The most heated argument in strength and conditioning today is to crunch or not to crunch. It's bewildering that this seemingly harmless, short ROM exercise could create such a rift between so many smart strength and conditioning professionals, yet the great crunch debate rages on.

At the center is research showing that repeated spinal flexion using cadaveric porcine spines resulted in herniated discs. This in vitro research seems to indicate that lumbar flexion is a potent herniating mechanism, and anti-crunch proponents have extrapolated from the data that humans possess a limited number of flexion cycles throughout their lifetimes.

Accordingly, they've gone out on a limb and recommended that spinal flexion exercises such as the crunch be avoided at all costs. While this may seem logical on the surface, there's more to this topic than meets the eye.

Someone needs to step up and grow some balls and address the 2000-pound gorilla soiling the carpet. We know dozens of respected strength coaches, physical therapists, personal trainers, researchers, and professors that all have serious doubts about the danger of crunches, yet none wish to discuss it out of fear of being chastised.

The line must be drawn here!

Humans have a basic need to fall into camps, rally behind a leader, believe in supernatural phenomena, and rebel against scientific principles. Throughout history scientists have been punished for questioning current dogma. Sadly, it's no different in the fitness industry.

Many fitness professionals have been seeking a culprit for low back pain and jumped aboard the anti-crunch bandwagon without question. These folks have adopted absurdly rigid views of the lumbar spine, believing that you should go through life moving this region as little as possible to spare insult to the spine, to the point of altering normal biomechanics in daily living.

Taking it a step further, they then intimidate others into jumping on the bandwagon and get downright emotional when confronted on the topic.

This is the antithesis of scientific thinking. We're just happy that we won't be house-imprisoned like Galileo for hypothesizing that the Earth wasn't the center of the universe.

After delving into the topic, reviewing the literature, and applying our critical thinking skills, we've concluded that like every other exercise, a reasonable dose of spinal flexion exercise is potentially good for you and need not be avoided.

We presented our position in a review paper published in the Strength & Conditioning Journal and while we won't rehash everything in the article, we do continue to question the recent "anti-crunch" movement and suggest a plausible alternative theory.

In our journal article we addressed the following issues, which will only be succinctly summarized below.

For more detailed explanations and citations, we encourage you to pull up the article and read it in its entirety.

  • Removal of muscle. The experiments used in the studies used to refute lumbar flexion exercise used porcine (pig) cervical spines with muscles removed, which alters spinal biomechanics.
  • No fluid flow in cadavers. Cadaveric spines don't function the same as living spines as fluid doesn't flow back into the discs as it does when tissue is alive.
  • Range of motion in porcine spine. Porcine cervical spines have smaller flexion and extension ranges of motion than human lumbar spines.
  • Doesn't mimic crunch exercise regimen. When most people do crunches, they might do a few sets of 10 to 20 reps or so and then wait a couple of days before repeating. This allows the discs to repair and remodel. In the studies used to bash lumbar flexion, thousands of nonstop cycles were performed, which does not replicate a strength and conditioning regimen. It should also be noted that cadaveric spines do not remodel while living spines do.
  • Genetics. In the world of intervertebral disc degeneration, the role of genetics is huge. It appears that some individuals are quite prone to disc issues while others are not, suggesting that optimal programming would require knowledge of genetic traits.
  • Range of motion in the crunch exercise. Many years ago, an NSCA journal article described the proper performance of the crunch exercise, which involved 30 degrees of total trunk flexion, most of this motion occurring in the thoracic spine, not the lumbar spine. If the lumbar spine doesn't approach end-range flexion in the crunch exercise, then the studies wouldn't be applicable to the crunch exercise.
  • IAP controversy. There's a chance that models used to estimate compressive forces during the crunch have been overestimated due to failure to take into account the role of intra-abdominal pressure (IAP). If this is the case, then the studies could have used too much compression along with the range of motion mentioned earlier, which would render the studies inapplicable to crunching. However, there's conflicting research in this area and it's likely that the effect isn't significant.
  • Increased fluid flow and nutrition to posterior disc. Lumbar flexion enhances nutrient delivery to discs by increasing nutrient-carrying fluids to the discs.
  • Increased remodeling of tissue. Proper doses of spinal flexion likely strengthens the disc tissues, which would therefore increase tolerance to lumbar flexion exercise and prevent future injury.
  • Sagittal plane mobility. Some studies have linked lack of spinal mobility to low back pain, however the literature is somewhat contradictory. At the very least crunches can prevent losses in spinal mobility, which might be important in low back pain prevention.
  • Rectus abdominis hypertrophy. When taking into account the entire body of knowledge on hypertrophy research, it's abundantly clear that dynamic exercise is superior to isometric exercise in increasing muscle mass. Much of this has to do with the increased muscular damage incurred from eccentric activity as well as the increased metabolic stress. Bottom line, if you want to optimize your "six-pack" appearance, spinal flexion exercises will certainly help to achieve this goal.
  • Performance enhancement. Contrary to what some have claimed, lumbar flexion is prevalent in many sport activities. Thus, concentrically/eccentrically strengthening the abdominals may very well lead to increased athletic performance.

Now let's look at some anecdotal evidence to support our claims.

If we're indeed "limited" in the number of flexion cycles, where does the number lie?

Several fitness professionals have suggested that humans possess a limited number of flexion cycles and believe that we should save these cycles for everyday living such as tying one's shoe rather than wasting them on crunches. Realizing that anecdotal evidence doesn't prove squat, it's still interesting to ponder and can provide a basis for theoretical rationale.

  • In 2003, Edmar Freitas, a Brazilian fitness instructor, performed 133,986 crunches in 30 hours, thereby setting a world record. This beat his previous record of 111,000 sit-ups in 24 hours set in the prior year.
  • Edmar Freitas
  • Manny Pacquiao, one of the world's best boxers, performs 4,000 sit-ups per day.
  • Manny Pacquiao
  • Finally, Herschel Walker, football legend, Olympic bobsledder, and current MMA hopeful, has been performing 3,500 sit-ups every day since he was in high school. He started doing sit-ups daily when he was 12 years old. Considering that he's now 49 years of age, this equates to 47,267,500 sit-ups. That's almost 50 million flexion cycles performed under compressive loading!
  • Herschel Walker

Granted, the argument could be made that perhaps all these individuals have screwed up spines and if you were to obtain MRI's from each you'd see appalling evidence of herniations and degeneration, but we doubt this is the case. Instead we believe that this is clear evidence that the spinal discs can remodel and become stronger over time to resist damage incurred from spinal flexion exercise.

Another argument could be made that these folks are "outliers" and their freakish genetics allow for such incredible flexion cycles. We disagree, and believe there are likely many individuals that have unknowingly met or exceeded this number in their lifetimes. Instead, we believe that this is evidence that muscular balance, abdominal strength, and flexion exercise can protect the spine.

If we cherry-picked select disc studies to determine which forms of exercise we do, we wouldn't be allowed to do literally anything.

We found 13 studies to indicate that spinal flexion is a bad idea. (Callaghan and McGill, 2001; Drake et al., 2005; Tampier et al., 2007; Drake and Callaghan, 2009; Marshall and McGill, 2010; Adams and Hutton, 1982; Adams and Hutton, 1983; Adams and Hutton, 1985; Lindblom 1957; Brown et al. 1957; Hardy 1958; Veres et al., 2009; Court et al. 2001). This means no crunches and sit ups.

We found 11 studies that showed that combinations of spinal loading aren't a good idea. (Gordon et al. 1991 (Flexion and Rotation); McNally et al. 1993 (Flexion and Anterolateral Bending); Shirazi 1989 (Lateral Bending and Rotation); Kelsey et al. 1984 (Flexion and Rotation); Adams et al. 2000 (Complex); Marshall and McGill, 2010 (Flexion/Extension and Rotation); Drake et al. 2005 (Flexion and Rotation); Veres et al., 2010 (Flexion and Rotation); Schmidt et al. 2007 (Lateral Bending and Rotation, Lateral Bending and Flexion, Lateral Bending and Extension); Schmidt et al. 2007 (Lateral Bending and Flexion, Lateral Bending and Rotation, Flexion and Rotation); Schmidt et al. 2009 (Lateral Bending and Flexion, Lateral Bending and Extension). This means no exercises such as twisting sit-ups or Russian twists.

We found a couple studies showing that spinal extension is a big no-no (Adams et al., 2000; Shah et al., 1978). This means no Supermans.

Several studies show that spinal rotation is bad (Krismer et al., 1996; Aultman et al. 2004; Farfan et al. 1970). This means no cable chopping motions. And you better not be performing these on a vibration platform as that would produce a double whammy to the spine. Vibration has been shown to be bad for the discs (Dupuis and Zerlett 1987).

There's research to suggest that lateral bending is bad for you (Costi et al. 2007; Natarajan et al. 2008). This means no side bending. Similarly, asymmetrical lifting has been shown to lead to negative results as well (Natarajan et al. 2008). This means no unilateral exercises or weighted carries.

Here's where things get interesting. We found 18 different studies showing that static or dynamic compression is a very bad idea (Virgin 1951; Liu et al. 1983; Lai et al. 2008; Lotz et al. 1998; Tsai et al. 1998; Iatridis et al. 1999; Lotz et al. 1998; Kroeber et al. 2002; MacLean et al. 2003; Hsieh and Lotz 2003; MacLean et al. 2004; Ching et al. 2004; Masouka et al. 2007; Veres et al. 2008; Lai and Chow 2010; Nakamura et al. 2009; Wang et al. 2007; Huang and Gu 2008).

This not only means no squats, deadlifts, and core stability exercises, it means no physical activity involving free weight exercise whatsoever, since muscular contractions create compressive loading on the spine. We should have known that though, as heavy lifting is bad for the back (Lee and Chiou 1994; Kelsey et al. 1984), as is overactivity (Videman and Battie, 1999). So now we know that weightlifting and intense training is out.

It goes on and on. We've found studies suggesting all sorts of every day activities are bad for the back, even sitting down and bed-rest. Maybe we should all just shoot ourselves before we become completely debilitated?

The theory goes something like this: Crunches shorten the rectus abdominis. Since the rectus abdominis spans from the sternum/rib cage to the pelvis, continually shortening the muscle will pull down your ribcage, ultimately resulting in kyphosis (i.e. a round-back posture). It's an interesting theory. It's also completely unfounded.

As with many theories, the essence of this claim is based on a kernel of truth. Specifically, placing a muscle in a shortened position for a prolonged time causes it to assume a shorter resting length. For example, if you immobilize your arm in a cast at a flexed position for several weeks, your arm will tend to remain flexed once the cast is removed.

This is due to an adaptive response whereby the elbow flexors (i.e. biceps, brachialis, etc.) lose sarcomeres in series while sarcomeres are added to the antagonistic extensor muscles (Toigo & Boutellier, 2006). This has been coined "adaptive shortening."

Perhaps you can see the flaw in hypothesizing that performing a crunch will shorten the rectus abdominis, namely, crunches aren't solely a shortening exercise! Rather, the crunch also includes eccentric actions where the rectus abdominis is returned to its resting length.

Thus, any potential negative effects of shortening contractions on sarcomere number would be counterbalanced by the lengthening effect of the eccentric actions. The net effect is no change in resting length.

Some anti-crunch proponents also argue that performing spinal flexion exercises (i.e. crunches) overly strengthens the rectus abdominis so that it overpowers its antagonists, thereby pulling down on the ribcage.

This is a straw man argument. Certainly it's true that an imbalance between muscles can cause postural disturbances — I'm sure you're familiar with guys who hit chest and arms every workout and end up so internally rotated that they have trouble scratching the back of their head. But this doesn't mean you shouldn't perform bench presses and arm curls.

The issue here is one of poor program design, not an indictment of specific exercises.

Regarding crunches, the same principle holds true. Sure, if you perform a gazillion crunches every day and don't train other muscle groups, you're setting yourself up for a postural disturbance.

But this is a non-issue if you adhere to a balanced routine. Performance of virtually any standing, non-machine based exercise will heavily involve the core musculature, particularly the posterior muscles that antagonize the rectus abdominis (Schoenfeld, 2010, Lehman, 2005). It also should be noted that the average person tends to have weak abdominals (Morris et al. 2006), so they could very well benefit from performing spinal flexion exercises.

To sum up, there is no convincing evidence that performing crunches as part of a total body resistance training routine will have any negative effects on posture.

Do crunches lead to additional dysfunction, such as breathing dysfunction and glute dysfunction?

If crunches did indeed pull down on the ribcage and induce kyphosis, then one could speculate that breathing and glute functioning could be compromised. However, as just mentioned, this likely isn't the case.

Anecdotally, hundreds of thousands of athletes in the past few decades have achieved terrific success in spite of doing crunches. If crunches did lead to shortening of the abdominals, given that a majority of people including athletes seem to display an anterior pelvic tilt in their daily posture, one could argue that it would be wise to perform crunches to pull up on the pelvis, which could theoretically decrease anterior pelvic tilt and lead to a more neutral lumbo-pelvic posture.

Are crunches a nonfunctional exercise?

Whenever one questions the implication that crunches are as dangerous as wake boarding in a tsunami, anti-crunchers quickly counter with something like, "Who cares if they're dangerous or not? Crunches aren't functional! They're a short range movement performed while lying on your back. They can't possibly transfer to anything."

As our mentor Mel Siff aptly stated:

"Functional training is any form of training that improves any relevant biomotor ability that does not come at the detriment to other biomotor abilities."

Anti-crunchers will point out that the crunch solely involves bodyweight and therefore isn't heavy enough to transfer to high-force or high-velocity movement. This is absurd, as it's very easy to hold a dumbbell at the upper chest to increase the exercise's intensity. You also can perform a kneeling rope crunch using a cable apparatus to increase training intensity.

What carries over best to functional activity depends on the task. Here's a chart that should help you determine optimal transfer of training.

Biomotor Ability Examples Exercise Category Examples
Trunk Rotary Stability, Strength and Power Throwing a football or baseball, throwing a discuss, swinging a bat, throwing a left hook Rotational exercises and anti-rotation exercises Woodchops, landmines, Pallof presses, cable or band chops and lifts
Trunk Lateral Bending Stability, Strength and Power Stiff-arming, posting up, landing a jab Lateral flexion exercises and anti-lateral flexion exercises Side bends, side planks, suitcase carries, cable or band side bends
Trunk Flexion Stability, Strength and Power Sitting up from a bench, bar gymnastics exercises, bracing for a punch to the midsection, resisting being pushed rearward as in sumo wrestling, throwing a soccer ball overhead Flexion exercises and anti-extension exercises Crunches, sit ups, hanging leg raises, planks, ab wheel rollouts, bodysaws
Trunk Extension Stability, Strength and Power Carrying heavy loads, picking up stuff off the ground, staying upright in the clinch Extension exercises and anti-flexion exercises Squats, deadlifts, back extensions, reverse hypers, farmer?s walks, Zercher carries

Basically, stability exercises appear to be better for stabilization tasks as well as tasks that require proper inner-core unit functioning, while strengthening exercises appear to be better for dynamic tasks and hypertrophy.

As you can see, an exercise like a weighted crunch could transfer quite well to a myriad of functional tasks, and therefore shouldn't be maligned for its applicability to functional or sports performance.

It's been stated by many practitioners that the discs don't heal. This is misleading. It's true that they're poorly vascularized and struggle to receive adequate nutrition, and it's true that disc tissue doesn't heal rapidly. Proteoglycan turnover may take 500 days (Urban et al. 1978) and collagen turnover may take longer (Adams and Hutton 1982).

Hence, the discs' rate of remodeling lags behind that of other skeletal tissues (Maroudas et al. 1975; Skrzypiec et al. 2007).

However, much evidence of disc-healing exists. Common sense would dictate that discs do heal, otherwise anyone that suffered a disc injury would never get better. We'd all just get progressively worse until we could no longer move.

Based on epidemiological studies, it's clear that there's an optimal window of spinal loading that is somewhere in between bed rest and overactivity (Videman et al., 1990).

A healthy balance has been shown to occur with spinal compression (Hutton et al. 1998; Lotz et al. 2002; Walsh and Lotz 2004; Wuertz et al. 2009; MacLean et al. 2005) and spinal rotation (Chan et al. 2011). Positive aspects of spinal bending have been shown to occur in the discs as well (Lotz et al. 2008; Court et al. 2001).

Furthermore, 16 different studies indicate that the spinal discs can repair and remodel themselves. While several papers reviewed the topic (Lotz 2004; Stokes and Iatridis; Adams and Dolan 1997; Porter 1987), others demonstrated that flexion damages can heal (Court et al. 2007), compression damages can heal (Lai et al. 2008; Korecki et al. 2008; MacLean et al. 2008; Hee et al. 2011), prolapses can reverse (Scannell and McGill 2009), herniations can improve (Girard et al. 2004; Wood et al. 1997), the outer annulus can strengthen (Skrzpiec et al. 2007), collagen within the disc can remodel to become stronger (Brickley-Parsons and Glimcher, 1984), and vertebrae, ligaments, and strengthen to resist loading (Porter et al. 1989; Adams and Dolan 1996).

Every time you move your spine you cause micro-damage to the tissues, which triggers both anabolic and catabolic processes. Ideally, you want to limit the amount of damage, as healing from larger-scale damage usually leaves the disc biomechanically inferior.

For example, annular damage is repaired by granulation tissue and the scar never regains its normal lamellar architecture (Hampton et al. 1989). End plate injuries heal with cartilaginous tissue (Cinotti et al. 2005; Holm et al. 2004), fibrocartilage replaces nucleus material (Kim et al. 2005), and healing of harmed tissue is often replaced by a thin layer of weaker fibrous tissue (Fazzalari et al. 2001).


The study of biomechanics is unique because you must not only consider forces and stresses on human tissues, you must also consider adaptive remodeling. The ideal situation in programming doesn't avoid stress; it keeps the body in eustress while avoiding distress, which ensures that anabolic agents inside of tissues exert more work than catabolic agents inside of tissues to promote full repair, recovery, and strengthening.

We hope that we've provided you some food for thought and encouraged you to rely on logic rather than emotion in decision-making involving exercise safety and program design. An effective practitioner weighs all available evidence and makes appropriate conclusions in a dispassionate manner, without adhering to rigidly held beliefs.

We believe that future research will help hone in on the safety of the crunch exercise and determine proper dose responses, but this research needs to be conducted on living humans and involve pre and post-MRI results with a training intervention that ensures proper crunch technique.

Bret Contreras is considered by many to be the world’s foremost expert on glute training. He has turbo-charged the fitness industry by introducing effective new exercises and training methods for optimal glute development. Follow Bret Contreras on Twitter
Brad Schoenfeld, PhD, is an internationally renowned fitness expert and widely regarded as one of the leading authorities on muscle development and fat loss. He is a lifetime drug-free bodybuilder, and has won numerous natural bodybuilding titles. Follow Brad Schoenfeld on Twitter