Blake Leeper v. World Athletics

II. OVERVIEW OF THE CASE

3. This dispute relates to the applicability and interpretation of WA Technical Rule 6.3.4, which came into effect on 31 January 2020. Under the preceding and related provision, Rule 6.2, “[a]ny athlete” is precluded from “giving or receiving assistance from within the competition area during an event”. Rule 6.3.4, which defines “assistance”, disallows “[t]he use of any mechanical aid, unless on the balance of probabilities the use of an aid would not provide them with an overall competitive advantage over an athlete not using such aid”1.

4. By application dated 24 December 2020, Mr Leeper sought WA’s permission to compete at WA-sanctioned events, including the 2021 Tokyo Olympics and related qualification events, on RSPs that give him a height of 185.42 cm (6’1”)2. On 26 April 2021, a five-member Mechanical Aids Review (“MAR”) panel established by WA rejected Mr Leeper’s application. In a reasoned decision (the “MAR panel decision”), the MAR panel held that Mr Leeper’s chosen RSPs conferred upon him an “overall competitive advantage” because they enabled him to run “taller” than his “natural height”, i.e. the height he would have been if he had fully intact biological limbs. The MAR panel thus concluded that Mr Leeper’s proposed blades were accordingly “not allowed” under Rule 6.3.4.

5. In these proceedings, Mr Leeper appeals against the MAR panel decision to the Court of Arbitration for Sport (“CAS”). He seeks an award from the CAS reversing the MAR panel decision and declaring him eligible to compete in all WA-sanctioned events, including the 2020 Tokyo Olympics taking place in 2021, using his proposed RSPs.

1 Prior to its amendment on 31 January 2020, WA Technical Rule 6.3.4 imposed the burden of proving the absence of an

overall competitive advantage on the athlete (“The use of any mechanical aid, unless the athlete can establish on the balance of probabilities that the use of an aid would not provide him with an overall competitive advantage over an athlete not using such aid”). See infra at paras 10 et seq. 2 The RSPs in question are J-shaped category 3 Ottobock 1E90 Sprinter RSPs.

VI. JURISDICTION

62. Article R47 of the CAS Code provides as follows:

“An appeal against the decision of a federation, association or sports-related body may be filed with the CAS insofar as the statutes or regulations of the said body so provide or as the parties have concluded a specific arbitration agreement and insofar as the Appellant has exhausted the legal remedies available to him prior to the appeal, in accordance with the statutes or regulations of the said sports-related body”.

63. Rule 3.1 of the WA’s Disputes and Disciplinary Proceedings Rules provides in material part:

“This Rule 3 relates to any legal dispute of any kind whatsoever arising between World Athletics on the one hand and any Member, Area Association, athlete, athlete support personnel or other person who is subject to the Constitution and/or any of the Rules or Regulations on the other hand, in relation to the Constitution and/or any Rule or Regulation and/or any World Athletics decision or act or omission, howsoever arising, that is not covered by the dispute resolution provisions of the Constitution or any Rules or Regulations (each, a ‘Dispute’). Subject to, and in accordance with Article 84 of the Constitution, a Dispute shall be submitted to arbitration before the CAS (Ordinary Arbitration Division or Appeal Arbitration Division, depending on the circumstances of the case), to the exclusion of any other court or forum”.

64. Both Parties agree that the CAS has jurisdiction over this appeal. The Parties confirmed CAS’s jurisdiction by signing the Order of Procedure.

65. The Panel is, therefore, satisfied that it has jurisdiction over this appeal.

VII. ADMISSIBILITY

66. Article R49 of the CAS Code provides as follows:

“In the absence of a time limit set in the statutes or regulations of the federation, association or sports-related body concerned, or of a previous agreement, the time limit for appeal shall be twenty-one days from the receipt of the decision appealed against. After having consulted the parties, the Division President may refuse to entertain an appeal if it is manifestly late”.

67. The MAR panel decision was issued on 26 April 2021. The Statement of Appeal was filed four days later, on 30 April 2021.

68. WA accepts that Mr Leeper’s appeal was filed within the time-limit and is therefore admissible.

69. The Panel is, therefore, satisfied that this appeal is admissible.

VIII. APPLICABLE LAW

70. Article R58 of the CAS Code provides as follows:

“The Panel shall decide the dispute according to the applicable regulations and the rules of law chosen by the parties or, in the absence of such a choice, according to the law of the country in which the federation, association or sports-related body which has issued the challenged decision is domiciled or according to the rules of law, the application of which the Panel deems appropriate. In the latter case, the Panel shall give reasons for its decision”.

71. Article R28 of the CAS Code provides that “[t]he seat of CAS and of each Arbitration Panel (‘Panel’) is Lausanne, Switzerland”. Article 176 of the Swiss Private International Law Act (“PILA”) provides that it “appl[ies] to arbitral tribunals that have their seat in Switzerland”.

72. Article 187 of the Act provides that “the arbitral tribunal shall rule according to the law chosen by the parties”.

73. In both their oral and written submissions, the Parties have expressly referred to and relied upon the WA Constitution, WA Competition Rules, WA Technical Rules, and WA Disputes and Disciplinary Proceedings Rules. WA also relies on “general principles of law drawn from a comparative or common denominator reading of various domestic legal systems” and, subsidiarily, upon the law of Monaco.

74. The Panel concludes that the law applicable to the merits is the WA Constitution, the WA Competition Rules, WA Disputes and Disciplinary Proceedings Rules (all in force from 1 November 2019) and WA Technical Rules (in force from 1 November 2019 and amended on 31 January 2020). There is no need to resort to Monegasque law given that it does not have any bearing on the Panel’s conclusions and reasoning below.

75. Regarding the application of general principles of law, the Panel notes that prior CAS panels have held that, given the “transnational character of international sporting competitions” and “the global effects of the actions and omissions of international federations”, national and international sports federations must conform to “a set of unwritten legal principles – a sort of lex mercatoria for sports or, so to speak, a “lex ludica” or “lex sportiva” (CAS 98/200, para 156; CAS 2014/A/3776, para 269). Such general principles may include “general principles of law drawn from a comparative or common denominator reading of various legal systems” (CAS 98/200, para 156), such as “fairness”, “good faith”, “prohibition of arbitrary rules and measures”, “venire contra factum proprium”, “nonretroactivity of laws”, “the right to be heard”, and “proportionality” (CAS 98/200, para 158). The only exception to their application is if “such general principles were in conflict with some national or transnational public policy provision applicable to a given case” (CAS 2014/A/3776, para 269; CAS

Accordingly, the Panel considers that general principles of law are also applicable to the present dispute as “rules of law [whose application] the Panel deems appropriate” under Article R58 of the CAS Code.

76. As to matters of procedure, the procedural rules of the CAS Code are applicable (R27 of the CAS Code), supplemented if need be in a manner consistent with mandatory rules of Swiss law, given that these proceedings are seated in Switzerland by virtue of R28 of the CAS Code.

IX. STANDARD OF REVIEW

77. Article R57 of the CAS Code provides, in part:

“The Panel has full power to review the facts and the law. It may issue a new decision which replaces the decision challenged or annul the decision and refer the case back to the previous instance”.

78. WA submits that, by virtue of this provision, this Panel is not restricted to deciding whether the MAR panel decision under appeal is wrong. Instead, it envisages a full de novo hearing of the merits of the case. Mr Leeper does not dispute this.

79. Accordingly, in accordance with Article R57 of the CAS Code, the Panel shall conduct a de novo determination of Mr Leeper’s eligibility to compete on his current RSPs.

X. MERITS

80. On the basis of the Parties’ requests for relief and the submissions advanced before the Panel in writing and orally, the Panel considers it helpful to identify the following issues as arising for determination:

- What is the effect of the Leeper I award on these proceedings?

- What is the proper comparator for assessing whether Mr Leeper, through his use of RSPs, has or does not have an “advantage” over an able-bodied athlete not using RSPs?

- Is the height of RSPs used by double-amputee runners determinative of their athletic performance, such that height can be regulated under Rule 6.3.4?

- Does Mr Leeper suffer disadvantage by running on RSPs, such as to offset an advantage conferred by the height of his RSPs?

- Is Mr Leeper running above his natural height on his current RSPs?

- Is there a less-intrusive alternative to altogether disallowing Mr Leeper’s RSPs in WAsanctioned events?

A. The effect of the Leeper I award on these proceedings

81. The Parties disagree on whether the Leeper I panel’s award is res judicata in these proceedings. Mr Leeper claims that, under the applicable Swiss rules, the Leeper I panel’s findings of fact and reasoning have no res judicata effect in these proceedings as they concern new facts and evidence, i.e. the use of shorter RSPs and expert evidence resulting from the Dallas tests. Nevertheless, WA contends that the general principle of res judicata, also recognised by Monegasque law, applies to these proceedings, and by virtue of it, rulings made by the Leeper I panel are res judicata. Alternatively, WA claims that, even if Swiss law is to be regarded as applicable, this requires the Panel to give res judicata effect to the Leeper I panel’s findings of fact and reasoning.

82. Neither party denies that res judicata is a principle of Swiss (procedural) public policy, which must be respected by an arbitral tribunal seated in Switzerland on pain of annulment under Article 190(2)(e) PILA.

83. The Panel notes that, under Swiss law, res judicata has two aspects, which are interrelated but separate: a negative aspect (i.e. that a claim finally adjudicated cannot be relitigated) and a positive or affirmative aspect (i.e. that a final adjudication of a claim must be respected in subsequent litigation). In both instances, res judicata applies only when the “parties” and the “object of the dispute” (i.e. the claims and relevant factual circumstances) are identical to those in the previous proceedings.

84. Mr Leeper argues that the present claim is different from that in Leeper I because it concerns RSPs of a different height, and thus no res judicata effect is to be given to Leeper I. WA, for its part, does not contest that these proceedings concern RSPs of a different height. However, it argues that the present proceedings concern the same Parties, the interpretation of the same rule, the same arguments by the parties and substantively the same evidence. Therefore, WA submits, the reasoning and the rulings in the Leeper I award are res judicata in these proceedings. In the alternative, the Leeper I award carries substantial precedential value, according to CAS jurisprudence.

85. The Panel agrees with Mr Leeper that the present claim is different from that in Leeper I, involving as it does RSPs of a different height and being directed against a different technical panel decision below. The Leeper I award does not, therefore, have a negative res judicata effect in these proceedings, i.e. it cannot prevent the adjudication of the present claim.

86. As to the potential affirmative res judicata effect of the Leeper I award, the operative section of that award is not of itself dispositive in the present case because, again, the specific claim in Leeper I concerned different RSPs. Nevertheless, in order to arrive at its ultimate decision, the Leeper I panel made a number of intermediate findings, including regarding the interpretation of an earlier version of Rule 6.3.4 and the MASH rule. The question, therefore, is whether the construction of Rule 6.3.4 and the application of the MASH rule adopted in Leeper I are to be regarded as part of the claim that was disposed of in Leeper I, such that these holdings would be binding on the Parties in the present proceedings. If the object of Leeper I is taken as the permissibility or otherwise of RSPs of a specific height, and nothing more, then intermediate

rulings of the Leeper I panel are not covered by res judicata. By contrast, if the object of Leeper I is to be regarded as being (or at least as including) the proper construction of Rule 6.3.4 and the applicability of the MASH rule, then findings on these issues are binding.

87. As the Panel has had the benefit of the Parties’ pleadings and full argument on the matters in issue, the Panel has been able to form its own view on those matters. There is no dispute that, under CAS jurisprudence, the reasoning of the Leeper I panel is to be carefully considered. The Panel agrees with that reasoning so far as relevant in the present case, and sets out in its own reasons the basis of that agreement. Accordingly, it is unnecessary to consider further the application, or otherwise, of res judicata.

B. The proper construction and scope of application of Rule 6.3.4

88. It is common ground that Mr Leeper, by competing in athletics on the international level, is bound by WA Competition and Technical Rules7.

89. WA Technical Rule 6.3.4 provides, in relevant part:

“For the purpose of this Rule, the following examples shall be considered assistance, and are therefore not allowed:

The use of any mechanical aid, unless on the balance of probabilities the use of an aid would not provide them [sc. athletes using such aids] with an overall competitive advantage over an athlete not using such aid”.

90. Rule 6.3.4 prohibits “[t]he use of any mechanical aid” that provides an athlete with an “overall competitive advantage”. It is common ground between the Parties that RSPs constitute “mechanical aids” within the meaning of Rule 6.3.4. The Panel will therefore approach the case on this agreed basis.

91. The Panel must determine two issues regarding the interpretation and application of Rule 6.3.4. First, what the appropriate comparator is for assessing whether a non-able-bodied athlete has an “advantage” over another athlete (whether able-bodied or not). Secondly, whether the height of a double-amputee athlete’s RSPs affects their athletic performance and can accordingly be regulated under Rule 6.3.4.

a. The notion of “advantage” under Rule 6.3.4

92. Mr Leeper submits that Rule 6.3.4 requires a comparison to be made between the disabled athlete who wishes to use RSPs and his/her actual competitors – in Mr Leeper’s case, any able-bodied athlete of elite world-class level not using RSPs. The Rule does not require, Mr Leeper argues, a comparison between the athlete who wishes to use RSPs and the performance that the same athlete would be capable of achieving if she or he had intact biological legs.

7 Rule 100 of the WA Technical Rules provides that “[a]ll International Competitions, as defined in the Generally Applicable

Definitions, shall be held under the Rules of World Athletics”. WA’s Generally Applicable Definitions define “International Competitions” as, inter alia, “[c]competitions included in the World Athletic Series” and “[t]he Athletics programme of the Olympic Games”.

This, Mr Leeper contended, was consistent with the natural meaning of the language of the Rule. And no other interpretation is practicable, as it is impossible to know how an amputee athlete like Mr Leeper would perform if he had intact biological legs. In Mr Leeper’s case, the Rule requires a comparison between his height and the height at which elite, world-class ablebodied male athletes of African descent run in 400m events. Mr Leeper presented the height data of top-ranked Black sprinters and claimed that those athletes ran as tall or even taller than Mr Leeper. Based on this information, he argues that the height of his RSPs did not give him a competitive advantage over other athletes.

93. WA submits that the Rule has only one logical, principled and workable construction. The notion of “advantage” requires a comparison between Mr Leeper’s performance on his RSPs and the performance he would have been able to achieve if he had intact biological legs. This interpretation of the Rule flows directly from its purpose, namely, to protect the integrity and fairness of competition by preventing athletes from using mechanical aids that do more than simply to compensate for the disadvantage caused by their disability.

94. As noted, the Leeper I panel held at paragraph 310 of its award that:

“[T]he only logical, principled and workable construction of the Rule is one that, in the case of disabled athletes who use a mechanical aid to overcome a disability, requires a comparison to be undertaken between the athlete’s likely athletic performance when using the mechanical aid and their likely athletic performance had they not had the disability which necessitates the use of that aid”.

95. At the hearing, the Panel sought clarification from the Parties on whether Rule 6.3.4 applies to all athletes (able-bodied and disabled alike) or only to disabled athletes. The Parties advanced divergent positions. Mr Leeper argued that Rule 6.3.4 applies only to disabled athletes, because it prohibits the use of an “aid”, which is “something you need to compete”. Rule 6.3.4 must be distinguished from Rule 6.4.4, which is designed to prohibit “any technology or appliance” capable of providing an athlete with an advantage. WA disagreed, submitting that, although Rule 6.3.4 may more frequently apply to disabled athletes than to non-disabled ones, it applies to all athletes.

96. In this respect, the Leeper I panel held that “[o]n its face, the Rule applies equally to all athletes; it draws no distinction between able-bodied athletes and athletes with a disability”. This is in contrast to “the eligibility rules that were the subject of the challenge in Chand (which expressly applied only to female athletes) and in Semenya (which expressly applied only to athletes who were recognised at law as female or intersex)”.

97. The Panel agrees with the Leeper I panel that, while “the practical effect of the Rule is likely to be significantly greater for disabled athletes”, the plain language of the Rule does not distinguish between able-bodied athletes and disabled athletes. The implication of this is as obvious as it is important. One must be careful not to construe Rule 6.3.4 as permitting aids which compensate for an athlete’s anatomic features to the point of equiparating an athlete’s competitive potential with the top athlete in the relevant discipline. Rule 6.3.4 cannot have the effect of allowing for an “arms race” to alter natural limitations in athletics.

98. Accordingly, the Panel is unable to accept the view that one must take as the comparator the height of elite competitors not running on RSPs. In a discipline such as Mr Leeper’s, the anatomy of an athlete is a factor in performance. Anatomy is taken as an unalterable fact of nature: there is no form of equalisation for anatomical features that may affect an athlete’s performance. Leg length is one such feature (as discussed below). If the Panel were to accept Mr Leeper’s view, then Rule 6.3.4 would allow athletes, both disabled and able-bodied, to use mechanical aids that alter their anatomy to emulate their competitors’ innate anatomical features and thus compensate for limitations resulting from their own anatomical features.

99. The Panel is unable to accept a reading of Rule 6.3.4 that would lead to such an outcome. The Panel, therefore, concludes that, in applying Rule 6.3.4 to an athlete such as Mr Leeper (who must run on RSPs if he is to run at all), the notion of “advantage” must be assessed by reference to Mr Leeper’s natural height, i.e. his height had his legs been intact below the knee.

b. The effect of RSP height on competitive performance

100. The next question is whether an increase in the height of an amputee athlete’s RSPs beyond his/her natural height (i.e. one’s height had one had intact biological limbs) is capable of providing the athlete with a performance advantage, and can, therefore be regulated under Rule 6.3.4.

101. The Parties’ experts agree on the basic mechanics of running speed. They agree that running speed is equal to the distance travelled per unit of time and involves ground reaction forces and contact lengths (i.e. the forward distance the body travels while the foot is in contact with the ground).

102. However, the experts disagree on whether the length of a double-amputee athlete’s RSPs affects ground contact length, and, as a result, directly affects their competitive performance in sprint running.

103. Mr Leeper’s experts opined that WA’s experts are wrong to suggest that prosthetic length directly affects maximum speed. Increases in the height of RSPs do not increase ground contact length and running velocity, because there are other biomechanical factors (e.g. forces and step-frequencies) that change speed. Dr Herr stated during the hearing that “running is a dynamic and complex activity”. In the words of Dr Beck, the matter is “[n]ot as simple as shorter or longer ground contact length”.

104. In order to substantiate their view, Prof Grabowski and her team presented in their Expert Report of 26 March 2021 the results from the Taboga 2020 Study, which “is the only study ever published that systematically changed prosthetic height and studied its effect on the running biomechanics and top speed of athletes with bilateral transtibial amputations”. In that study, the authors measured the treadmill-running top speed of five athletes with bilateral transtibial amputations, each running on three different RSPs (different models, different stiffness levels, and different heights). They increased the RSP height used by one of the participants by up to 16 cm. That participant ran 1.1 m/s slower with a 10 cm increase, 1.8 m/s slower with a 12 cm increase, and 2.8 m/s slower with a 14 cm increase. Similarly, the authors increased RSP height for another

participant by up to 7 cm. That participant ran 1.4 m/s slower using 2 cm taller RSPs and ran 1.25 m/s faster using 3 cm shorter prostheses. Based on these results, Prof Grabowski and her colleagues concluded that there was no significant relationship between prosthetic height and maximum speed.

105. In addition, Dr Herr claimed that the notion that an increase in leg length increases speed has never been proven in a scientific, peer-reviewed study. It also lacks scientific vigour because it assumes that all other factors are held constant, while there are many other variables that can affect an amputee athlete’s performance if their RSPs are lengthened. (For instance, if the height of the RSPs is increased, then they become heavier, which slows down maximum velocity.) He said that WA’s experts have not taken such variables into account.

106. For its part, WA submits that height should be regulated under Rule 6.3.4 because longer legs are indeed advantageous to sprint-running performance, and such advantage can be quantified. According to Profs Weyand and Bundle, the scientific understanding of the mechanical basis of sprint running confirms that there is a direct relationship between limb length and maximum sprinting speed. They expressed the basis of this relationship in mathematical terms, explaining that the “most critical determinant” of the maximum sprinting speed that an athlete can attain (i.e. an athlete’s performance over a 400m race) is “the minimum time of foot-ground contact”. Since running speed is equal to the distance travelled per unit of time, a runner’s speed can be expressed as the distance their body travels while their foot is in contact with the ground (“ground contact length”) divided by the duration of that contact (“ground contact time”). The ground contact length of an athlete is a close function of the length of their legs: the longer a runner’s legs, the further forward the runner moves each time they make contact with the ground. For most runners, contact lengths are generally equal to or slightly longer than their leg lengths. Therefore, all things being equal, a lengthening of the leg results in a direct increase in ground contact length.

107. In their Expert Report of 5 April 2021, Prof Tweedy and Dr Connick opined that the data in the Taboga 2020 Paper do not prove that RSP height has no effect on maximum speed. They pointed out that the authors of that paper considered “a small sample size”, used “an inappropriate statistical technique”, and misinterpreted the results. The study also included confounding factors. For example, the subjects ran on RSPs of length and stiffness they were used to in only one of the 15 configurations tested. But an amputee athlete may take a while to familiarise (“acclimatise”) himself or herself with different height, models and configuration of the RSPs. Indeed, that was Mr Leeper’s justification in Leeper I for not being able to shorten his RSPs. If one removes this confounding factor of unfamiliarity and only looks at the maximum speed achieved by each subject when using her usual RSPs at the usual stiffness, then, of the three heights investigated (0, +2 cm and +4 cm), four out of five athletes achieved maximum speed when running at +2 cm or +4 cm. While cautioning that the evidence from these limited data “should be interpreted cautiously”, Prof Tweedy and Dr Connick concluded that “the direction of the effect is the opposite to that declared by the authors [of the Taboga 2020 Paper]”, tending to support the hypothesis that increased RSP length permits higher maximum velocity.

108. WA’s experts also testified that, in addition to the experimental data in the Taboga 2020 Paper, observational evidence also demonstrates that an increase in leg length increases maximum

speed. Prof Tweedy and Dr Connick examined the data of three of the eight bilateral transtibial amputee finalists from the 400m race at the Rio 2016 Paralympic Games after these athletes were required to decrease the lengths of their blades by several inches, under the MASH rule put in place in 2018. They noted that the maximum speed of these runners dropped by up to six seconds after reducing the height of their RSPs. Similarly, Profs Weyand and Bundle compiled height data for the best 45 individual performers over a 14-year period in Olympic running events over distances between 100m and 3,000m. Their data showed that the best individual performers in all sprint events “exceed the sex-specific height norms of the larger population”, meaning that “being taller and having longer legs clearly is advantageous for sprint running”. They added that the magnitude of the advantage derived from increased height and leg length “becomes larger as the duration of the sprint race increases”.

109. Given these experimental and observational data, Profs Weyand and Bundle examined whether a decrease in the height of Mr Leeper’s RSPs resulted in a decrease in his top speed. They noted that, during tests in Boulder in 2018, Prof Grabowski had measured Mr Leeper’s maximum speed on his 5 cm taller RSPs (giving Mr Leeper a leg length of 109 cm) at 11.4 m/s. Based on their mathematical formula, Profs Weyand and Bundle predicted that a 5 cm decrease in the height of Mr Leeper’s RSPs would lead to a drop in maximum speed of 0.5 m/s, to 10.9 m/s. They tested this hypothesis at tests in Dallas in 2021, which, they claimed, confirmed that Mr Leeper’s maximum speed was 10.9 m/s on his 5 cm shorter RSPs. In this way, they confirmed that leg length directly affects an athlete’s maximum speed.

110. WA’s experts claimed that, as a corollary, the height of RSPs should be controlled under Rule 6.3.4. This is because increasing the length of RSPs can give an athlete a longer stride length. Non-amputee runners, by contrast, are unable artificially to increase the length of their lower limbs in order to increase their stride length. As a result, the use of RSPs by a double-amputee runner such as Mr Leeper may confer an advantage that is both “unnatural” and “unfair” and should therefore be assessed under Rule 6.3.4. In that connection, Prof Weyand disagreed with Mr Leeper’s experts that the additional weight of lengthier RSPs will decrease speed. On the contrary, the light carbon-fibre material that RSPs are made of enables sprinters to reposition their limbs in 20% less time than a non-amputee runner, an enormous advantage for sprinting. Prof Tweedy and Dr Connick also questioned the existence of any peer-reviewed studies proving that mass affects top speed.

111. Mr Leeper’s experts, for their part, disagreed that non-amputee runners cannot change their running height. In their Expert Report of 26 March 2021, they submitted that, as biomechanics govern running performance, the effect of leg- or RSP length should be assessed when an athlete is running and not when they are standing. Non-amputee runners can lengthen their legs during running through plantar flexion, i.e. by pointing away their toes. By lengthening their legs in such a way, non-amputee runners can contact the ground earlier. Passive RSPs used by amputee runners cannot flex in the same way. To test this hypothesis, Prof Grabowski and her team experimentally compared the biological and prosthetic leg lengths of seven athletes with unilateral transtibial amputation. They recorded the biological and prosthetic leg length of each athlete at touchdown, midstance, and take-off. They concluded that non-amputee runners can lengthen their leg prior to ground contact by

approximately 6 cm. In addition, non-amputee athletes can race with permitted (“sanctioned”) shoes that have 2 cm-thick soles, increasing their overall running leg length.

112. Having considered the evidence before it, the Panel accepts the evidence of WA’s experts that, all things being equal, leg length is a factor that affects a runner’s maximum speed. The Panel considers persuasive the scientific evidence offered by WA’s experts to the effect that a direct relationship exists between leg length and running speed. It accepts that a significant component of a runner’s potential top speed is their contact length, which is determined by their leg length (see supra at para 106). The Panel also acknowledges that this direct relationship is reflected in the observational evidence put forward by WA’s experts. The population data presented by Profs Weyand and Bundle of the 45 best 400m sprinters over a 14-year period of Olympic running events show that the top performers (male and female) were taller and had longer legs compared to the general population (see supra at para 108). This shows that leg length plays at least some role in enhancing a runner’s performance.

113. Both Parties agree that the Taboga 2020 Paper is the only experimental study that collates data on the effect of RSP height on an athlete’s maximum speed. The Parties disagree on the cogency of the design of the study and the conclusions drawn by the authors. Profs Grabowski and Taboga, and Dr Beck relied on the Taboga 2020 Paper in support of the proposition that the height of an athlete’s RSPs does not determine their maximum speed in a 400m event. However, the Panel notes that the Taboga 2020 Paper involved changing not just the height of the RSPs but also the model and stiffness of the RSPs. As Prof Tweedy explained, “an experiment that gives high quality evidence is one where you have got […] experimental control”. To produce reliable results, the Taboga 2020 Study should have manipulated only one variable, i.e. height of the RSPs. The Panel accepts Prof Tweedy and Dr Connick’s evidence that, when only height is manipulated (i.e. when the subjects used their usual model of RSPs, at the usual stiffness), four out of five subjects attained their best maximum speed running at +2 cm and +4 cm. Therefore, the Panel accepts Prof Tweedy’s view, namely that the conclusion that leg length affects maximum velocity is “as strong as any conclusion” one can draw from the Taboga 2020 Paper.

114. The Panel also notes that there were other confounding factors in the Taboga 2020 Study. For instance, the subjects in study were not given time to familiarise themselves with running at taller heights. This appears to be a significant weakness. Runners using prosthetics need at least some time to get acclimatised to new RSP heights. Prof Weyand gave the example of Mr Alan Oliveira, a Paralympian athlete from Brazil, who stated that it took him at least two weeks to get used to his lengthened RSPs before the 2012 Paralympics, after which he was able to run much faster. The Panel also recalls that one of the main justifications offered by Mr Leeper for being unable to run at his MASH height was because he was not able to acclimatise himself (see supra at para 48). Similarly, he gave witness evidence that running on his 5 cm shorter RSPs still felt unnatural and uncomfortable, after months of using them (see supra at para 52). An adjustment/acclimatisation period was not accounted for in the Taboga 2020 Study.

115. The Panel is unable to draw any conclusions regarding the expert testimony presented by Mr Leeper on the effect the increased weight of longer RSPs may have on running speed. The Panel has not been offered any published, peer-reviewed scientific evidence specifically

showing that the incrementally greater weight materially affects a double-amputee runner’s maximum velocity. Similarly, the Panel is not able to assess the unpublished, non-peerreviewed study of Prof Grabowski’s team on the ability of non-amputee runners to increase their leg length through plantar flexion. That study was conducted with athletes with unilateral transtibial amputation, a specific class of athletes with different gait mechanics and performance abilities, rather than non-amputee athletes.

116. The Panel is also unable to accept Profs Weyand and Bundle’s comparative analysis of Mr Leeper’s performance in Boulder (on longer blades) and in Dallas (on shorter blades). Those tests were designed differently from each other and were conducted under different testing protocols and conditions, as both sides’ experts acknowledged. For example, Prof Weyand agreed that Mr Leeper’s performance in top-speed treadmill trials were recorded in Boulder and Dallas using different testing methods at each location. For the sake of clarity, the Panel does not intend to cast any doubt on the validity of the Boulder and Dallas tests taken individually. Scientists design their tests differently, based on their own training and experience. But, given the differences between the two tests, the Panel is unable to draw safe conclusions by comparing those two tests. To establish a true and reliable comparison, one would need to measure Mr Leeper’s top speed on both heights in tests designed and conducted using the same methodology.

117. For the above reasons, the Panel concludes that it is a legitimate purpose to employ Rule 6.3.4 to control height conferred through the use of RSPs.

C. Whether Mr Leeper’s proposed RSPs give him an overall competitive advantage

118. The next question is whether WA has established, on a balance of probabilities, that the specific RSPs chosen by Mr Leeper confer, through extra length, an overall competitive advantage that he would not have had if he had intact biological legs. As part of this inquiry, the Panel needs to determine two issues. First, whether Mr Leeper suffers any disadvantages by running on RSPs, and if so, whether such disadvantages outweigh any performance advantage he may have. Secondly, whether Mr Leeper is running above his natural height on his current RSPs.

a. The existence of any overall disadvantage of running on RSPs

119. Mr Leeper contests WA’s claim that he has an overall competitive advantage over able-bodied athletes in the 400m event. He contends that, far from enjoying any such advantage, he suffers a net disadvantage by running on RSPs, which disadvantage must be weighed against any alleged advantage he may have. Mr Leeper’s experts claimed the following:

- He has a disadvantage at the starting blocks: while able-bodied athletes can use muscular power in their legs and ankles to push out of the blocks, Mr Leeper’s RSPs prevent him from exerting maximal force at the start. Due to this biomechanical disadvantage, he begins every race at an immediate disadvantage compared to his able-bodied competitors. Prof Grabowski and her team tested Mr Leeper’s performance at the

starting blocks during the 2018 Boulder tests and quantified his acceleration disadvantage at 1.41 seconds.

- He has a disadvantage when running on the curves, as Mr Leeper’s RSPs are designed for straight-running movement and not curved movement. This places him at a disadvantage when running around the curves of a track, which make up about half of the 400m race. Prof Grabowski and her team quantified this disadvantage at 0.4 second.

120. WA’s experts agreed that Mr Leeper accelerates more slowly than able-bodied athletes at the starting blocks. However, they disagreed on the cause and magnitude of this disadvantage.

- Referring to video footage of Mr Leeper racing against his competitors, they claimed that Mr Leeper’s amputee and non-amputee competitors are able to lean forward and push more horizontally at the starting blocks. By contrast, Mr Leeper’s disproportionately long RSPs require him to assume a more upright position as the block-push commences, lest he lose balance. This vertical position reduces the amount of force he is able to exert on the starting blocks. Therefore, Mr Leeper’s disadvantage results from the disproportionate length of his RSPs, which impedes him from exerting immediate maximal force. It is not caused by an inherent limitation of RSPs as his experts claim.

- According to WA’s experts, the Dallas tests confirmed that Mr Leeper’s use of shorter RSPs reduced his disadvantage substantially.

- WA’s experts disagreed with Prof Grabowski’s team regarding the extent of the disadvantage. They quantified Mr Leeper’s disadvantage at 0.5 second, as opposed to 1.41 seconds predicted by Mr Leeper’s experts. WA’s experts also claimed that Mr Leeper benefits from a slower, lower exertion start. The gradual acceleration at the start leaves him less fatigued and enables him to run faster in the second part of the race.

121. WA’s experts claimed that Mr Leeper does not have any disadvantage on the curves. To prove their point, they first relied on actual race data from World Championships in Seville (1999), Berlin (2009) and London (2017) of 23 performances by some of the most accomplished 400m runners in the world. Although Prof Grabowski claims that Mr Leeper has a 0.4 second disadvantage on the curves, he is in reality the fifth-fastest person among these 23 performances to complete the second turn of the 400m race. Notably, the data show that the current 400m World Record holder and the 2004 400m Olympic gold medallist are both slower than Mr Leeper during this curved portion of the race. They also tested Mr Leeper’s curve-running in Dallas, which indicated that his running speeds around the curve (3.8% slower) than on straight segments are similar to the performances of non-amputees (~3.0% slower). This superiority of Mr Leeper’s curve-running ability is made possible by the effect his lengthened limbs have on his top running speed.

122. WA’s experts go on to say that even if one accepts that Mr Leeper has a cumulative disadvantage of 1.8 seconds in a 400m race, his overall advantage would still outweigh that disadvantage. The results of the Dallas tests showed that the 10 cm difference between Mr

Leeper’s MASH height and his current RSPs leads to a difference in top speed of 1.0 m/s, which equates to an advantage of 4.7 seconds over 400m. Therefore, even discounting the disadvantage that Mr Leeper’s experts claim he has, his overall advantage would still be 2.9 seconds over 400m.

123. WA’s experts also claimed that bilateral transtibial amputees running on RSPs on the whole have an advantage over non-amputees. In their Expert Report of 31 May 2020, WA’s experts, Profs Weyand and Bundle, referred to the all-time Paralympic 400m list for male transtibial amputees (both unilateral and bilateral) to show that the fastest seven athletes were all bilateral transtibial amputees. They claim that these performances provide compelling evidence of the overall or net competitive advantage that bilateral amputees have in sprint-running events. While their RSPs slow their sprint starts to an extent, the degree to which they dominate this list would be unattainable if their blades did not provide them with an advantage that more than offsets the modest disadvantage at the start. The same is true in Mr Leeper’s case. The variables on which Mr Leeper’s experts’ conclusion is based do not indicate a single source of advantage for Mr Leeper; yet, Mr Leeper’s personal record is faster than 99.88% of the elite non-amputee 400m runners. These data are at odds with Mr Leeper’s experts’ conclusion that bilateral transtibial amputees experience a disadvantage compared with non-amputee athletes.

124. According to the UQ Experts, evidence from a range of different sources provides reasonably strong support for the hypothesis that Mr Leeper’s RSPs give him a performance advantage over athletes not using such RSPs.

- RSPs are specifically designed for running fast, and have several design features to optimise running performance.

- In general, RSPs are much lighter than biological limbs. As a result, they “are easier to swing through the air due to their decreased inertia compared to biological legs”. The lightweight design of RSPs has previously been highlighted as a possible factor in Mr Pistorius’s ability to reposition his limbs during the swing phase of sprinting to achieve exceptionally high step-frequencies. In addition, it takes less force and less work to accelerate a small mass, and it is possible that lighter limbs may enhance running economy, which is the amount of energy required to maintain a given speed. Mr Leeper’s RSPs weigh 2.5 kg in total, which is a lower mass and weight advantage of 57% compared to an intact limb.

- RSPs are made of carbon fibre and do not fatigue as biological legs do during a 400m race. It is plausible that this feature of RSPs may enhance performance.

- An important feature of all RSPs is that, compared to biological legs, athletes have the option of improving running performance by customising the features of their RSPs (design, weight, materials, and length). Non-amputee athletes do not have this option. And with the exception of length, RSP customisation is largely unregulated by sporting authorities.

125. The Panel notes that it is common ground that RSPs may slow down a runner at the starting blocks (although the experts disagree as to the cause and extent of that disadvantage). It is disputed whether they may slow down a runner on the curves, compared to a runner on biological legs. In support of their respective positions, both sides’ experts have relied on experimental evidence from both the Boulder and the Dallas tests. As noted at paragraph 116 above, the Panel declines to draw any conclusions by directly comparing the results from the Boulder and Dallas tests. Those tests were conducted using different testing protocols and under different testing conditions.

126. There is, however, observational evidence in the record that is relevant to the issue. In their Joint Supplementary Expert Report of 10 July 2020, WA’s experts presented data on the split times for the second curve (between 200m and 300m) for Mr Leeper and 400m World Championship finalists from 1999, 2009, and 2017. These data show that Mr Leeper is the fifth-fastest runner to complete the second curve, ahead of non-amputee runners such as Mr Wayde van Niekerk (the current world and 2016 Olympic record holder in the 400m event), Mr Jeremy Wariner (former world champion and Olympic gold medallist in the 400m event), and Mr Steven Gardiner (2019 world champion in the 400m event). In the Panel’s view, this evidence supports the assertion that Mr Leeper does not suffer a disadvantage on the curves compared to able-bodied runners.

127. There is also evidence in the record that Mr Leeper does not suffer an overall disadvantage compared to unilateral amputee runners, who are limited by their single biological legs. There is no dispute that double-amputee runners can readily adjust the individual features of their RSPs such as design, material, weight, and length. It is also undisputed that RSPs are much lighter than biological limbs and that they do not fatigue an athlete in the same way as biological legs do. Therefore, it appears that these features of RSPs, set at the discretion of a double-amputee runner, may enable him or her to offset any disadvantage he or she may suffer in a 400m race. In this regard, the Panel acknowledges the evidence submitted by WA’s experts on the performance gap between bilateral transtibial amputees (who can adjust the individual features of both of their RSPs) and unilateral transtibial amputees (who can only do so by reference to their biological leg). The all-time Paralympic 400m list for male transtibial amputees shows that the fastest seven athletes are all bilateral transtibial amputees. The difference between the fastest time for a double amputee (Mr Leeper) and the fastest time for a unilateral amputee (Mr Michail Seitis) is 5.24 seconds – a considerable difference, even larger than the performance difference between male and female non-amputee 400m world-record holders. The Panel considers the extent of this performance difference as evidence that RSPs do provide an overall advantage to double-amputee runners, which more than offsets any disadvantages they may have.

128. According to Prof Grabowski, Prof Taboga, and Dr Beck, Mr Leeper obtains no advantage on his RSPs; he only suffers disadvantages. The Panel finds it difficult to reconcile this view with Mr Leeper’s overall performance. The evidence before the Panel shows that Mr Leeper is able to achieve times quicker than 99.88% of the elite non-amputee 400m runners. The Panel agrees with the Leeper I panel (Leeper I award, para 374) that a range of factors may explain how Mr Leeper was able to achieve such times. It is possible that they are the result of natural talent, training, and sheer determination, meaning that Mr Leeper would be the

fastest 400m sprinter if he had biologically intact legs. However, it is more likely that Mr Leeper was able to achieve these times because his RSPs confer on him a net performance advantage, which more than counterbalances any disadvantage in respect of initial acceleration and curve-running.

129. On the totality of evidence before it, the Panel concludes that, overall, bilateral transtibial amputees like Mr Leeper do not suffer a performance disadvantage compared to non-amputee runners.

130. Given the Panel’s finding that the height of RSPs is a factor that affects a runner’s maximum speed (see supra at paras 112 and 117), the next question before the Panel is whether Mr Leeper is running above his natural height on his current RSPs.

D. Whether Mr Leeper is running above his natural height on his proposed RSPs

131. The new MASH rule, which was adopted in 2018, serves to calculate the maximum permitted length of RSPs used by double-amputee athletes. Beyond that length, a double-amputee athlete using RSPs is considered to be running unnaturally tall and, therefore, to have a disallowable advantage. Mr Leeper started competing against able-bodied athletes from June 2017 onwards. Accordingly, WA applied the MASH rule to Mr Leeper to determine his natural height. The USATF, effectively Mr Leeper’s federation, also employs the same methodology, notably for Olympics-qualification races. It officially informed Mr Leeper in June 2018 that the IPC had issued a new MASH rule in January 2018 and that he would need to submit his revised MASH height certification in order to be eligible to compete in the 2018 USATF Track and Field championships. Thus, Mr Leeper had notice of the application of the MASH methodology to him since mid-2018 at the latest.

132. Using the MASH rule adopted by the IPC/WPA, WA calculated Mr Leeper’s MASH at 174.44 cm. Accordingly, it claimed that Mr Leeper was running significantly taller on his new RSPs, which give him a height of 185.42 cm. Mr Leeper does not contest the arithmetic calculations, but he does contest the use of the MASH rule in the first place, on various grounds.

133. The Panel has carefully reviewed both Parties’ evidence on the application of the new MASH rule, which it summarises below.

134. According to WA, the 2018 MASH rule is directly relevant to the question of whether Mr Leeper has an overall competitive advantage over non-amputee athletes, seeking as it does to ensure that Mr Leeper’s RSPs do not enable him to run taller than would have been the case if his legs were intact.

135. Prof Tweedy and Dr Connick provided a detailed history of the MASH rule and its evolution in their expert reports and hearing testimony. The MASH rule has been used by the IPC/WPA, the global governing body for Paralympic athletics, since 1993. This rule regulates the height of RSPs used by bilateral lower-limb amputees so that they are proportionate in length to the rest of their body. It does so by predicting their natural height, i.e. the height bilateral-amputee athletes would have if their legs were intact, and by disallowing RSPs that

make them stand taller than their natural height. In this way, it safeguards competition integrity and ensures that double-amputee athletes (who can self-select the height at which to run) do not have an unfair advantage over non-amputee athletes (who are limited by the height of their biological legs), unilateral amputee athletes (who are limited by the height of their single biological leg), or other bilateral-amputee athletes.

136. The methodology of predicting height under the MASH rule has evolved over the years. Prior to 2017, the IPC/WPA MASH rule predicted natural height based on two upper body measurements: the demispan (the distance from the middle of the sternal notch to the tip of the middle finger) and forearm length. This methodology came under scrutiny at the 2012 Paralympic Games, when the bilateral-amputee athlete Mr Oliveira defeated Paralympic champion Mr Pistorius. Following the race, Mr Pistorius complained that “[t]he regulations allow athletes [to] make themselves unbelievably high” and that “[w]e aren’t racing a fair race”.

137. In 2013 and 2014, Prof Tweedy and his team became aware of a study by Dr Alicia Canda that proposed a more reliable method for estimating height. Dr Canda measured the body segments of 545 anatomically intact Spanish Caucasian males and 502 anatomically intact Spanish Caucasian females using internationally accepted standards. From these measurements, she derived a series of 51 equations (27 for males and 24 for females), each of which used different combinations of body segments to estimate standing height. One of these 51 regression equations did not require the length of the lower leg to be known. According to the UQ Experts, this particular equation, called the “Canda-1” equation, can predict the height of double-amputee males with great accuracy. The “Canda-1” equation required summing four anatomical measurements (thigh length, upper arm length, forearm, and sitting height) after each of them was multiplied by an empirically determined coefficient8. The Canda-1 equation is based on the well-established biological principle that, while there is considerable variability in the standing height of human beings, their height remains in proportion to the rest of their body. It therefore incorporates measures of both the torso (where, on average, a White subject will have a longer measurement) and limb segments (where on average a Black subject will have longer measurements) to obtain predictive accuracy. So, if a male has a shorter trunk and longer limbs (as do Black athletes), any disadvantage resulting from a shorter trunk would be offset by relatively longer measurements of their upper arm, forearm and thigh.

138. As the Canda-1 equation was derived from a Spanish Caucasian population, Prof Tweedy and Dr Connick conducted a study to determine whether “it can be validly applied to predict standing height in geographically distant, genetically diverse populations”. As noted in their Expert Report of 16 March 2021, they based their study on the scientific premise of the American Association for Physical Anthropology (“AAPA”) that natural biological variations between individual human beings are caused by variation in genetic make-up. Human genetic variation is driven by geographic distance and not race. On this premise, the UQ Experts validated the Canda-1 equation in population samples from two locations that were geographically distant from Spain – Japan and Australia. For comparison, they also tested the upper-body method in these populations. According to the UQ Experts, the Canda-1 equation provided the most valid

8 Canda-1 equation = -5.272 + (0.998 × sitting height) + (0.855 × thigh) + (0.882 × upper arm) + (0.820 × forearm).

estimations of standing height across Japanese and Australian samples. Their published study showed the following:

- The error of the estimated stature in the Australian population sample was not significantly different from the error of the estimated stature in the Japanese sample (probability value greater than 0.05). This was even though the Japanese males on average had a significantly shorter thigh length (41.5 cm) compared to the Australian males (45.5 cm), and were significantly shorter in standing height (168.1 cm) than their Australian counterparts (179.5 cm). Based on these data, the authors concluded that the Canda-1 equation could be used to predict height across different populations, including people with a different range of body proportions, such as people of African descent.

- The upper-body method under which double-amputee athletes were assessed prior to 2019 systematically over-estimated standing height by 7 cm. By contrast, the Canda-1 equation had a systematic bias of only 0.2 cm.

139. Based on Prof Tweedy and Dr Connick’s scientific evidence, the IPC/WPA changed the MASH methodology from the upper-body method to the Canda-1 equation, with effect from 1 January 2018. The MASH rule adopted by the IPC/WPA adds a pure error factor of 1.91 cm to the height predicted by the Canda-1 equation to account for any natural biological variation9. According to the new MASH rule, Mr Leeper’s natural height is 174.44 cm (172.53 cm before the pure error factor of 1.91 cm is applied).

140. To prove that the Canda-1 equation measures the standing height of Black athletes accurately, Profs Weyand and Bundle tested it against the data collated by Dr Tanner in his 1964 study titled “The Physique of the Olympic Athlete”. In that study, Dr Tanner collected two sets of data. First, in the Appendix of his study, he recorded the race and sitting height for each athlete from a total of 200 male athletes who competed either at the 1958 Commonwealth Games in Cardiff or the 1960 Olympic Games in Rome. Second, he included a sequence of three photographic plates for each one of a subset of 104 of the athletes. Profs Weyand and Bundle extracted the measurements of thigh, upper arm, and forearm lengths from the photographs of 100 out of 104 athletes, using photographic analysis. For the fourth measurement required for the Canda-1 equation, the sitting height, they relied on the measurements that Dr Tanner recorded in his Appendix for those 100 athletes. These included 76 White athletes, twelve Black athletes, and nine Asian athletes. The experts then input the required four measurements for these 100 athletes in the Canda-1 equation. They found that the Canda-1 equation predicted the height of all 100 athletes accurately, within an average of 2.8 cm. Specifically, the Canda-1 equation overestimated the height of ten out of twelve Black athletes. It underestimated the height of two Black athletes, by 0.2 cm and 1 cm respectively (once the pure error factor of 1.91 cm was applied).

141. WA’s experts agree with Mr Leeper that Black athletes have different body proportions from White athletes. As noted above, Black athletes have longer limbs while White athletes have

9 New MASH rule = -5.272 + (0.998 × sitting height) + (0.855 × thigh) + (0.882 × upper arm) + (0.820 × forearm) +

1.91 cm of pure error factor.

longer sitting heights. However, the magnitude of such anthropometric differences is very small. Dr Tanner’s data underscore this fact. Dr Tanner quantified the differences between White and Black athletes in relation to their sitting height and leg length. He recorded an average difference of 1 cm in sitting height and an average difference of 3.2 cm in leg length (with a leg length to sitting height ratio of 0.90 in White athletes and 0.94 in Black athletes). By contrast, the leg length to sitting height ratio for Asian subjects was much smaller than was the case for White and Black athletes. Yet, as Prof Tweedy and Dr Connick showed, the new MASH rule predicted the stature of Asian subjects with accuracy. The new MASH rule mitigates any marginal anthropometric differences through the 1.91 cm pure error factor.

142. Finally, WA’s experts argued that in practice the new MASH rule predicts accurately the standing height of Black athletes. Prof Tweedy’s team reported that the MASH rule accurately predicted the height of Mr Marko Cheseto, a Kenya-born bilateral transtibial amputee athlete. Mr Cheseto’s known biological height was 6’0” (182.88 cm). After both his legs were amputated following a frostbite incident, the new MASH rule predicted his natural height at 5’11.75” (182.14 cm). The rule, therefore, underpredicted Mr Cheseto’s height by only 0.74 cm.

143. For his part, Mr Leeper contends that the IPC’s/WPA’s MASH rule is based on scientific studies that did not include as subjects any Black persons of African descent. Citing various studies, including the 1964 study by Dr Tanner, he argues that it is well-established in the scientific community that Black persons have different limb proportions than persons of other genealogical backgrounds, which differences affect the performance of these athletes in sprinting competitions. Therefore, it is “scientifically unsound” to state that Mr Leeper runs at an unnaturally tall height based on the MASH rule. In particular, Mr Leeper argues as follows:

- The MASH rule is based entirely on a study conducted with Spanish Caucasian subjects, in which not a single Black person was involved. The empirically defined co-efficients in the MASH rule were selected without using any morphological data from Black persons of African heritage. These percentages cannot mathematically account for the anthropometric differences that exist between Black persons and persons of other genealogical backgrounds.

- The rule was later validated in Japanese and Australian populations but not in a Black population. To apply the MASH rule to Black athletes, WA would have to conduct a study with Black subjects. Indeed, WA’s experts agreed that it would be better to conduct such a population study. However, they instead rely on the photographs published in Dr Tanner’s study to argue that the MASH rule reliably predicts the stature of Black athletes. Not only are these photographs 60 years old, they concern a small sample size of twelve Black athletes. In addition, WA’s experts’ reconstruction of measurements from these two-dimensional photographs is inaccurate and unreliable.

- Mr Leeper disagrees with the measurements used by WA’s experts’ in Mr Cheseto’s example. He claims that Mr Cheseto stated in a radio interview that his pre-amputation height was 6’1” (185.42 cm), while his MASH-calculated height is 5’9” (175.26 cm). The

MASH rule, therefore, underpredicted Mr Cheseto’s height by a full four inches (10.16 cm).

144. Mr Leeper’s main complaint is that the new MASH rule was neither based on nor validated by reference to a population study involving Black athletes. The Panel acknowledges that the new MASH rule was derived from a dataset underlying an anatomical study of Caucasian male and female athletes published in 200910. However, before being adopted by the IPC and the WPA, the Canda-1 equation underlying the MASH rule was validated in small groups of populations in Japan and Australia. The UQ Experts conducted this validation exercise on the scientific premise that geographic distance, not race, is the main driver of genetic variability in the bodily proportions of different populations. This premise, which is not contested by Mr Leeper, was established by the AAPA, the world’s leading professional organisation for physical anthropologists. Based on this premise, the UQ Experts concluded that, as the MASH rule successfully predicted height in Japanese and Australian populations, two populations farthest from Spain, it would also successfully predict height in other geographically distant and genetically diverse populations, including people of African descent. In the absence of contrary evidence, the Panel accepts the scientific premise before it, along with the resulting conclusions advanced by the UQ Experts.

145. Mr Leeper claims that the MASH rule could only be considered sound if it had been validated specifically with reference to athletes of African descent. However, as noted in the preceding paragraph, the only scientific evidence before the Panel is that human genetic variation (which causes anthropometric differences) is driven by geographic distance, not race. Mr Leeper has not challenged the scientific validity of this evidence through affirmative evidence in these proceedings. Indeed, certain studies cited by Mr Leeper’s experts have at least partially confirmed this premise. For instance, the study by Dr Cherice Hill (Leeper Exhibit 31) states that “[r]ace is a sociocultural categorization not based on any single characteristic, trait, or gene. Thus, the human species cannot be strictly divided into biologic or genetic races”. Therefore, according to the evidence before the Panel, the validation of the MASH rule on the basis of geographic distance is more methodologically apt than race-oriented validation.

146. Furthermore, the Panel notes that there is some evidence in the record which lends a degree of support to the proposition that the MASH rule predicts the lower-leg length of Black athletes reasonably accurately.

147. For instance, WA’s experts provided anecdotal evidence that the MASH rule accurately predicted the natural height of Mr Cheseto, whose pre-amputation height was known. The Panel recognises that there is some disagreement between the Parties’ experts on the measurement of Mr Cheseto’s MASH and pre-amputation heights. While WA’s experts state that Mr Cheseto’s MASH height, as published in the WPA Classification Master List, is 5’11.75” (182.14 cm), Mr Leeper’s experts claim that his MASH height is 5’9” (175.26 cm). The Panel is bound to accept Mr Cheseto’s MASH height as published in the official records, which is 5’11.75” (182.14 cm). The experts also disagree on Mr Cheseto’s pre-amputation

10 Its validity for that segment of the world population is not being contested in these proceedings (or elsewhere, so far as

the Panel is aware).

height. WA’s experts state that Mr Cheseto’s actual height was 6’00” (182.88 cm) whereas Mr Leeper’s experts claim that it was 6’1” (185.42 cm). The Panel is unable to form a view on the actual, pre-amputation height of Mr Cheseto. However, even if the Panel were to accept Mr Cheseto’s actual height as reported by Mr Leeper’s experts, then the MASH rule underpredicts Mr Cheseto’s height by 3.28 cm. This is in stark contrast to Mr Leeper, who seeks to use RSPs that give him an additional leg length of 10 cm compared to his MASH height.

148. The Panel also notes that there is partial evidence in the record, derived from Dr Tanner’s data (both raw measurements and photographic analysis), that the MASH rule accurately predicts the height of Black elite athletes. The Tanner data showed that the Canda-1 equation overpredicted the height of ten of the twelve Black athletes and underpredicted the height of two Black athletes by a slight margin. Any disadvantage in the latter hypothesis would be offset by adding the 1.91 cm pure error factor in the MASH rule.

149. In the Panel’s view, the key point about this body of evidence, limited as it is, is that it does not cast doubt on the MASH methodology, especially given the pure error factor of 1.91 cm. In Mr Leeper’s case, his MASH height is 174.44 cm (5’9”), some 10 cm (2”) lower than his height on his proposed RSPs. This is a considerable margin, more than five times the 1.91 cm error factor.

150. The Panel is accordingly satisfied, on the basis of the scientific evidence before it, that the MASH rule provides a reliable method of estimating standing height in geographically distant populations with different anthropometric proportions, including people of African descent. The Panel is also satisfied that Mr Leeper’s proposed RSPs confer upon him a standing height comfortably beyond any margin of error that one might allow for on the evidence before the Panel.

151. Thus, on the evidence before it, the Panel concludes that WA has established, on a balance of probabilities, that, on his chosen RSPs, Mr Leeper is running approximately 10 cm taller than his natural height. Through this extra length, Mr Leeper’s RSPs confer upon him an overall running-speed advantage that Mr Leeper would not have if he had biologically intact lower limbs.

152. The Panel notes, however, the consensus among the experts that it would be sound and desirable for the MASH rule to be specifically validated by reference to African athletes or athletes of African descent. The Panel strongly encourages WA to conduct such a population study as soon as restrictions related to the COVID-19 pandemic allow. The Panel also strongly recommends that, in conducting the study, WA considers how best to organise the work to avoid any suggestion of confirmation bias.

E. Proportionality considerations

153. The Panel has determined that Mr Leeper’s chosen RSPs, through an extra length of 10 cm, give him an overall competitive advantage within the meaning of Rule 6.3.4 (see supra at paras 149–151). The Panel turns now to consider whether there is a less-intrusive alternative to disallowing Mr Leeper’s chosen RSPs in WA-sanctioned events.

154. The panel in Leeper I determined the onus of proof as to the lawfulness or unlawfulness of the Rule. In that context, the panel considered the approach that the CAS panels had taken in Chand and Semenya, where it had been determined that, for a rule that discriminates on the basis of a protected ground, the burden of proof shifted from the athlete to the IAAF (now known as WA) to establish that the relevant regulation was necessary, reasonable and proportionate. The panel in Leeper I concluded that the same approach should logically apply in a case where a regulation promulgated by the IAAF discriminates unlawfully against athletes with disabilities. The Leeper I panel concluded at paragraph 317 of its award that, although the provisions of the Rule in issue, both in that case and here, “apply to all athletes who wish to use mechanical aids during IAAF-sanctioned competitions, the Rule was enacted with the specific objective of regulating the ability of athletes with disabilities to participate in IAAF-sanctioned competitions while using mechanical aids to overcome the limitations that arise as a result of their disabilities”. The panel continued, at paragraphs 318 and 319, to observe that while Rule 6.3.4 is not directly discriminatory on its face, it is indirectly discriminatory in its practical application, in that “it is likely to affect a larger number and a much greater proportion of disabled athletes than able-bodied athletes”.

155. That reasoning was not challenged before this Panel.

156. The panel in Leeper I then turned to the question of whether the Rule pursued a legitimate objective and concluded that it did, this being “to ensure the fairness and integrity of competitive athletics by ensuring that the outcome is determined by competitors’ natural talent, training and effort, and not by the use of mechanical aids which confer an artificial competitive advantage over athletes who are not using such aids” (at para 332).

157. In that analysis, however, the panel observed that, in formulating the Rule, there was a “lack of attention and concern regarding the impact of the Rule on the rights and interests of disabled athletes” and, specifically, the IAAF did not enact a bespoke rule to deal specifically and directly with the position of disabled athletes who need to use prosthetic aids as a substitute for missing biological limbs (at para 331). The panel later noted (at para 349) as part of its detailed analysis that “the significant risk of disabled athletes being unfairly prevented from competing, and the risk that disabled athletes who derive no competitive advantage from their prosthetic aids will be forced to incur irrecoverable expenses in proving [that their mechanical aids do not confer any overall athletic advantage] both weigh heavily in any assessment of the overall necessity, reasonableness and proportionality of the Rule” .

158. The above observations, which were not the subject of adverse submission or comment before this Panel, are relevant to the questions of necessity, reasonableness and proportionality.

159. The panel in Leeper I concluded that it was not necessary, reasonable or proportionate to impose the burden of proof on the disabled athlete to establish the effect of a mechanical aid on their athletic performance and competitiveness. As noted, WA now accepts that, according to Rule 6.3.4 as it now stands, it bears the burden of proof of establishing that Mr Leeper derives an overall competitive advantage from the use of the particular prosthetic aid.

160. During closing submissions, the Panel raised the question of proportionality with the Parties, specifically the proportionality in its application to Mr Leeper in light of his evidence that he

could not adjust to new prostheses that conformed to his MASH height in time for selection for, or competing in, the Tokyo Olympic Games. The Panel asked whether there was any other way in which Mr Leeper could compete. The Panel requested the Parties to consider this question, adjourned the hearing and reconvened to hear further submissions.

161. The Parties’ submissions are summarised as follows.

162. WA submitted that:

- The Leeper I panel found that the purpose of Rule 6.3.4 is to pursue a legitimate objective (at para 332). This Panel should proceed on the same basis. It is for Mr Leeper to show that there is a less-restrictive means of achieving that legitimate objective. He has not even attempted to identify a less-restrictive means, let alone substantiate it by competent evidence.

- Mr Leeper has never argued that Rule 6.3.4 operates disproportionately in this case. As a result, this was not a case that WA was ever required to meet. In any event, Rule 6.3.4 does not operate disproportionately in this case:

o Mr Leeper argues that he has been running on MASH non-compliant blades for many years and cannot now reduce them to his MASH height. However, Mr Leeper’s testimony at the hearing that he tried running on MASH-compliant blades for seven to ten days but that he “threw [his] back out” was never mentioned before. There is absolutely no competent medical evidence in the record to that effect. Mr Leeper has explained that it has taken him several months to acclimatise to blades 5 cm shorter than his previous blades (at issue in Leeper I). One now knows that he can run safely on those blades. Therefore, his attempt to run on MASH-compliant blades for only seven to ten days cannot demonstrate that he is not capable of familiarising himself with such blades if given sufficient time.

o If it is found that WA has met its burden under Rule 6.3.4, the result would be that Mr Leeper is unable to compete in four events (the quadrennial Olympic Games and the biennial World Outdoor Championships, World Indoor Championships, and World Relay Championships). However, Mr Leeper may still compete in all other WA-sanctioned international-level events, including Diamond League events, Continental Tour events, and Indoor Tour events, with his results listed separately. Mr Leeper may still compete in national-level events, without restriction from WA, subject only to any eligibility conditions imposed by the national association. This limited impact confirms the proportionality of the rule, i.e. that it does no further than what is necessary to achieve the legitimate objective of Rule 6.3.4.

- Mr Leeper never had any legitimate expectation that he would be permitted to run in WA or USATF events on his non-MASH compliant blades. He knew that the MASH rule was implemented to prevent double amputees from gaining an advantage by “running tall”. He knew that his national federation, USATF, had adopted the new

MASH rule for its competitions. Mr Leeper was not allowed to run in the 2018 USATF National Championships at all, because he was not MASH-compliant, and he was allowed to run in the 2019 USATF National Championships only under protest. He also knew that WA had a rule that prohibited the use of blades that gave the user an overall competitive advantage.

- WA cannot conceive of any mechanism that would be appropriate to enable Mr Leeper to compete in the Olympic trials at MASH plus 10 cm. The entire point of Rule 6.3.4 is to ensure fairness by ensuring that everyone is running on a level playing field. Any individual exception would swallow the rule: there will not be a level playing field if Mr Leeper runs above his MASH height.

163. Mr Leeper submitted that:

- If the Panel concludes that WA has met its burden of proving that Mr Leeper’s proposed RSPs provide him with an overall competitive advantage, then it should determine whether the sanction of prohibiting Mr Leeper from competing this year on his current RSPs would be an excessive and disproportionate sanction in this particular case. The burden of proof on proving proportionality rests on WA, which has not met it.

- Rule 6.3.4, as interpreted by WA to bar Mr Leeper from competition on his RSPs with able-bodied athletes, is grossly disproportionate to any legitimate objective. He is currently not able to compete at a lower height without risking injury.

- He relied for many years on the fact that, prior to 2018, the MASH rule permitted him to compete at a greater height. WA never adopted the MASH rule, and gave no warning to Mr Leeper or any amputee athletes that the MASH rule would even apply to WA’s events until the Leeper I proceedings began in 2020. The only procedures that WA adopted regarding Mr Leeper’s RSPs never mentioned the MASH rule. Mr Leeper’s only fair inference from this obvious omission was that the MASH rule did not apply and that the question of “overall advantage” would be assessed in the round, based on scientific issues.

- WA also wrongly contends that the rule has only a “limited impact” because Mr Leeper is unable to compete in only four events. But denying Mr Leeper his greatest ambition to compete at the highest level on the world stage is the “limited impact”.

- The appropriate, proportional relief would be to permit him to compete on his current RSPs in all WA events, including the Olympics if he qualifies, with the option of a possible reconsideration of that determination in future years if WA takes appropriate steps to set proper, non-discriminatory limits on the length of RSPs with which Mr Leeper and other double-amputee athletes would have to comply. That proportionate relief would permit WA potentially to apply the MASH rule to Mr Leeper but only if it has meanwhile been validated for athletes of African heritage through a new population

study conducted by scientists having no connection with WA or any other interest that bring into question their neutrality.

164. The Panel notes as a preliminary matter that Mr Leeper did not explain to the Panel why, as he contended, Rule 6.3.4 was not a necessary, reasonable or proportionate way of achieving WA’s legitimate objective of ensuring the fairness and integrity of competition. The Panel, given its findings in Sections X.B–C, is unable to accept Mr Leeper’s contention.

165. The Panel’s proportionality analysis, then, hinges on one important question: whether there exist any practical alternatives to disallowing Mr Leeper’s current RSPs. Both Parties have (at the Panel’s urging) considered alternatives to total disallowance. However, neither Party has identified a workable alternative. Mr Leeper submitted that he should be permitted to compete on his current RSPs, which decision could be reconsidered in the future. WA raised other possible alternatives, as did the Panel, including a time handicap, or starting further back at a distance calculated to equate to the advantage. For each suggestion, WA explained why it would be unworkable and unfair to the other athletes in a 400m race. Mr Leeper did not adopt any of WA’s suggestions.

166. In the absence of workable alternatives, the Panel declines to adopt a novel alternative to the requirement that, in order to compete in WA-sanctioned events, Mr Leeper must run at his MASH height. Given this finding, the Panel need not assess the Parties’ other arguments made in relation to proportionality.

167. The Panel adds that it has not received medical evidence that Mr Leeper risks injury by running on his MASH height. The Panel acknowledges that Mr Leeper will not have time to acclimatise himself to MASH-compliant RSPs for the US trials for the Tokyo Olympics; but he will have time to acclimatise himself for the 2022 Indoor and Outdoor World Championships.