Once upon a time, you were told to always keep your knees behind your toes when you're doing anything like a squat. Maybe you have a gym buddy, call him Gallant, who still follows that advice, sticking his butt out to keep his balance and watching those knees in the mirror. Meanwhile, Goofus threw the rule out years ago, and changes the subject when Gallant tells him he's ruining his knees. Both can trade stories about people who have bad (or good) knees and how they squat, but we're here today to answer the question: who has science on their side?

First, the reason behind the rule: keeping your knees behind your toes is supposed to protect your knee joints. The study that kicked off the classic advice was done in 1974 and was one of the first computer analyses of biomechanics. Among twelve weightlifters, the ones with the most forward motion in their knees also had the greatest shear forces. (Imagine that your femur wants to slide right off your tibia as you're bending your knee. That would be a shear force. ) Your ligaments and tendons keep this from happening, but is the strain so great that those tissues are prone to injury?

We'll come back to that question in a minute, but first let's look at a hot-off-the-presses study that flips the question on its head: Could Gallant's "restricted" squat technique be putting him at risk for injury? The study is far from conclusive, but these Swiss researchers think the answer is yes. Here's why.

30 men (all students with weightlifting experience) did squats in front of a set of video cameras, with 71 reflective stickers marking the position of various body parts, from shoulders to feet, with particular attention to the back. The men were asked to squat while keeping their knees in front of their toes (watching video of themselves) and to also do "unrestricted" squats where any knee position is allowed.

Computers assembled a 3D model based on the video, and the researchers could measure joint angles on the model. Based on the angle, the researchers could calculate the forces on the knee, hips, and lower back. (This is a standard technique in kinematic research, but it doesn't measure strain on tendons and ligaments directly.)

The results? Keeping your knees behind your toes causes a "compensatory motion in the trunk," in other words, sticking your butt out. That causes greater forces on the back (including both the lower back and the thoracic, or ribcage, area) and less work being done by the legs. So not only is Gallant putting his back at risk, he's also not strengthening his legs as well as he could be.

This agrees with earlier research. A study from 2003 found that restricting squats reduces knee forces slightly, but increases the forces on the hips and low back tenfold. Changing your form doesn't just affect your knees; it shifts stress to other body parts, possibly putting you more at risk for injury.

The studies have some caveats, though. For one thing, the Swiss study only tested weights up to 50 percent of the subject's body weight. (The 2003 study went up to 100 percent, while the classic from 1974 had lifters squatting whatever they liked, around 200-300 percent of their body weight.) All used only men as their subjects, and there is evidence that men's and women's knees perform differently during squats. A study that asked athletes to do single-leg squats found that the women's anterior cruciate ligaments were being strained more than the men's. (Women are known to be more at risk for ACL injuries in general, although the reasons aren't fully understood.) Could women benefit from restricting their double-leg, weighted squats even if men don't? We don't know.

Since this is a real world question, we're interested in real world results. And that's where the research falls short: none of these studies look at actual injury rates among athletes, which is a tougher question to answer. If Gallant gets injured and Goofus doesn't, or vice versa, we can't be sure that squat technique was the reason. There will be other differences in how they work out, what they eat, and what their anatomy was like to start with. It takes huge studies to sort out those variables, and studies that big have their own problems.

So, is this myth busted? Sort of. Restricting your squat does seem to reduce forces on the knee – but at the expense of your hips and lower back. Which one matters more in real life? We don't have enough data to say for sure, but the Swiss team ends their study with some words for Gallant: "Practitioners should not be overly strict ... in coaching against anterior knee displacement during performance of the squat."