, it very well might be that somebody pulling sumo might find a benefit out of a little bit of a heel in the same way that a conventional puller would, but I’d be pulling things out of my ass if I said that I knew for sure.

MK: If weightlifting shoes facilitate greater quad recruitment from the floor, it seems flat shoes are more advantageous in deadlift assistance exercises like Romanian deadlifts and good mornings that seek to exclude the quads. In deadlift assistance exercises like pulls on blocks and pulls with 35s entailing greater knee extension, weightlifting shoes seem more advantageous.

MR: These all seem to be reasonable conclusions.

MK: What is the maximum heel height that you have found to be compatible with keeping the bar over the middle of the foot?

MR: I like somewhere between 0.625 inches and 0.75 inches. That’s the difference between the height under the ball of the foot and the height under the heel. We have investigated this at length, and we don’t think that you can get somebody in position with the bar over the middle of the foot with a 1.75-inch heel.

MK: You mentioned the sumo deadlift. What are your thoughts on the sumo deadlift as opposed to the conventional stance?

MR: As I said, the sumo is a way to increase the verticality of the back angle. This shortens the moment arm between the hip and the bar by effectively shortening the length of the femur when the stance width is increased and by shortening the moment arm between the hip joint and the scapulas when the back angle is made more vertical. It is interesting that there are some people who can’t deadlift efficiently with a conventional stance. For example, some people have femurs and tibias that are so long relative to their back length that they can’t get their back angles much above horizontal without a stance that sufficiently shortens the effective femur length. We found a woman at one of our seminars whose femurs were four inches longer than her back. She had never been able to deadlift conventionally without having the bar ten inches in front of her mid-foot at her start position. When she was in the conventional stance with the bar over her mid-foot, her hips were actually much higher than her scapulas. We put her in a sumo stance, and she got her back at what would be a normal angle for a more proportioned person. She set a PR in her deadlift by 10 kg that day.

MK: In order to have the bar over the middle of the foot, what variances have you found in the distance the bar is from the shins in the set up with people of different foot sizes?

MR: Essentially none. I teach 35 people at least twice a month to deadlift using this model. I’ve seen everyone from a girl with a women’s shoe size of four to a guy with a men’s shoe size of 17. All were in essentially the same place with the bar about an inch away from the shin when the shin is vertical. It’s surprising that there’s that little variability. But, nonetheless, if you line up about an inch from the bar when you’re standing straight up, you’re going to be in the right place. Human foot and leg proportions are remarkably consistent no matter how big or small they are.

Another thing that we have recently begun to understand about this is the necessity for most people—and especially bigger guys—to point the toes out to activate the hip and make room for the gut between the femurs in order to express the correct back angle. The correct back angle is, of course, where the scapulas are directly over the bar. Generally, the thinner you are, the less important it is. But the heavier you are, the more important the toes-out position is.

What we find is that if you take your vertical jump stance right under the bar (relatively narrow stance), place the middle of the foot under the bar, and then point your toes out maybe 15–20 degrees, you make it much easier to express the correct back angle when you’re in thoracic and lumbar extension. It also enables you to place a little bit of tension on the adductors with that external rotation. To a certain extent, you can call them into hip extension because the adductors are hip extensors. They’re part of the posterior chain. When you point your toes out and point your knees out a little bit, you actually tighten them up and get a little bit of help out of those muscles. If you look at old videos of Vince Anello, he figured this out (http://www.youtube.com/watch?v=rQD9mHTLFCA).

For more information on activating the hips, see Rippetoe’s forthcoming article in CrossFit Journal on “Active Hip.”