No large-scale study has measured what fraction of the general population can perform a full split, so there is no reliable single number to quote. The honest answer is that only a small minority of untrained adults can drop into a split cold, while a much larger group could get there with consistent training, and some people will never reach the floor no matter how much they stretch. The reason no clean statistic exists is that “doing the splits” depends on an interplay of joint structure, connective tissue properties, nervous-system tolerance, sex, age, and training history. Understanding those factors tells you far more than a percentage ever could.
Why No One Has Measured This Directly
Flexibility research tends to measure specific joint ranges of motion in degrees, not whether someone can perform a particular party trick. A front split requires roughly 130 degrees of hip flexion on the front leg and about 30 degrees of hip extension on the back leg, along with adequate hamstring and hip-flexor length. A side (or “middle”) split demands around 130 to 140 degrees of combined hip abduction. Researchers measure these individual ranges with goniometers or motion-capture systems in clinical and athletic settings, but they rarely ask a random sample of adults to attempt a full split on camera and then report the success rate. The closest proxy data come from studies on generalized joint hypermobility and hip range of motion in various populations.
What Hypermobility Data Tell Us
Generalized joint hypermobility, the condition where multiple joints move well beyond average ranges, gives a rough ceiling estimate. A systematic review of hypermobility in children, adolescents, and young adults found that prevalence ranged from about 9% to 36% depending on the ethnic group studied and the scoring threshold used.1PubMed Central. Hypermobility prevalence, measurements, and outcomes in childhood, adolescence, and emerging adulthood: a systematic review That wide range reflects real biological variation across populations, not just measurement quirks. People of South Asian and West African descent, for example, tend to score higher on hypermobility scales than people of Northern European descent.
But hypermobility is measured at the elbows, knees, wrists, fingers, and spine. It does not directly test the hip splits position. You can be hypermobile at the wrist without having unusual hip range. Still, the data confirm that a meaningful fraction of the population is born with connective tissue that permits extreme joint motion, and those individuals have a built-in head start toward splits-level flexibility even without specific training.
The Sex Gap in Hip Flexibility
Women are, on average, more flexible at the hip than men. A study of asymptomatic college freshman athletes found that males had significantly less hip flexion and internal rotation than females across multiple clinical tests.2PubMed Central. Gender-Dependent Differences in Hip Range of Motion and Impingement Testing in Asymptomatic College Freshman Athletes This gap shows up even in a young, athletic population where everyone is reasonably active. In an untrained general population, the difference tends to be even more pronounced.
Several factors drive this. Hormonal differences affect connective tissue compliance: estrogen promotes collagen turnover in a way that keeps ligaments and tendons somewhat more pliable. Pelvic shape differs too. Women generally have a wider, shallower acetabulum (the hip socket), which permits greater range before bone contacts bone. These structural advantages mean that, all else being equal, a larger proportion of women than men could train to a full split within a given timeframe. Estimates from flexibility coaches and dance instructors commonly suggest that the majority of healthy women under 40 could achieve a front split with six months to a year of dedicated stretching, while men of the same age and health might need considerably longer or hit a harder structural ceiling.
How Age Changes the Picture
If you could measure splits capability across all age groups, the curve would decline steeply after the teenage years. Children are famously bendy, and many kids under 10 can do a split with little practice. By middle age, resting muscle stiffness increases substantially. A review of age-related changes in the muscle-tendon complex identified several culprits: alterations in the extracellular matrix surrounding muscle fibers, increased tendon rigidity, fascial thickening, and changes in how the nervous system activates muscles, particularly greater co-contraction of opposing muscle groups.3PubMed Central. Increase of resting muscle stiffness, a less considered component of age-related skeletal muscle impairment
Research on older women specifically confirmed that maximum passive stretch range and the torque the muscle-tendon unit can tolerate both decline with age.4Physical Therapy. Influence of Age on Length and Passive Elastic Stiffness Characteristics of the Calf Muscle-Tendon Unit of Women That study examined the calf, not the hip, but the underlying tissue changes are systemic. Your hamstrings, hip flexors, and adductors undergo the same stiffening process. This means the percentage of people who could do a split if they tried drops significantly with every decade past adolescence, not because training stops working but because the starting point keeps getting farther from the goal.
Bone Structure Sets a Hard Limit
One of the most underappreciated facts about splits is that some people are structurally blocked by their skeleton. The shape of the femoral neck, the depth and angle of the hip socket, and the orientation of the femoral head all vary from person to person. In a deep socket or one that angles forward (a condition sometimes called pincer-type morphology), the thighbone simply runs into the rim of the socket before the leg reaches full abduction or extension. No amount of stretching can change bone geometry.
You can get a rough sense of whether bony impingement limits you by comparing your passive range (someone else gently pushing your leg) with your active range (you moving it yourself). If both hit a hard stop at about the same point and you feel a pinching or blocking sensation deep in the joint rather than a muscle stretch, that is likely bone on bone. Estimates vary, but experienced flexibility coaches report that a meaningful minority of adults, perhaps 10 to 15 percent, have hip anatomy that physically prevents a full side split regardless of soft-tissue flexibility. The front split is more forgiving because it demands range in two different directions (flexion and extension) rather than extreme abduction on both sides simultaneously.
What Limits You Might Not Be What You Think
People who have tried and failed to get their splits often blame tight muscles. The reality is more nuanced. Research on stretch tolerance suggests that the fascial layers surrounding muscle play a larger role than the muscle tissue itself. A study on the quadriceps found that the deep fascia’s taut surface appeared to drive the sensation of stretch tolerance more than the underlying muscle belly did.5Journal of Anatomy. Stretch tolerance and elastic passive reaction of the quadriceps femoris seem to depend more on the fascia profundis taut surfaces than on the underlying stretched muscle
In practical terms, this means that the “I can’t stretch any further” feeling is partly a nervous-system signal originating from fascial stretch receptors, not necessarily an indication that the muscle fiber itself is at its physical limit. Consistent stretching appears to recalibrate that tolerance threshold over weeks and months, allowing the nervous system to permit a greater range before sounding the alarm. This is why gains in flexibility often feel sudden and nonlinear: you plateau, then seemingly overnight your body lets you go a few degrees deeper. The tissue did not change overnight. Your brain’s threat assessment did.
Can You Train Your Way to a Split?
The encouraging news is that stretching training produces large gains in range of motion for most people. A systematic review and meta-analysis of chronic stretching programs found that regular stretching produced large improvements in range of motion overall. When people stopped stretching, they retained a moderate portion of those gains even after a detraining period, though some of the improvement did fade.6PubMed Central. Detraining Effects Following Chronic Stretching Training on Range of Motion: A Systematic Review and Meta-Analysis That means flexibility is genuinely trainable and not purely a genetic lottery, but it also means you need to keep stretching to hold onto your gains.
For the specific muscles involved in a split, both static stretching (holding a position for 30 to 60 seconds) and PNF stretching (a contract-relax technique) are effective. A review comparing the two approaches for hip flexion, the primary motion in a front split, found that both methods increased range of motion and neither was clearly superior to the other.7PubMed. The Effectiveness of PNF Versus Static Stretching on Increasing Hip-Flexion Range of Motion For the side split, which demands hip adductor length, a study on recreational football players did find a significant advantage for PNF stretching on abduction range and groin flexibility.8Journal of Ecophysiology and Occupational Health. Efficacy of Static Stretching Versus Proprioceptive Neuromuscular Facilitation (PNF) Stretching on Hip Adductor Flexibility in Recreational Football Players The practical takeaway: if you are specifically chasing a middle split, adding PNF work for the inner thigh is probably worth the extra effort.
Realistic timelines depend enormously on where you start. Someone in their 20s with average flexibility and no structural limitations might reach a front split in three to six months of daily stretching. A middle split typically takes longer, often a year or more. Someone starting in their 40s or 50s with below-average flexibility might need two years or more for a front split, and a full middle split may not be in the cards at all if hip socket geometry intervenes.
Dancers and Gymnasts as a Case Study
The populations where splits are most common are dancers, gymnasts, martial artists, and figure skaters. These athletes offer a useful natural experiment. Dancers, for instance, develop dramatically increased flexibility throughout the spine, pelvis, and hips through years of training. Yet when researchers tested them on standard hypermobility criteria that exclude the trained motions, like elbow hyperextension, dancers were about as flexible as the general population.9PubMed Central. The Hyperflexible Hip Managing Hip Pain in the Dancer and Gymnast This finding is revealing. It suggests that dancers’ extreme hip flexibility comes mostly from training-induced adaptation of the soft tissues and nervous system around the hip, not from an inherently more elastic connective tissue type throughout their bodies.
That said, selection bias is real. People who gravitate toward dance and gymnastics as children often have above-average baseline flexibility, and those who progress to elite levels tend to be the ones whose anatomy cooperates. So while training clearly does the heavy lifting, the starting material matters too. A child with deep hip sockets and short femoral necks might love dance class but never achieve the oversplits that her teammates manage.
Risks of Forcing the Issue
Pushing aggressively toward a split carries real injury potential, especially for the hip joint. Capsular laxity and hip hypermobility are recognized causes of labral tears, the painful injuries to the cartilage ring lining the hip socket.10PubMed Central. A comprehensive review of hip labral tears Dancers and gymnasts have elevated rates of these tears precisely because their training demands repetitive extreme ranges. For a recreational adult chasing splits as a fitness goal, the risk is lower simply because the loading is less frequent, but it is not zero.
Hamstring strains are the other common casualty. The long head of the biceps femoris, which runs from your sitting bone to below your knee, is under enormous stretch during a front split. If you load it aggressively before it has adapted, a partial tear at or near the tendon attachment is possible. The typical warning sign is a sharp, sudden pain high in the back of the thigh, distinct from the slow-building discomfort of a good stretch.
A sensible approach involves:
- Warming up first: Stretching cold tissue is riskier than stretching warm tissue. Five to ten minutes of light cardio before a flexibility session makes a measurable difference in how far your muscles can safely go.
- Progressive overload: Add depth gradually over weeks, not within a single session. Forcing a new range in one sitting is how strains happen.
- Listening to the type of discomfort: A broad, pulling sensation along the belly of a muscle is generally safe stretch. Pinching in the front of the hip joint, sharp pain near a tendon, or anything that feels like it is in the joint rather than the muscle is a signal to back off.
- Balancing flexibility with strength: A joint that can reach extreme ranges but has no muscular control in those ranges is an unstable joint. Strength work through the new range, such as slow leg lifts near your end range, helps protect the tissues.
Putting Rough Numbers on It
Given everything above, here is a reasonable back-of-the-envelope picture for the general adult population. Among untrained adults who have never specifically worked on flexibility, the fraction who can sit into a full front split right now is likely in the low single digits, perhaps 2 to 5 percent, weighted heavily toward younger women and people with naturally hypermobile connective tissue. For a middle split, the untrained number is even smaller, probably under 2 percent, because the anatomical demands are stricter.
Among people who actively train flexibility, like yoga practitioners, dancers, martial artists, and dedicated stretchers, the proportion who can do a front split is much higher. In a serious adult ballet class, for example, nearly everyone can do a front split and most can do a middle split. But those populations are intensely self-selected and have spent years building that range. Among the broader group of people who do yoga a few times a week, front-split capability varies widely and a full middle split remains uncommon.
The theoretical ceiling, the percentage of healthy adults who could achieve a front split if they trained consistently for a year or more, is probably somewhere around 60 to 80 percent. The remainder would be limited by hip socket depth, femoral morphology, or connective tissue conditions that respond poorly to stretching. For a middle split the theoretical ceiling is lower, maybe 40 to 60 percent, because bony impingement of the femur against the acetabular rim in abduction is more common than people realize.
Why the Side Split Is So Much Harder
People often wonder why the front split comes more easily than the middle split, and the answer is mostly architectural. In a front split, one hip flexes while the other extends. These are the directions your hip is designed to move most freely, since walking and running involve exactly this forward-and-back pendulum. The hip socket is relatively unconstrained in the sagittal plane.
A middle split requires both hips to abduct toward 90 degrees or beyond simultaneously. The lateral rim of the acetabulum is often more prominent and the femoral neck can impinge against it at high abduction angles. You also have to lengthen the adductor muscles on both sides at once, and the adductor magnus, the largest of the inner-thigh muscles, has a substantial tendinous component that resists elongation more stubbornly than a fleshy muscle belly would. People with a retroverted (backward-tilted) hip socket tend to have an easier time with the middle split, while those with anteverted (forward-tilted) hips find the front split more accessible. You do not get to choose your version, and imaging would be the only way to know for sure which type you have.
Flexibility You Lose When You Stop Stretching
A common frustration is achieving a split and then losing it during a break from training. The detraining data confirm that this is real but not catastrophic. After people stopped a stretching program, they retained a moderate amount of the range they had gained, though they did lose some of the peak improvement.6PubMed Central. Detraining Effects Following Chronic Stretching Training on Range of Motion: A Systematic Review and Meta-Analysis In practical terms, if you trained to a full split and then stopped stretching entirely for a few months, you would likely come up a few inches short of the floor but still be significantly more flexible than when you started. Maintaining a split with minimal effort, say two or three brief stretching sessions per week, appears to be far easier than building the range in the first place.
This use-it-or-lose-it dynamic is another reason why the “what percent can do a split” question resists a clean answer. Flexibility is not a fixed trait like height. It is a moving target that responds to what you do with your body week to week. Someone who was a competitive gymnast at 16 and hasn’t stretched in 20 years probably can’t do a split today, yet might regain it faster than someone starting from scratch thanks to residual tissue adaptation and motor memory. The percentage of people who can do a split at any given moment depends entirely on who has been stretching lately.