4.17 PNF, PAILs, and RAILs

Try this before you read another word

Sit on the floor with one leg straight out and reach towards your foot. Stop when you feel a proper stretch, and make a mental note of exactly where your fingertips get to.

‌

Now push your heel down into the floor, as hard as is comfortable, for seven seconds. Relax completely. Reach again.

‌

Further, isn't it? Almost everyone reaches noticeably further on the second attempt, often by a surprising margin. And the whole thing took about ten seconds.

‌

Now the strange part. Nothing in your hamstring structurally changed in those ten seconds. No tissue lengthened, nothing "released".

‌

So what did change? That question is the key to this whole article, and the answer is not the one most textbooks give.

‌

‌

What you just did has a grand name: proprioceptive neuromuscular facilitation, or PNF. Stretch, contract the stretched muscle without moving, relax, sink deeper.

‌

That little loop, in various arrangements, is what PNF boils down to in a stretching context. The name comes from a much broader rehabilitation system developed in the 1940s and 1950s for patients with neurological conditions, in which stretching was a minor component.

‌

Most of that system stayed in the clinic. The contract-relax family of techniques is the part that escaped into general fitness use, and it's the part this article is about.

‌

The whole story in six lines

• PNF generally produces the largest acute range increase of any stretching method. For range right now, nothing beats it.
• The classic "autogenic inhibition" explanation is not supported by the electromyography evidence. Altered stretch tolerance is the better-supported mechanism.
• The variants differ in one important way: whether you actively move into the new range or get placed there. Active variants produce range you can actually use.
• progressive angular isometric loading (PAILs) and regressive angular isometric loading (RAILs) add a progressive contraction into the new position and then out of it, with the aim of converting passive gain into active control. That aim is a reasonable extension of the evidence for active PNF variants rather than something the branded protocol has been tested for on its own.
• Over long training programmes, the advantage of PNF over static stretching narrows considerably.
• You don't need a partner. A strap, a wall or the floor can provide the resistance for nearly everything.

‌

If you only remember one thing, make it the last two lines of that list. This method is brilliant for right now, ordinary for the long run, and entirely doable on your own.

‌

Anatomy of one repetition

Look back at Figure 1, because every named variant is just a rearrangement of those five steps. Get the base sequence into your head and everything else in this article becomes much easier to keep straight.

‌

Step one: ease into a comfortable stretch and hold it for around ten to twenty seconds. Mild sensation only. This is your baseline, the mark you'll measure progress against.

‌

Step two: contract the muscle you're stretching against something that won't move. A partner, a strap, the floor. Five to ten seconds at moderate effort, and the joint doesn't move at all, because this is an isometric.

‌

Step three: relax. Properly. Give it two to three full seconds, because incomplete relaxation is the most common reason the technique fails to produce a deeper position.

‌

Step four: move into the new position, usually by a noticeable margin, and hold for twenty to thirty seconds.

‌

Step five: repeat the cycle two to four times. Each round yields a smaller increment, and beyond three or four repetitions the returns are minimal while the fatigue cost keeps climbing.

‌

The variants, minus the alphabet soup

The named variants confuse everyone, because the differences are small and the names are nearly identical. So ignore the acronyms for a moment and ask one question: who produces the movement into the new range?

‌

‌

In hold-relax and standard contract-relax, the answer is "not you". After the contraction, a partner or a strap moves your limb into the deeper position. You're passive during that phase, so the range you gain is passive range.

‌

In contract-relax-antagonist-contract, mercifully shortened to CRAC, you do the moving. After the contraction, you actively contract the opposing muscle and pull yourself into the new range.

‌

That one change matters more than it sounds. Because you produced the movement, the range is active from the outset. It's yours from the first second.

‌

PAILs and RAILs, drawn from the Functional Range Conditioning system, extend the same idea (Spina, 2013). After reaching the new position, you progressively contract into the stretch, then progressively contract in the opposite direction to actively hold the new end position.

‌

Think of it as CRAC with a work ethic: structured, higher-effort, and explicitly obsessed with owning the range you've just acquired.

‌

• Hold-relax An isometric contraction of the stretched muscle, then the partner moves the limb deeper. The simplest and most passive version.
• Contract-relax A contraction against resistance, then a partner or strap moves you deeper. The standard technique in most textbooks, and still largely passive.
• CRAC A contraction of the stretched muscle, then the opposing muscle actively pulls you into the new range. Produces active range.
• PAILs and RAILs A progressive isometric contraction into the stretch (PAILs), followed by a contraction in the opposite direction to hold the new position (RAILs). The most demanding, and the most focused on active control.

‌

Here's the concrete version, using the hamstring reach you did earlier. Contract-relax: push the heel down, relax, and let a strap or partner take you deeper. CRAC: push, relax, then pull yourself deeper with the opposing muscles instead of being pushed.

‌

PAILs and RAILs: stay exactly where you are and do all the pushing and pulling without moving an inch. Same stretch, three levels of ownership.

‌

Why fuss over active versus passive? Because sport happens actively. A kick requires producing force at the acquired range, not merely reaching it, which is exactly why kicking disciplines have leaned on contract-relax methods for decades.

‌

‌

Why the textbook explanation is wrong

Ask why the floor trick works and almost every textbook tells the same story: autogenic inhibition. Contracting the muscle stimulates its Golgi tendon organs, the story goes, which reflexively inhibit that muscle, so it relaxes more deeply afterwards and allows more range.

‌

It's a neat story, and it's still taught almost everywhere. It just doesn't survive contact with the measuring equipment.

‌

‌

When researchers actually measured muscle activity during and after the contract-relax sequence, they generally found the opposite of what the theory predicts. Electromyographic activity in the stretched muscle is often higher after the contraction, not lower (Chalmers, 2004; Sharman et al., 2006).

‌

The timing is wrong too. The Golgi tendon organ reflex is far too short-lived to explain an effect that persists through a twenty or thirty second stretch (Chalmers, 2004).

‌

Reciprocal inhibition, where contracting the opposing muscle reduces activity in the stretched one, fares somewhat better. But the effect is small and inconsistent, it may contribute mainly in the CRAC and PAILs variants, and it cannot account for the size of the acute change (Sharman et al., 2006).

‌

The best-supported explanation is the same one behind ordinary stretching gains: altered stretch tolerance. The contraction changes how much stretch sensation your nervous system is willing to accept, which lets you move further at the same tissue tension (Weppler & Magnusson, 2010; Mitchell, 2009).

‌

This fits the facts neatly. Tissue stiffness changes very little acutely, the effect is largest immediately, and it decays over the following hour (Magnusson, 1996; Konrad et al., 2017).

‌

In other words: your hamstring didn't get longer on that second reach. Your nervous system simply agreed to let you go further.

‌

This does not make PNF less useful. It works, reliably, and it produces the largest acute range increase of any stretching method. It simply works for a different reason than most people are taught, and that reason has implications: If the effect is tolerance-based, then it needs to be repeated regularly and paired with strength work to become durable.

‌

Sit with that for a second, because it changes how you should train. Tolerance-based gains are rented, not owned. They need regular repetition, and they need pairing with strength work, before they become something you keep.

‌

It also tells you when to deploy the trick. The effect is biggest immediately and fades over the following hour, so use it close to the session that needs the position, not hours before.

‌

‌

What the evidence actually says

How big is the advantage, really?

PNF is frequently presented as clearly superior to static stretching. The evidence supports that for acute effects, and supports it much less for long-term outcomes.

‌

‌

Acutely, PNF generally produces the largest single-session range increase of the three main methods, and this is a consistent finding (Hindle, 2012; Thomas, 2018).

‌

So if you need more range right now, for a session or an assessment, it's the best tool available. No argument.

‌

Zoom out to a training programme of six weeks or more, though, and the picture changes. Meta-analyses comparing stretching modalities generally find that PNF, static and other methods produce broadly similar long-term range improvements (Thomas, 2018).

‌

Any long-term advantage to PNF is small and inconsistent across studies. Worth knowing before you build a routine around partner work.

‌

And here's why that matters practically: PNF is considerably more effortful than static stretching, frequently requires a partner, and is harder to sustain as a daily habit.

‌

If the long-term outcome is similar, the method that actually gets done consistently is the better one for most people. Boring, but true.

‌

Every session of PNF asks for effort, focus, and often another human. Static stretching asks for almost nothing, which is precisely why it's easier to keep showing up for.

‌

• Acute range gain: PNF clearly leads.
• Long-term range gain over 6 or more weeks: broadly comparable across methods.
• Adherence and convenience: static and solo methods lead comfortably.
• Active range produced: the active variants lead, which is a genuine and underrated advantage (Hindle, 2012). The direct evidence here is for CRAC-style contract-relax-antagonist-contract work; PAILs and RAILs inherit it by resemblance rather than by separate testing.

‌

One-line recap: PNF wins today, everything roughly draws over months, and the active variants win at the thing nobody measures — range you control.

‌

‌

No partner? No problem

Most PNF instruction assumes a partner, and that assumption is the single largest barrier to using it. Good news: almost every application can be adapted.

‌

A strap, a wall, the floor, or your own opposite limb can play the part of the resistance. Five worked examples:

‌

1. Hamstrings. Lie on your back with a strap around the foot. Push the heel down against the strap for 7 seconds, relax, then draw the leg closer.
2. Hip flexors. Half-kneeling with the rear foot against a wall. Push the rear knee down into the floor for 7 seconds, relax, then shift the hips forward.
3. Calves. Standing with the forefoot on a step. Push the ball of the foot down for 7 seconds, relax, then allow the heel to drop further.
4. Chest and shoulders. Forearm on a doorframe. Press into the frame for 7 seconds, relax, then rotate further away.
5. Adductors. Seated with the soles of your feet together. Press the knees up against your own hands for 7 seconds, relax, then allow the knees to descend.

‌

How hard should you push? Gentler than you'd think. Research comparing contraction intensities has generally found that moderate efforts, in the region of twenty to sixty per cent of maximum, produce similar range outcomes to maximal contractions, with considerably less risk and fatigue (Feland & Marin, 2004).

‌

That finding is quietly brilliant, because maximal contractions at end range are exactly where partner-assisted PNF has occasionally caused injuries. Half effort buys you the range without the risk.

‌

A cue worth stealing: push into the strap, not through it.

‌

‌

PAILs and RAILs in proper detail

This variant deserves its own treatment, because it directly attacks the problem with all passive stretching: gaining range you cannot use.

‌

One caveat before the detail, because it is easy to lose. What has been studied directly is PNF in general — the contract-relax and CRAC families, and the mechanisms behind them — together with end-range isometric loading. PAILs and RAILs are a branded application of those principles, and no trial has compared them head to head with conventional CRAC or loaded end-range work. The evidence for them is therefore indirect, and any claim that the branded version is superior is currently unsupported.

‌

The sequence begins conventionally. Reach a passive end position and hold it for around two minutes, which is noticeably longer than most PNF protocols.

‌

Then, over roughly ten to twenty seconds, progressively build a contraction into the stretch, pushing towards the direction that would take you out of the position. That's the PAILs contraction. Build it gradually to a high effort and hold for ten to twenty seconds.

‌

Then, without leaving the position, switch sides. Contract in the opposite direction, actively trying to pull yourself deeper into the stretch using the opposing muscles. That's the RAILs contraction.

‌

It's usually much harder than it sounds. The first few attempts often end in cramp, which is a normal response to asking a muscle to contract at a very short length.

‌

1. Reach the passive end position and hold for 90 to 120 seconds.
2. Build a progressive contraction into the stretch over 10 seconds, then hold at high effort for 10 to 20 seconds.
3. Without changing position, switch to contracting in the opposite direction, actively pulling deeper, for 10 to 20 seconds.
4. Relax briefly, then explore whether a slightly greater passive range is now available.
5. Repeat two to three times. This is demanding work, and more than three cycles per position is rarely productive.

‌

Mapped onto the floor stretch from the top of this article: settle into the reach for two minutes, then drive the heel down into the floor, building gradually. That's your PAILs.

‌

Then, staying exactly where you are, switch to actively pulling your torso deeper into the reach. That's your RAILs, and yes, it will probably try to cramp.

‌

About the cramping: it's common in the RAILs phase, generally harmless, and it usually stops occurring after a few sessions as the muscle adapts to producing force in a shortened position.

‌

If it's severe or persistent, reduce the effort level and build more gradually. Cramping is normal; pain is not.

‌

Safety, and when to leave it alone

Be clear about what you're handling here. PNF is the highest-intensity stretching method, and it carries correspondingly more risk than the alternatives, particularly in the partner-assisted forms (Sharman et al., 2006).

‌

The central problem with partner work is that the person applying the force cannot feel what the person receiving it feels. A partner pushing to what looks reasonable can exceed your tissue tolerance before you've had a chance to say anything.

‌

Most PNF injuries reported in practice come from exactly this situation. So set the rules before anyone touches your leg.

‌

• Prefer strap and wall variants over partner variants where possible; that way you control the force.
• If you do use a partner, you control the depth verbally, and the partner never pushes beyond what's asked.
• Use moderate contraction intensities of twenty to sixty per cent rather than maximal efforts.
• Warm up thoroughly first. PNF on cold tissue is a poor idea.
• Avoid it entirely on an acutely strained muscle, and follow clinical guidance on timing after injury.

‌

And four situations where the answer is "not today":

‌

• Not appropriate acutely after a muscle strain. High-tension work on healing tissue can impede recovery. Timing is a clinical decision.
• Caution with joint instability. PNF produces the largest acute range increase available, which is the opposite of what an unstable joint needs.
• Caution in hypermobility. Adding further passive range to someone who already has too much is counterproductive. The PAILs and RAILs variant is more defensible here, because it emphasises active control.
• Stop for neurological symptoms. Tingling, numbness or radiating pain is not a stretch sensation. It means stop, not push on.

‌

One-line recap: the person who can feel the stretch is the only person qualified to decide how deep it goes.

‌

Where it fits in your week

Long-term outcomes are broadly comparable across methods. So the case for PNF rests on exactly two things: the acute effect, and the active control produced by the CRAC family, of which PAILs and RAILs are a branded version rather than a separately validated one.

‌

Use the acute effect when you need range now. If today's session requires a position you can't currently reach, a couple of contract-relax cycles will produce that position, with less of a force penalty than an equivalent duration of static stretching would.

‌

Use the active variants when the goal is durable, usable range. PAILs and RAILs, or CRAC, performed twice weekly on one or two priority positions, combined with end-range strength work, is a coherent programme.

‌

That combination, active PNF plus loading, is where the method genuinely earns its place.

‌

1. Once or twice weekly per position, not daily. It's demanding, and daily use produces fatigue without extra benefit.
2. Two to three cycles per position. Beyond that, the increments are negligible.
3. Always warm up first.
4. Always follow with active work in the new range, whether that's the RAILs contraction or separate end-range strength work.
5. Don't use it immediately before power or strength work, since the acute force decrement applies here as it does to static stretching (Behm, 2016).

‌

And the practical caveat worth repeating until it's annoying: the best stretching method is the one that gets done.

‌

If a partner-assisted protocol means the work happens twice in six weeks, a simpler solo method performed five days a week will outperform it comfortably. That's two exposures against thirty over the same six weeks, and no acute advantage survives arithmetic like that.

‌

Your ready-made protocols

The solo contract-relax protocol

Two positions, twice weekly, after training or as a separate warm session. About twelve minutes, start to finish.

‌

• Warm up for five minutes with light movement.
• Position one: reach the passive end position and hold 20 seconds.
• Contract against a strap or the floor at 30 to 50 per cent effort for 7 seconds.
• Relax completely for 3 seconds, then move deeper and hold 25 seconds.
• Repeat the contract-relax cycle twice more, then move to position two.

‌

The PAILs and RAILs protocol

One or two positions, twice weekly. More demanding and considerably longer per position, so treat it like training rather than a wind-down.

‌

• Warm up thoroughly. This is not a cold-tissue method.
• Reach the passive end position and hold for 90 to 120 seconds.
• Build a progressive contraction into the stretch over 10 seconds, hold at high effort for 15 seconds.
• Switch to contracting in the opposite direction, actively pulling deeper, for 15 seconds.
• Relax, explore the new range, and repeat twice.
• Expect cramping in the second phase for the first few sessions.

‌

Making the gains stick

Here's the bit everyone skips. Because the mechanism is largely tolerance-based, the acquired range decays unless it's used and loaded.

‌

This step is not optional, and it's where most PNF programmes quietly fail. Win the range, then keep it:

‌

1. Immediately after the session, perform two or three end-range isometric holds in the new position.
2. Within the same week, include a loaded exercise that passes through the new range.
3. Reassess passive and active range every four weeks.
4. If passive range rose but active range did not, reduce the PNF volume and increase the strength work.
5. Once the target range is reached, maintain it with one session per week.

‌

Do that, and the ten-second party trick from the start of this article stops being a trick. It becomes range you keep.

‌

Sport applications

Martial arts and kicking disciplines have used contract-relax methods for decades, and the active variants suit them well because kicks require producing force at the acquired range rather than merely reaching it.

‌

Gymnastics and dance use PNF extensively, usually in a partner-assisted form within a supervised setting, and always alongside substantial active strength work at the same ranges.

‌

Rehabilitation settings use PNF in its original and broader form, of which stretching is only a small part. Timing after injury is a clinical decision rather than a training one.

‌

Field-sport athletes generally get more from loaded end-range work than from PNF, since their limitation is usually control at range rather than passive range itself.

‌

Swimmers and overhead athletes should be cautious. Many already have generous passive shoulder range, and the acute range increase produced by PNF is the opposite of what a loose shoulder needs.

‌

Common mistakes

• Believing the autogenic inhibition explanation. The electromyography evidence does not support it. The technique works through altered stretch tolerance, which means the gains need repetition and loading to become durable.
• Using maximal contractions. Moderate efforts of twenty to sixty per cent produce similar range outcomes with considerably less risk. Maximal contractions at end range are where partner-assisted PNF has caused injuries.
• Letting a partner decide the depth. The partner cannot feel what the athlete feels. The athlete should control the depth verbally, and the partner should never push beyond what is asked.
• Relaxing incompletely between phases. The most common reason the technique fails to produce a deeper position. Take a full two to three seconds and release the contraction entirely.
• Doing it daily. It is demanding and the returns do not scale with frequency. Once or twice weekly per position is appropriate.
• Performing more than three cycles. Each cycle yields a smaller increment. Beyond three or four the gains are negligible and the fatigue cost is not.
• Never loading the new range. Because the mechanism is tolerance-based, the range decays unless it is used. Pair every session with end-range isometrics or loaded work.
• Using it before power or strength work. The acute force decrement applies here as it does to prolonged static stretching. Place it after training or in a separate session.

‌

Coaching cues

• "Push into the strap, not through it."
• "Relax completely before you move."
• "Half effort is enough."
• "Pull yourself deeper rather than being pushed."
• "Own the new range before you leave the position."
• "Cramping in RAILs is normal; pain is not."
• "Warm first, always."
• "Twice a week, not every day."

‌

FAQs

What does PNF stand for?

Proprioceptive neuromuscular facilitation. It began as a broad rehabilitation approach developed in the 1940s and 1950s for patients with neurological conditions, in which stretching was a small component. What survives in general fitness use is the contract-relax family of stretching techniques.

‌

Is PNF better than static stretching?

For acute range gain in a single session, yes, consistently. For long-term range development over six weeks or more, the difference is small and inconsistent across studies. Since PNF is more effortful and often needs a partner, the simpler method that actually gets done regularly is usually the better practical choice.

‌

Why does contracting the muscle let me stretch further?

Not for the reason usually taught. The classic autogenic inhibition explanation, involving Golgi tendon organ reflexes, is not supported by the muscle activity measurements, which often show higher activity after the contraction rather than lower. The best-supported mechanism is altered stretch tolerance: The contraction changes how much stretch sensation your nervous system accepts.

‌

Can I do PNF alone?

Yes, for almost every position, using a strap, a wall, the floor or your opposite limb as the resistance. This is generally preferable to partner work because you control the force, and most reported PNF injuries come from a partner pushing further than the athlete intended.

‌

How hard should the contraction be?

Twenty to sixty per cent of maximum is sufficient. Research comparing contraction intensities has generally found similar range outcomes from moderate and maximal efforts, and moderate efforts carry considerably less risk and fatigue cost.

‌

What are PAILs and RAILs?

Progressive angular isometric loading and regressive angular isometric loading, from the Functional Range Conditioning system. After reaching an end position, you build a progressive contraction pushing out of the stretch, then switch to contracting in the opposite direction to actively pull yourself deeper. It is a structured, higher-effort development of the contract-relax-antagonist-contract technique, with more emphasis on actively owning the new range. The supporting evidence is indirect: it comes from studies of PNF and end-range isometrics in general, not from head-to-head trials of the branded protocol.

‌

Why do I cramp during the RAILs contraction?

Because you are asking a muscle to contract hard at a very short length, which is mechanically difficult and unfamiliar. It is common in the first few sessions and generally harmless, and it usually stops as the muscle adapts. If it is severe, reduce the effort and build more gradually.

‌

How often should I do PNF?

Once or twice weekly per position, with two to three cycles each. It is demanding, the returns do not improve with daily use, and the acute force decrement means it should not be placed immediately before power or strength work. Follow every session with active work in the new range, or the gain will decay.

‌

Recommended videos

Each video below was chosen because it covers a specific part of this article in more depth than text alone allows.

‌

‌

‌

‌

‌

‌

Related reading on FitXplor

• 4.11 Static Stretching
• 4.12 Active and Passive Mobility
• 4.14 End-Range Strength
• 4.15 Isometric Mobility
• 4.16 Controlled Articular Rotations (CARs)
• 4.9 Understanding Range of Motion
• 4.13 Loaded Mobility
• 1.5 The Nervous System and Athletic Performance

‌

References

Sharman, M. J., Cresswell, A. G., & Riek, S. (2006). Proprioceptive neuromuscular facilitation stretching: Mechanisms and clinical implications. Sports Medicine, 36(11), 929–939. Read on PubMed

Chalmers, G. (2004). Re-examination of the possible role of Golgi tendon organ and muscle spindle reflexes in proprioceptive neuromuscular facilitation muscle stretching. Sports Biomechanics.

Mitchell, U.H. et al. (2009). Acute stretch perception alteration contributes to the success of the PNF contract-relax stretch. Journal of Sport Rehabilitation.

Hindle, K.B. et al. (2012). Proprioceptive neuromuscular facilitation (PNF): its mechanisms and effects on range of motion and muscular function. Journal of Human Kinetics.

Konrad, A. et al. (2017). The accumulated effects of foam rolling combined with stretching on range of motion and physical performance. Journal of Sports Science and Medicine.

Feland, J.B., Marin, H.N. (2004). Effect of submaximal contraction intensity in contract-relax PNF stretching. British Journal of Sports Medicine.

Behm, D.G. et al. (2016). Acute effects of muscle stretching on physical performance, range of motion, and injury incidence. Applied Physiology, Nutrition, and Metabolism.

Thomas, E. et al. (2018). The relation between stretching typology and stretching duration: the effects on range of motion. International Journal of Sports Medicine.

Weppler, C.H., Magnusson, S.P. (2010). Increasing muscle extensibility: a matter of increasing length or modifying sensation? Physical Therapy.

Magnusson, S.P. et al. (1996). Mechanical and physical responses to stretching with and without preisometric contraction in human skeletal muscle. Archives of Physical Medicine and Rehabilitation.

Konrad, A., Stafilidis, S., Tilp, M. (2017). Effects of acute static, ballistic, and PNF stretching exercise on the muscle and tendon tissue properties. Scandinavian Journal of Medicine & Science in Sports.

Spina, A. (2013). Functional Range Conditioning system materials. Functional Anatomy Seminars.