The two-finger rule at the noseband: what the taper gauge and pressure sensors show

A rule without a fixed measure

The two-finger rule is the best-known rule of thumb in riding: two fingers should fit under the closed noseband. It sounds precise and is not. Neither the thickness of a finger, nor the place where it is laid, nor the direction in which it is pushed in is fixed anywhere. Two slim fingers and two strong fingers describe the same space in very different millimetres. It is exactly this vagueness that has led several federations to move to a standardised taper gauge.

How little the rule is understood in common is shown by an international survey with more than three thousand responses. Of those who check the fit at all, about 62 per cent tested on the bridge of the nose, a good 21 per cent under the chin and about 10 per cent at the side of the head. In a Canadian survey only a little over half of those asked regarded the bridge of the nose as the correct site. Two people can therefore both consider the same noseband compliant while measuring different things.

Why is a finger not enough as a measuring instrument?

Because it is not a measure but a body part. Finger thickness, the point of contact and the direction of insertion change the result, and none of these three quantities is defined in the rule itself. A taper gauge fixes shape and dimensions and only thereby makes two checks comparable at all. About the pressure that builds up under the strap, it too says nothing.

Two gauges are in circulation, one from research and one from the international rulebook, and they are more alike than their reputation suggests. A reply published in 2026 measured both: the research gauge has the larger cross-section at around six square centimetres, while the rulebook gauge is one millimetre taller. Referred to the circumference of the closed noseband, that gives 3.2 against 3.4 centimetres. In practice, the authors write, the one gauge therefore leaves no more room than the other.

How tightly nosebands are actually fastened

The most frequently cited field measurement dates from 2017. A single investigator measured the space under the noseband in 750 horses at competitions in Ireland, England and Belgium, always with the same tapered gauge and always in the middle of the bridge of the nose. In 44 per cent of the horses no space remained, in 7 per cent half a finger, in 23 per cent one finger and in 19 per cent one and a half. Two fingers were reached by 7 per cent, and a single noseband lay above that. The tightest fitting was found in eventing, followed by the dressage arena.

Other countries produce other figures. In Canada nineteen officiating judges measured 551 horses at 32 events with the same type of gauge: 71 per cent reached the value for two fingers, a further 19 per cent at least the value for one. In the Netherlands, shortly after a two-finger requirement was introduced, 59 per cent of riders met it. Such figures are not comparable: they come from different years, from different people and in part from announced equipment checks.

What happens when the gauge becomes compulsory is shown by a Czech survey of 238 dressage combinations, from the lowest class up to Grand Prix. With the official gauge in the hands of the steward, every horse but two was compliant. The spread thereby disappeared, and the actual question of the study could no longer be answered: without differences in fitting there is no link to behaviour left to examine. What explained the conflict behaviours was the level of the class.

What pressure sensors measure under the strap

For a few years now it has not only been the space that is measured, but the pressure. In a British measurement series, eight upper-level dressage horses trotted in a straight line with four types of noseband and five tightness levels each. On the bridge of the nose the mean pressure was 1.6 kilopascals at two fingers of space and rose through 2.9 and 3.1 to 4.2 and finally 6.4 kilopascals with the strap closed. The rise became meaningful only below one finger; between two fingers and one and a half the analysis found no difference.

A second measurement series on eight horses linked pressure to movement for the first time. Peak pressure was around 26 kilopascals at two fingers, about half as much again at one finger and more than four times as high with no space at all. At the same time stride length shortened, by about 6 per cent at one finger and about 11 per cent with the strap closed. A shorter stride is a signal, not a demonstrated harm: what follows from it the study did not investigate.

The one published figure that looks like a limit comes from a dead horse. On a cadaver head the noseband was tightened in eight steps, with sensors on the nasal bone, in the middle of the bridge of the nose and inside the mouth. Below the equivalent of 1.4 fingers the values jumped, up to 403 kilopascals at the lateral nasal bone. The authors conclude from this that the old two-finger rule should be retained. That head, however, lacked muscle tone, blood flow, breathing, a bit and movement.

The dispute is about the site of measurement

Perhaps the most important observation from these measurements appears in no rulebook. In the trot measurement the pressure at the mandible was higher than on the bridge of the nose for all four designs, about 9.1 against 2.8 kilopascals for the plain cavesson. A second study found the same ranking in the standing and chewing horse, but states expressly that the difference was not confirmed throughout. The site that is checked is therefore the less loaded one.

Out of this came the proposal to measure at the side of the head. In 100 horses, three designs presented in random order at five tightness levels were measured with a calliper at three lateral sites. The authors write that a lateral point could complement the measurement on the bridge of the nose, not replace it. Their most awkward result for the two-finger rule lies elsewhere: between two fingers and one and a half this method too found no difference at two of the three sites.

The objection came in 2026 as a commentary signed by eight specialists. Its core: at the upper jaw there is so much soft tissue that a gap measured there can understate the actual tightness. Moreover, the research gauge comes with a validated method that prescribes the bridge of the nose; changing the site without revalidating the method makes the figures incomparable. The reply counters that older requirements themselves referred to one finger between the cheek and the strap. Each side accuses the other of conflicts of interest.

What has been tested on the living horse

Pressure is a surrogate measure. Whether a horse suffers under it has to be tested on the animal itself, and there are few, small trials for that. In a randomised trial with twelve horses, four settings were compared, from the open strap to a fitting with no space at all. Heart rate, heart rate variability and eye temperature shifted towards a stress response only at the tightest setting. Chewing was already reduced at one finger. Most striking was what happened after unbridling: yawning, swallowing and licking increased markedly, which the authors read as a pent-up need being made good.

A German trial with sixteen horses examined the upper airways, randomised, blinded and with an endoscope during ridden work. The measured pharyngeal diameter did not change meaningfully: a narrowing of the airway in the strict sense is therefore not demonstrated. Other findings in the pharynx did change, among them the amount of secretion and a more pronounced collapse of the pharyngeal wall. The score of a pain ethogram was higher with the tight strap, but stayed below the threshold at which pain is assumed in every individual horse.

Against the simple reading stands a study from the other camp. Eight dressage horses took up a treat and chewed it at every tightness level, including with no space at all, generating around 100 newtons and more than 40 kilopascals, without eye temperature or blink rate rising. The authors conclude that these markers indicated no burden during chewing. The order, however, always ran from loose to tight, and both markers are regarded as insensitive.

Findings in the mouth, the skin and the bone

Whether tight nosebands leave traces was examined by a Danish survey of 3,143 combinations after competition. A looser fitting went together with fewer findings in the mouth there. The second half of the same result is rarely quoted: where the noseband was absent altogether, the risk of lesions at the corners of the mouth was more than twice as high as with the loosest fitting. The relationship is therefore not a straight line. Blood or lesions at the corners of the mouth were found in 9.2 per cent of the horses, and the findings increased with the level of the class.

For the bone, a study of 144 riding horses is readily cited: 37.5 per cent showed changes of the nasal bones on radiographs. The authors state expressly, however, that their work demonstrates no causal link with any item of equipment. A second finding is instructive: by palpation the investigators found bony deposits in more than 80 per cent of the horses, on radiographs in fewer than 10 per cent.

The everyday observations come from the international survey: 18.6 per cent of respondents reported at least one physical or behavioural abnormality, most often bald patches under the strap. These are self-reports without veterinary examination.

Design of noseband and bridle

Not every design behaves the same way. In the Irish field measurement the combined noseband with a flash strap was the most common design and at the same time markedly tighter than three other designs, among them the plain cavesson and the Micklem bridle. In the trot measurement the plain and the Swedish cavesson did not differ from each other in pressure, whereas the values on the bridge of the nose did differ between designs overall.

A common short circuit concerns the double bridle. In a comparison of eleven dressage horses, in which the noseband was set to two fingers in both bridles, the pressure under the strap did not differ between snaffle and double bridle. With the double bridle the pressure at the poll was higher, which fits its greater weight, and rein tension was lower than with the snaffle. In the Czech competition survey, by contrast, horses in a double bridle showed proportionally more conflict behaviour. The two findings do not necessarily contradict each other: they measure different things in different horses.

What is settled and what remains open

Three statements hold. First, pressure under the noseband rises steeply below about one finger of space, and several independent measurements have found this in agreement. Second, neither the taper gauge nor the calliper reliably distinguishes two fingers from one and a half. Third, every compliance figure hangs on the instrument, on the site of measurement and on whether the check was announced.

Equally clear is what is missing. None of the pressure studies measured pain or welfare alongside; the authors state this themselves. There is no pressure value tested in living, ridden horses above which harm arises. The only published figure of that kind comes from a single cadaver head. A trial that set out specifically to find a stress marker found a difference in the blood only between the loosest and the least favourable condition, and calls its own set-up a mild model.

A 2026 review draws from this not a rule of measurement but a rule of decision: as long as the evidence is missing, a practice should not count as harmless merely because its harm has not been demonstrated. That is a stance, not a study result, and it is to be read as such.

Field measurements of noseband tightness: instrument, site and blind spot

Original analysis

Limitations and uncertainty

• The instrumented measurement series rest on eight to sixteen horses per trial, mostly well-schooled dressage horses that are used to their equipment. No Swiss data on noseband fitting or pressure could be found in this search.
• None of the pressure and locomotion studies measured pain or welfare alongside; the authors state this themselves. There is no pressure value tested in living, ridden horses above which harm arises.
• The only published numerical value, the equivalent of 1.4 fingers, comes from a single cadaver head without muscle tone, blood flow, breathing, a bit or movement. It holds for that skull and that design, not in general.
• Two of the central measurement studies were challenged in 2025 by collective letters in the same journal, both author groups replied, and conflicts of interest are alleged against both camps. The dispute about the site of measurement is open.
• Compliance rates from different countries are not comparable: instrument, site, categories and whether the check was announced all differ, and several measurements were taken during official equipment checks.
• Of the 24 studies used here, only three reach the level of a randomised trial, a systematic review or a meta-analysis. The field consists largely of cross-sectional surveys, controlled measurement series without randomisation and position pieces; the editorial requirement of at least half at that level cannot be met here.

Open questions

• From what pressure and over what duration does harm arise in the living, ridden horse? No threshold has so far been tested against clinical endpoints.
• Is the mandible the site that carries the relevant load, and what would change if the check were made there instead of on the bridge of the nose?
• Does introducing a taper gauge change fitting for good, or only on competition days under supervision?
• Do tight nosebands explain part of the radiographic findings on the nasal bones, or do those have quite different causes?

Frequently asked questions

So how tight may the noseband be?

There is no scientifically established limit. What can be measured is that pressure rises steeply below about one finger of space, and that between two fingers and one and a half neither the taper gauge nor the calliper found a difference. The only published numerical value, the equivalent of 1.4 fingers, comes from a single cadaver head without muscle tone, breathing or movement, and does not transfer to the living horse without further ado. What applies at competition is set out in the rulebook of the relevant federation; what suits an individual horse belongs in a conversation with a qualified person.

Is a flash noseband more harmful than a plain cavesson?

What is established is that it sits tighter in practice, not that it does harm. In the field measurement of 750 competition horses, the combined noseband with a flash strap was the most common design and at the same time markedly tighter than three other designs. In the pressure measurement on the ridden horse, the designs differed in the values recorded on the bridge of the nose, without any one of them being identified as harmful. None of these studies measured pain or welfare alongside. The design therefore mainly changes how easily a very tight fitting is reached.

Where should the two fingers actually be placed?

That is precisely the point in dispute, and it is not settled. In an international survey about 62 per cent checked on the bridge of the nose, a good 21 per cent under the chin and about 10 per cent at the side of the head; in Canada only a little over half of those asked regarded the bridge of the nose as the right place. The research gauge comes with a validated method that prescribes the bridge of the nose. A 2025 proposal to measure laterally as well was criticised in 2026 by eight specialists, on the grounds that the soft tissue there understates tightness. For competition, only the requirement of the federation counts.

Can you tell from the horse that the noseband is too tight?

Only to a limited extent, and not in the graded way one would expect. In a trial with twelve horses, changes in heart rate and eye temperature appeared only with the strap fully closed, while chewing was already reduced from one finger onwards. Clearer still was what followed unbridling: yawning, swallowing and licking increased abruptly. In a second trial with sixteen ridden horses, the score of a pain ethogram was higher with the tight strap, while staying below the usual threshold in every horse. A horse showing no obvious signs is therefore no proof that the fitting is right.

Is it true that tight nosebands cause bone damage?

That is not demonstrated. In a study of 144 riding horses, 37.5 per cent showed changes of the nasal bones on radiographs, but the authors themselves state that their work demonstrates no causal link with equipment: details of the tack used and a comparison group were missing. The cadaver experiment does show very high local pressures with a tight fitting, but on a dead head and without any follow-up over time. The two together give a plausible hypothesis, not proof. It is also notable that palpation in the same study found far more bony deposits than the radiographs did.

Would it be better to leave the noseband off altogether?

The evidence does not support that conclusion. In the Danish survey of 3,143 competition combinations, a looser fitting did go together with fewer findings in the mouth, but the complete absence of a noseband went with a more than twice as high risk of lesions at the corners of the mouth. The relationship is therefore not a straight line, and leaving it off is not the continuation of loosening it. The same survey also found no difference between horses with and without a bit. This is a snapshot taken after competition, from which no cause can be derived.

Sources

1. Doherty O, Casey V, McGreevy P, Arkins S. Noseband Use in Equestrian Sports - An International Study. PLoS One, 2017 (Cross-sectional study | Horse)DOI 10.1371/journal.pone.0169060
   Descriptive survey of 750 competition horses in Ireland, England and Belgium: 44 per cent left no finger of space and only about 7 per cent reached two fingers, with the authors stating expressly that data on the short- and long-term consequences are lacking and that they therefore demonstrate no harm.
2. Doherty O, Conway T, Conway R, Murray G. An Objective Measure of Noseband Tightness and Its Measurement Using a Novel Digital Tightness Gauge. PLoS One, 2017 (Laboratory study | Horse)DOI 10.1371/journal.pone.0168996
   Metrological work on the development of a tension gauge: for the same gauge settings the normal force on the bridge of the nose ranged from 7 to 95 newtons, but laterally only from 1 to 28 newtons, showing that an identical setting means very different forces depending on where it is measured.
3. Fenner K, Yoon S, White P, Starling M. The Effect of Noseband Tightening on Horses' Behavior, Eye Temperature, and Cardiac Responses. PLoS One, 2016 (Randomised trial | Horse)DOI 10.1371/journal.pone.0154179
   In twelve horses randomly allocated to four settings, a rise in heart rate, a fall in heart rate variability and a rise in eye temperature appeared only with the strap fully closed, while chewing was already reduced from one finger onwards and a marked catch-up behaviour followed unbridling.
4. MacKechnie-Guire R, Murray R, Williams JM, Nixon J. Noseband type and tightness level affect pressure on the horse's face at trot. Equine Veterinary Journal, 2025 (Controlled trial | Horse)DOI 10.1111/evj.14420
   In eight dressage horses ridden at trot the pressure at the mandible consistently exceeded that on the bridge of the nose, and the rise on the bridge of the nose became meaningful only below one finger of space, with no difference between two fingers and one and a half; behaviour and physiology were not recorded.
5. Clayton HM, Murray R, Williams JM, Walker V. Facial pressure beneath a cavesson noseband adjusted to different tightness levels during standing and chewing. Equine Veterinary Journal, 2025 (Controlled trial | Horse)DOI 10.1111/evj.14451
   Eight dressage horses took and chewed a reward at all five tightness levels including the closed strap, generating around 100 newtons and more than 40 kilopascals, without any rise in eye temperature or blink rate, which is why the authors see no burden for these markers.
6. Doherty O, Conway R, McGreevy P. Using an Equine Cadaver Head to Investigate Associations Between Sub-Noseband Space, Noseband Tension, and Sub-Noseband Pressure at Three Locations. Animals (Basel), 2025 (Laboratory study | Horse)DOI 10.3390/ani15142141
   On a single cadaver head the pressures jumped below the equivalent of 1.4 fingers, reaching 403 kilopascals at the nasal bone, which is why the authors expressly wish to retain the inherited two-finger rule.
7. Hopkins E, Whitrod S, Marlin D, Blake R. Tight nosebands apply high pressures on the horses' face and alter stride kinematics. Journal of Equine Veterinary Science, 2025 (Controlled trial | Horse)DOI 10.1016/j.jevs.2025.105654
   In eight horses the peak pressure rose from around 26 kilopascals at two fingers to around 41 at one finger and around 116 with no space at all, while stride length fell by about 6 and then 11 per cent; neither pain nor welfare was measured.
8. MacKechnie-Guire R, Clayton H, Williams J, Marlin D. Measuring Noseband Tightness on the Lateral Aspect of the Horse's Face. Animals (Basel), 2025 (Controlled trial | Horse)DOI 10.3390/ani15040537
   In 100 horses the lateral measurement with a calliper found no difference between two fingers and one and a half at the nasal and the maxillary site, and the authors expressly propose the lateral point as a complement to, not a replacement for, the measurement on the bridge of the nose.
9. Henshall C, McGreevy P, Shea G, Doherty O. Commentary on MacKechnie-Guire et al. Measuring Noseband Tightness on the Lateral Aspect of the Horse's Face.2015,, 537. Animals (Basel), 2026 (Other | Horse)DOI 10.3390/ani16030412
   Argumentative commentary without new data: at the upper jaw the soft tissue could make the actual tightness appear less than it is, and changing the site of measurement without revalidating the method would make the figures incomparable; five of the eight signatories are connected with the organisation that distributes the research gauge.
10. MacKechnie-Guire R, Clayton H, Williams J, Marlin D. Reply to Henshall et al. Commentary on "MacKechnie-Guire et al. Measuring Noseband Tightness on the Lateral Aspect of the Horse's Face.2025,, 537". Animals (Basel), 2026 (Other | Horse)DOI 10.3390/ani16142131
    Reply without new experimental data: the two gauges increase the circumference of the strap by 3.2 against 3.4 centimetres and leave practically the same amount of room, older requirements themselves referred to one finger at the side, and the authors accuse the opposing side of an undeclared conflict of interest.
11. Fialova S, Kuritkova D, Sobotkova E. Stress Responses in Dressage Horses: Insights from FEI Noseband Measurements Across National Competition Levels. Animals (Basel), 2026 (Cross-sectional study | Horse)DOI 10.3390/ani16030518
    With the official gauge, 236 of 238 dressage combinations were compliant, so that the missing spread made any link with behaviour impossible to test; what did explain the findings was the level of the class, and horses in a double bridle showed proportionally more conflict behaviour.
12. Visser EK, Kuypers MMF, Stam JSM, Riedstra B. Practice of Noseband Use and Intentions Towards Behavioural Change in Dutch Equestrians. Animals (Basel), 2019 (Cross-sectional study | Horse)DOI 10.3390/ani9121131
    Shortly after a two-finger requirement was introduced, only 59 per cent of riders met it, and the belief that the rule serves welfare ranged from 38 to 89 per cent depending on the group, which is why the authors call for objective and transparent checking rather than a rule alone.
13. Weller D, Franklin S, Shea G, White P. The Reported Use of Nosebands in Racing and Equestrian Pursuits. Animals (Basel), 2020 (Cross-sectional study | Horse)DOI 10.3390/ani10050776
    In a survey with 3,040 responses, 62.1 per cent checked the fit on the bridge of the nose, 21.5 per cent under the chin and 10.4 per cent at the side, and 18.6 per cent reported at least one abnormality, most often bald patches under the strap; all the information is self-reported.
14. Merkies K, Copelin C, Small N, Young J. Noseband Fit: Measurements and Perceptions of Canadian Equestrians. Animals (Basel), 2022 (Cross-sectional study | Horse)DOI 10.3390/ani12192685
    Of 551 horses measured at 32 Canadian competitions, 71 per cent reached the value for two fingers and a further 19 per cent the value for one, while only 51.5 per cent of those asked regarded the bridge of the nose as the correct site, which is why the authors call for education first.
15. Uldahl M, Clayton HM. Lesions associated with the use of bits, nosebands, spurs and whips in Danish competition horses. Equine Veterinary Journal, 2019 (Cross-sectional study | Horse)DOI 10.1111/evj.12827
    In 3,143 combinations examined after competition, a looser fitting went together with fewer findings in the mouth, but the complete absence of a noseband went with a more than twice as high risk of lesions at the corners of the mouth, which occurred overall in 9.2 per cent of the horses.
16. Perez-Manrique L, Leon-Perez K, Zamora-Sanchez E, Davies S. Prevalence and Distribution of Lesions in the Nasal Bones and Mandibles of a Sample of 144 Riding Horses. Animals (Basel), 2020 (Cross-sectional study | Horse)DOI 10.3390/ani10091661
    In 144 riding horses, 37.5 per cent showed radiographic changes of the nasal bones, but the authors state expressly that their work demonstrates no causal link with any item of equipment; palpation moreover found far more bony deposits than the radiographs did.
17. Scholler D, Wittenberg J, Zablotski Y, May A. Do tight nosebands have an effect on the upper airways of horses?. Veterinary Medicine and Science, 2024 (Randomised trial | Horse)DOI 10.1002/vms3.1478
    In a randomised, blinded trial with sixteen ridden horses the measured pharyngeal diameter remained unchanged, while the amount of secretion, the lateral deviation of the epiglottic folds and collapse of the pharyngeal wall increased with the tight strap and the pain ethogram score rose, without exceeding the threshold in any horse.
18. Scholler D, Zablotski Y, May A. Evaluation of Substance P as a New Stress Parameter in Horses in a Stress Model Involving Four Different Stress Levels. Animals (Basel), 2023 (Controlled trial | Horse)DOI 10.3390/ani13071142
    Substance P proved unusable as a stress marker in the horse; serum cortisol did rise across the four levels but differed only between the loose strap and the combination of tight strap and endoscope, and the authors themselves describe their set-up as a mild stress model.
19. MacKechnie-Guire R, Clayton H, Williams J, Marlin D. Comparison of Rein Forces and Pressure Beneath the Noseband and Headpiece of a Snaffle Bridle and a Double Bridle. Animals (Basel), 2025 (Randomised trial | Horse)DOI 10.3390/ani15071058
    In eleven dressage horses with the noseband set to two fingers, the pressure under the strap did not differ between snaffle and double bridle, while pressure and force at the poll were higher with the double bridle and rein tension lower; only the order of the two bridles within the same horse was randomised.
20. Copelin C, Merkies K. Riding with care: A review of factors that influence the welfare of the ridden horse and a case for the application of the precautionary principle in equestrian pursuits. Journal of Equine Veterinary Science, 2026 (Other | Horse)DOI 10.1016/j.jevs.2026.105801
    Narrative review without data of its own: it argues that a practice should not be treated as harmless for as long as its harmlessness has not been shown, and cites over-tightened nosebands as an example of equipment with unavoidable pressure.
21. Doherty O, Fenner K, Winther Christensen J, McLean A. Comments on MacKechnie-Guire et al. (2024): Noseband type and tightness level affect pressure on the horse's face at trot. Equine Veterinary Journal, 2025 (Other | Horse)DOI 10.1111/evj.14549
    Collective letter from seven specialists without data of its own; no abstract is held in the register, so that it serves here solely to establish that the trot measurement was formally challenged in the same journal.
22. Wilkins C, Christensen JW, Doherty O, Fenner K. Comments on Clayton et al. (2024): Facial pressure beneath a cavesson noseband adjusted to different tightness levels during standing and chewing. Equine Veterinary Journal, 2025 (Other | Horse)DOI 10.1111/evj.14548
    Collective letter from seven specialists without data of its own; no abstract is held in the register, so that it serves here solely to establish that the chewing study too was formally challenged in the same journal.
23. MacKechnie-Guire R, Murray R, Williams JM, Nixon J. Response to comments on: Noseband type and tightness level affect pressure on the horse's face at trot. Equine Veterinary Journal, 2025 (Other | Horse)DOI 10.1111/evj.70085
    Reply from the challenged author group, without data of its own; no abstract is held in the register, so that it serves here solely to establish that the criticism received a published response.
24. Clayton HM, Murray RC, Williams JM, Walker V. Response to comments on 'Facial pressure beneath a cavesson noseband adjusted to different tightness levels during standing and chewing'. Equine Veterinary Journal, 2025 (Other | Horse)DOI 10.1111/evj.70087
    Reply from the challenged author group, without data of its own; no abstract is held in the register, so that it serves here solely to establish that this criticism too received a published response.