The court of arbitration for sport jurisprudence on match-fixing: a legal update

The Court of Arbitration for Sport (CAS) jurisprudence on manipulation of sports competitions has vastly evolved from its initial award in RSC Anderlecht in 1998, to now Labuts in August 2020. Alongside, international and national regulations, as well as sporting regulations, including, most recently, the Council of Europe’s Macolin Convention on the Manipulation of Sports Competitions, have sought to effectively tackle the omnipresent, ever-growing phenomenon of competition manipulation. Against this backdrop, this article briefly outlines the existing legal landscape on manipulation, followed by a chronological detailing of each CAS issued award. The key aspects of defining such sanctionable behaviour, select issues of standard of proof and types of evidence which are admissible and relied on, as well as the manner and quantum of sanction are then analysed. Ultimately, noting empirical trends across these awards, questions on ne bis in idem, proportionality of sanctions and legal certainty across CAS jurisprudence are raised.


Introduction
The manipulation of sports competitions 1 is certainly not a new phenomenon; on the contrary, it has been reported for centuries. For example, in August 1774, a London newspaper, The Morning Chronicle, described the alleged fixing of several cricket matches on the Artillery Ground in northern London and deplored the fact that the "game of cricket has too long been perverted […] to excessive gaming and public dissipation". 2 Today, in part due to globalization, match-fixing has become a huge cross-border phenomenon. It is currently estimated that billions of dollars are circulated every year through crime syndicates, with more than 80 countries reporting match-fixing offences to the INTERPOL during the last few years. In a recent operation called "Veto" conducted by Europol, 3 a total of 425 officials, players and criminals, from more than 15 countries, were suspected of being involved in the manipulation of more than 380 professional football matches in Europe, Africa, Asia, and South and Central America. 4 More importantly, these activities were part of a sophisticated organized crime network, which generated over €8 million in betting profits and involved over €2 million in corrupt payments to those involved in the matches. 5 The INTERPOL also reported that its operation, SOGA ("SOccer-GAmbling"), against illegal betting carried out between 2007 and 2014 resulted in more than 8400 arrests, the seizure of almost USD 40 million in cash and the closure of around 3400 illegal gambling dens which handled bets worth almost USD 5.7 billion. The operations have successfully removed a major source of proceeds for organized crime syndicates. 6 Match-fixing occurs in many sports, with regular recurrence reported in sports such as football, cricket, tennis, boxing, basketball, badminton, ice hockey, handball, volleyball but also in bridge, darts or eSports. 7 This global nature and tremendous growth of matchfixing recently gave birth to a specific international convention-with a worldwide reach-adopted by the Council of Europe in 2014. Known as the Convention on the Manipulation of Sports Competitions (CETS No. 215, hereafter, the "Macolin Convention"), it entered into force on September 1st, 2019. In parallel to this international effort, the Court of Arbitration for Sport ("CAS") cases dealing with manipulation of competitions have become increasingly abundant and have set important milestones in the fight against this phenomenon.
In this complex context, this article strives to capture, describe and analyse the relatively rich CAS jurisprudence related to match-fixing, while also focusing on several key issues such as the definition of competition manipulation, selected evidentiary issues and sanctions.

Terminology
Several terms, such as "match-fixing", "competition fixing", "manipulation of competitions" and so forth are used at academic 8 and legislative 9 levels to describe the fraudulent alteration of a sports competition or event. Terminology used in national regulations is more diverse and includes concepts such as "sporting fraud" (e.g. India), "corruption of players" (e.g. Malta), "bribery in sporting contests" (e.g. USA) or "corrupt activities relating to sporting events" (e.g. South Africa). 10 The Macolin Convention-which is the most prominent international legal instrument combating this phenomenon-uses the terms "manipulation of sports competitions".
In this paper, the terms "manipulation of sports competitions" and "match-fixing" are used interchangeably, except when otherwise indicated.

Duality of sanctions for match-fixing: a brief reminder
The manipulation of sports competitions is sanctionable at different levels.
The first level consists of disciplinary (sporting) sanctions, applied by the relevant sports bodies according to their internal punitive system (known as "sport justice").
At the second level are state sanctions, applied by public authorities ("state justice"). Depending on the applicable national law, the latter may be of a civil, administrative, disciplinary or criminal nature. For instance, the fixing of a football match within the Swiss championship may be sanctioned both by the Swiss national football federation 11 and by civil 12 and criminal law. 13 The CAS has made it clear that "disciplinary sanctions imposed by associations are subject to civil law and must clearly be distinguished from criminal penalties". 14 Obviously, in the CAS's view, sporting and criminal sanctions may be complementary. 15 The legal nature of "sport sanctions"-which may include, inter alia, warnings, bans, relegations, fines and other penalties, 16 has been clarified by the Swiss Federal Tribunal ("SFT") in the Gundel case, 17 as statutory, which is a form of contractual sanction. 18 This article focuses on the procedure and sanctions for tackling match-fixing cases, initiated and applied by sport bodies and reviewed by the CAS on appeal.

Legal basis for the fight against match-fixing
The manipulation of sports competitions being a global phenomenon, the fight against it is regulated at different levels.

International level
At an international level, three international conventions tackle match-fixing, either through general instruments aimed at fighting against corruption and transnational organized crime [United Nations Convention Against Corruption 19 ("UNCAC") and United Nations Convention Against Transnational Organized Crime 20 ("UNTOC")], or through a specific instrument dedicated to the manipulation of sports competitions (the Macolin Convention). These international instruments are needed notably in order to enhance cooperation between different countries, promote good governance and set international standards. In the UNCAC framework, "match-fixing" can appear in a form of six corruption criminal offences: active and passive bribery in the public sector (Art. 15-16 of the UNCAC); active and passive trading in influence (Art. 18 of the UNCAC); and active and passive bribery in the private sector (Art. 21 of the UNCAC).
The UNTOC is the main international instrument in the fight against transnational organized crime, which has, as its purpose, to "promote cooperation to prevent and combat transnational organized crime more effectively". Based on the idea that if crime can cross borders, so too must law enforcement, it can be applied to match-fixing if transnational elements and organized criminal groups are involved.
However, "countries seem [...] not to be able to fully utilize the potential and the added value of the UNTOC and the UNCAC to effectively combat match-fixing. The absence of several important elements of match-fixing from the scope of application of these Conventions might leave different offences unaddressed". 21 The most promising international instruments thus seem to be the Macolin Convention, which came into force on 1 September 2019. To date (October 2020), the Macolin Convention has been signed by 38 States (including by one non-European State, i.e. Australia) and ratified by seven of them. 22 As its Explanatory Report clearly mentions, this convention completes the framework of the UNCAC and the UNTOC, as it specifically deals with cases involving the manipulation of sports competitions which may occur outside any transnational crime network and without any acts falling within the definition of corruption having been committed. 23 It is important to mention that the Macolin Convention involves all relevant stakeholders, namely public authorities, sports organizations and sports betting operators. It establishes important mechanisms of information exchange through national platforms (Art. 13 of the Macolin Convention) and cooperation regarding law enforcement.

National level
Second, at a national level, around thirty countries have adopted or are currently in the process of enacting specific legislation criminalizing the manipulation of sport events, 24 many of which are inspired by the Macolin Convention.
While these national regulations differ from each other in terms of definition and scope of the match-fixing offence, as well as in terms of applicable sanctions, they all share a common perspective on the seriousness of this offence and on the need to effectively tackle it in practice. 25

Sport organizations
Thirdly and importantly, the fight against match-fixing is also regulated at the level of (international) sport organizations.
According to our ongoing research on this topic, 34 of the 35 International Federations (IFs, or bodies governing sporting disciplines included in the summer and winter Olympics) have adopted specific provisions or the International Olympic Committee's ("IOC") Olympic Movement on the Prevention of the Manipulation of Competitions of 2016, addressing match-fixing, notably under the angle of prevention, education, reporting, investigating and sanctioning.
These specific provisions have been scrutinized by the CAS in its increasingly abundant jurisprudence, which we address hereunder.

A brief inventory of the CAS jurisprudence on match-fixing
The section below summarizes each CAS award presented in chronological order based on the date of each award, with key facts and findings of each Panel.

Köllerer, CAS 2011/A/2490 57
The Appellant, a professional tennis player, had, on the basis of testimonies, been sanctioned with permanent ineligibility and fined USD 100,000 by the Anti-Corruption Hearing Officer ("ACHO"), for three counts of attempted match-fixing, inviting other players to deliberately lose matches and offering them USD 10,000 each time. 58 The Panel was satisfied that it was more likely than not that the Appellant had indeed attempted to fix several matches, even if unsuccessful. 59 (1) Oral testimonies of fellow players and the Appellant's own manager (not the Appellant's own inconsistent one), 60 (2) voice recognition and (3) the lack of incentive to frame him were relied on (given his contention of impersonation). 61 The Panel upheld the life-ban citing tennis' susceptibility to fixing (fewer athletes needing to be corrupted) and the deterrent effect of exemplary punishment, 62 but considered any additional financial penalties disproportionate, his difficult financial situation already having been affected. 63

Savic, CAS 2011/A/2621 64
The Appellant, a professional tennis player, was sanctioned with permanent ineligibility and fined USD 100,000 by the ACHO 65 for offering a fellow competitor USD 30,000 to lose the first set against himself, on which he could win the remaining two 66 . The Panel relied on (1) witness testimony of the player approached, consistent before different fora,

Butt, CAS 2011/A/2364 77
In the same context as Asif (above) the Appellant, captain of the Pakistani cricket team, was sanctioned for not reporting (but not for actual fixing) with 10 years of ineligibility (five suspended, conditional on no further breach, as above), 78 with parallel conviction by English criminal courts. 79 (1) Accuracy of the information given to the reporter and its recording, (2) timing of contact between the Appellant and agent, and (3) critical contribution to executing the fix, among other facts, were relied on by ICC's Tribunal. The Panel upheld the sanction (established liability not having been contested), 80 reiterating deference to sports bodies' expertise (with only rare deviation on fact 81 ) to gauge both disproportionality (here, held proportionate, given (1) captaincy, (2) prior bans, (3) unchallenged establishment as "ring-master", (4) leniency in parallel criminal proceedings, (5) no apology issued, (6) establishment of financial benefit being unrequired, and thus irrelevant to sanction); 82 and irrationality ("obviously or self-evidently unreasonable or perverse", a high threshold, not met here given ICC's rationale for length of sanction being relative to other code provisions, impact on the game and prioritization of offences 83 ). 84

Olympiakos Volou, TAS 2011/A/2528 85
Based on allegations of fixing and illegal betting by the President of Olympiakos Volou in one match in the Greek Super League, UEFA's Appeals Body sanctioned the club with ineligibility to participate in the 2011-2 Europa League. The Hellenic Football Federation (HFF) had found only the President, but not players, involved in manipulation. 86 The Panel considered HFF's decision, the President's preventive detention and reports on betting, 87 to conclude that UEFA had proven the club's involvement to justify ineligibility, 88 even if players were not involved. 89 They confirmed UEFA's ability to rely on national federation findings, and held that pending procedure before other fora was not grounds to suspend UEFA/CAS decisions. 90

Metalist, CAS 2013/A/3297 91
Having been found guilty of fixing prior, 92  and 2281 concerning the club was challenged before the SFT on 12 August 2013, in which stay was consequently granted (the club contending it needed to be observed in this proceeding as well)-para 2.13.
League by UEFA's Appeals Body based on proved fixing by its official (with strict liability for the club). 93 The Panel held that the Appeals Body was not bound by, or need not enforce prior stayed proceedings irrespective of likely outcome, if otherwise comfortably satisfied in its own determination. 94 Evidence not otherwise admissible could be admissible here given limited resources, even if secured inappropriately (as long as within public policy limitations) to curb fixing. 95 The Panel upheld sanctions 96 as (1) proportionate ("reasonably required in search of a justifiable aim" given the importance of fighting fixing, preserving confidence and integrity, and deterrence) 97 burden to show disproportionality being on the club; 98 ; and (2) not discriminatory (equal treatment of same facts). 99

Besiktas, CAS 2013/A/3258 100
The CAS appeal arose from a UEFA Appeals Body decision holding Besiktas ineligible for the 2013-4 Europa League based on its officials' involvement in fixing the 2011 Turkish Super Lig final. 101 The Panel noted UEFA's ability to use other decisions to corroborate, supplement and confirm, without being bound, when applied on a case by case basis. 102 The (1) club's sporting and financial interest in winning, 103 (2) recorded conversations with tendency to avoid using direct language/reference to fixing, (3) private meetings held, 104 (4) oral and written testimonies of those familiar with events and the accused, 105 (5) incentives to frame/lie and (6) plausibility of alternative explanation, inter alia, 106 were relied on to find the officials, and thus Besiktas, directly/indirectly involved in fixing, the Panel upholding the issued sanctions. 107 Lack of clarity on direct/indirect involvement was held to not be fatal as an eligibility determination was not sanctionary in nature. 108

Fenerbahçe, CAS 2013/A/3256 109
This CAS appeal arose from a 2013 UEFA Appeals Body decision sanctioning Fenerbahçe with exclusion from two UEFA club competitions for which they qualified 110 (reduced from three, by the CDB 111 ) due to proven instances of bribes awarded to lose games in five matches, based on parallel criminal action initiated by Turkish authorities. 112 Finding that clubs could be sanctioned even if evidence adduced was insufficient to sanction specific individuals (before or concurrently), 113 the Panel found, inter alia, that (1) wire-taps (each assessed for credibility, and of "crucial importance"), as adequate to draw inference from, and corroborative to facts otherwise asserted; 114 (2) lack of suspicious on-field behaviour and silent match-reports, or lack of financial records showing money transferred were not proof of fixing; 115 (3) there was no evidence that the match was actually influenced 116 and or of effective change in the outcome was needed; 117 and (4) (2) the Appellant was responsible for the fix (noting that (a) a witness's criminal past did not impact credibility, but motivation to make false statements-which the players possessed-did, 133 (b) the footballer's specific identification thrice before UEFA and German authorities; 134 and (c) his key on-field errors which led to the first goal). 135 On sanctions the Panel noted that both life-time and shorter bans had been awarded by federations for this offence, though CAS had consistently upheld life bans awarded by federations. 136 Yet, the ban was mitigated back to 10 years ineligibility, as the Panel found no proof of individual involvement in "actual implementation" which was more relevant to sanctioning (a mere error not proving intent to allow a goal). 137

Eskişehirspor, CAS 2014/A/3628 138
The CAS appeal arose from a UEFA Appeals Body decision sanctioning Eskişehirspor with ineligibility for the 2014-5 Europa League based on findings of fixing against club officials/players in the Turkish Super Lig (Fenerbahçe's case above). 139 The Panel, held, inter alia, that the law could not predict all acts, 140 i.e. actions, otherwise legal, could influence match outcome. In this case, bonuses from third parties to play well had influenced competition, skewed player motivation, and thus could imply undue advantage, infringing fair play. 141 Relying on (1) wiretaps (presumed to be coded deliberately), (2) secret meetings (drawing inference from their nature), (3) timing and nature of connected acts, (4) corroborating parallel judgements (irrespective of whether final as evidence was considered anew) and (5) noting that evidence not otherwise admissible could be used given the objective of fighting fixing offences, the Panel found matches fixed and the club liable through acts of its official, irrespective of culpability. 142 Sanctions were upheld as justified, proportionate, connected to objectives and not contrary to principles of law, 143 with factors such as lack of negligence/fault considered irrelevant for mitigation given the period of mandatory ineligibility (one year) and nondisciplinary nature of sanctions. 144

de la Rica, CAS 2014/A/3467 145
The Appellant, a professional tennis player, was sanctioned with ineligibility for any ATP event for 5 years, and a fine 121 Para 270 and 297. 122 Being the document where clubs were required to verify/selfdeclare that no fixing was ongoing prior to being admitted to any UEFA competition and the legal basis for eligibility considerations, it being impossible to independently sanction with no legal basis for it-para 215. 123 Paras 568-571-two year period is usual for "standard" offences. The offences here were "particularly serious" due to the number of matches, multiple involved high-ranked officials and top administration orchestrating the process-thus a sanction from the "higher region of the spectrum" was considered warranted. Yet, it could not go beyond the Appeals Body decision ("ultra petita" as UEFA has not filed an independent appeal against that), making two years appropriate-paras 577-578. 124  Here, the CAS appeal arose from FIS's Hearing Panel's 157 suspension of the Appellant, a British-Thai professional skier, for 4 years from its events world-wide for manipulation of results of four 2014 giant slalom Olympic qualifying events she was involved in organizing, 158 based on factors including the events' dates' proximity to qualification deadlines, number of competitors, among others (for example, "Thai National Junior Championships" having the Appellant as a sole competitor, aged 35). 159 The FIS Council cancelled results of these events, noting that the Appellant should not have qualified, asking the IOC to take further action as seen fit. 160 The Panel overturned the decision, finding actions such as (1) irregular starts (due to inconsistent testimony and absent witnesses), (2) asking others to ski slowly (proven but not proved connected to manipulation), match-fixing related ineligibility (akin to licensing and fairplay related eligibility) to be a separate admissions phase determination process before competition started (whereon, the replacement rule is no more applicable). 167 Panathiakos was thus held to lack standing 168 and certainty of selection as a replacement. 169

Pakruojo, CAS 2015/A/4351 170
In this case, based on (1) BFDS and other reports of betting variations in two matches lost to FK Siauliai, 171 (2) the club's form, league positions and turnover rate, (3) player behaviour on and off-field and (4) other factors (loss despite return of a top-scoring player, concession of a penalty, leading bookmakers' removing this match from the market) the LFF's Disciplinary and Appeals Committees sanctioned players with match disqualifications between 8 and 12 months and the club with a fine (strict responsibility). 172 The Panel relied on the (1) BFDS reports (with its expert explanation and demonstrated betting abnormalities) and (2) expert analysis of on-field behaviour (circumstantial to statistics) with no plausible alternative explanation, to conclude "presumed" fixing (needing evidence of fixing and connected player behaviour). 173 Noting that a "well-reasoned sanction" was to be upheld unless "grossly disproportionate", 174 the EUR 4500 fine for Pakruojo's strict liability per match 175 was confirmed.

Skënderbeu, CAS 2016/A/4650 176
The UEFA Appeal's Body decision holding Skënderbeu ineligible (based on BFDS reports of multiple fixed matches) to participate in the 2016-7 Champions League was appealed to CAS. 177 The Panel held that one year's eligibility was not disproportionate, illegal or contrary to public policy, based on the value-based objective of preliminary administrative eligibility determination. 178 They considered quantitative data sets of BFDS reports "valuable evidence if corroborated by further evidence" 179 (here, by video footage of conduct 180 and of betting operators), 181 not definitive to assess fixing by itself but definitive when seen with qualitative expert analysis 182 (of indirect fixing through betting patterns). 183 Noting the (1) similarity to evidence gathering and proof in doping matters, 184 and (2)  Paras 137-141. The Panel noted that players often did not have information in advance, or it was withheld from them until the last minute due to fear of them using the information to skew betting odds, but they assisted in manipulation through on-field actions, for gaining financial benefit for themselves and third parties, used information likely to compromise integrity, did not report the schemes, and did not denounce the behaviour. The Panel also noted that this was likely done due to their non-payment of dues, no professional alternatives, fear of repercussions, and need to provide for themselves and dependents-paras 138-139.

Fantoni and Nunes, CAS 2016/A/4783 214
The EBL's Disciplinary Committee, in its appealed decision, had found two professional bridge players guilty of manipulation by having pre-arranged an illicit method of communication effectively used to exchange information 215 by placing the lead card "vertically" to indicate having unseen high honours (A, K, Q) in that suit and otherwise placing the cards horizontally. 216 They were (1) banned from EBL events for five years, (2) banned from playing as a team for life and (3) issued a fine of CHF 20,000 each. The Panel, (1) noting misleading inconsistencies/false positives in the code, (2) relying on expert opinion on utility thereof to win, (3) the inability to comprehend the entire code and (4) data selection and analysis flaws in assessment of the code, (5) with no proven patent advantage, 217 among other factors, reversed the decision based on inadequate evidence adduced before it, stating however, that "this did not mean the players were not guilty" 218 and imposing costs, as they considered that the player's "unusual" behaviour had led to the proceedings. 219

Skënderbeu II, CAS 2017/A/5272 220
The AFA's Ethics Committee had found Skënderbeu matches manipulated for betting gains, despite no establishment of the required direct or indirect player involvement, resulting in removal of their 2015-2016 Kategoria Superiore title, point deduction for the next season and fine of ALL two million, based on just BFDS data on over 50 matches, including local ones, and connected player behaviour. 221 In this appealed decision, the Committee held that where the regulations were inadequate, in disciplinary matters, a competent authority had power to act as a legislator. 222 The Panel found the "negative" formulation of the non-requirement of player involvement to find a club liable (in absence of what was, in-fact, required for liability) problematic for the level of certainty 223 necessary for disciplinary sanctions, 224 with a lower threshold only permissible for administrative/provisional cases. 225 BFDS data were considered a "valuable tool", but its admissibility was not considered in detail given the lack of legal basis for sanctioning in the first place. 226 Even if considered, the Panel found that another club, and not Skënderbeu, would be liable. 227

Lao Toyota, CAS 2018/A/5500 228
AFC's Entry Control Body ("ECB") held Lao Toyota FC, winners of the 2017 Lao Premier League, ineligible for AFC Cup playoffs due to manipulation proven before the ECB, but the AFC's Disciplinary Committee ("DC") had decided and dismissed the same claim before under different regulations. 229 The Panel held that clubs could not be tried twice by the same federation on the same facts (in this case, the DC's prior disciplinary process and ECB's later administrative one). 230 The ECB could not "cure" the defect in the DC decision, the second decision being both unforeseeable to the accused persons in the rules, and no reservation having been made to this effect by the DC prior. 231

Keramuddin, CAS 2019/A/6388 232
This case is not causally related to match-fixing but gives some interesting insight on the Panel's concept of a certain "hierarchy" of seriousness of offences in sport. the gravity of such (fixing/bribery) offences was "far less severe" than those affecting human rights, dignity and integrity of young players in this case. 235

Labuts, CAS 2018/A/6075 236
In the latest CAS case concerning match-fixing, the Appellant, a Latvian goalkeeper playing for Athlone Town AFC in the second-tier league in Ireland, was banned for one year from any football-related activities for allegedly participating in the fix of a match lost 3:1 against Longford Town FC. 237 The FAI Disciplinary Committee considered, based on suspicious betting activity in relation to several aspects of the match, that (1) the match was fixed, and (2) that the Appellant had taken part in the fix, by deliberately failing to stop two goals (at 39′48″ and 92′31″) from the opposite team. 238 The player strongly denied the accusations and no evidence was provided as to any financial irregularities in the player's situation. 239 The three experts called to judge upon the goalkeeper's defensive mistakes were not unanimous in distinguishing between a deliberate behaviour and mere manifestations of inadequacy and fatigue. 240 It was, however, established that the Appellant made a save (at 87′) and that two central defenders of the Appellant's team were also underperforming, which notably contributed to the second goal being scored. 241 Based on these facts and expertise, the Panel was comfortably satisfied that the match was fixed, but not that the Appellant was actually involved in the fix. 242 In the light of these awards, we shall examine in the following lines the way in which the CAS has described the notion of match-fixing, the most relevant evidentiary issues extracted from these cases and the sanctions applied for the different manipulation offences.

The notion of competition manipulation in the CAS jurisprudence
As previously noted, CAS jurisprudence has amplified in match-fixing cases, thus giving the Panels several opportunities to carve out the definition of match-fixing at a judicial level, in parallel to the legislative efforts to define this concept (notably, by the Macolin Convention).
As a reminder, in the Macolin Convention, the manipulation of sports competitions is defined as an "intentional arrangement, act or omission aimed at an improper alteration of the result or the course of a sports competition in order to remove all or part of the unpredictable nature of the aforementioned sports competition with a view to obtaining an undue advantage for oneself or for others" (under Art. 3). Importantly, this definition covers not only manipulations altering the result of a match or competition, but also any manipulation which influences "the natural and fair course" thereof, notably through a foul, penalty or action on the field altering the intermediate result or phase of the game. This is also called "micro-manipulation" or "spot-fixing". Finally, it should be noted that Art. 7 of the Convention encourages sports organizations and competition organizers to: (1) prohibit competition stakeholders from betting on sports competitions in which they are involved; (2) prohibit competition stakeholders from misusing inside information; and (3) require competition stakeholders to report any suspicious activity immediately.
CAS jurisprudence has addressed the issue of matchfixing in a similar manner, using a wide angle and a broad definition of match-fixing. In the following lines, we will mention the main elements, ratione materiae and ratione personae, extracted from this jurisprudence.

Ratione materiae
As has been pointed out by several authors, 243 the material scope of the definition of match-fixing in CAS jurisprudence is quite amorphous. Notably, many types of unlawful or unfair conduct may result in match-fixing; certain conduct may be used to conclude "presumed" match-fixing 244 and conduct otherwise lawful might be considered illegal in a manipulation context. 245 To date, among others, the following elements are set as indicators of match-fixing in the CAS jurisprudence: • bribing or attempting to bribe a referee (Lisboa and Guimaraes, CAS 2008/A/1583); 246 • approaching competitors and offering them money (EUR 10,000 each time) in order to deliberately lose a tennis match (Köllerer, CAS 2011/A/2490); • approaching a competitor and offering him money (USD 30,000) 246 The evolution of the definition under applicable UEFA regulations, making up a large number of the CAS awards described above, was examined in detail in Garcia Silvero (2018, p. 9). match, while winning the second and third set, and the match (Savic, CAS 2011/A/2621); • receiving a bribe (between EUR 50,000 and EUR 60,000) as a referee, in order to fix several football matches (O, CAS 2010/A/2172); • receiving a bribe as a goalkeeper, in order to fix several football matches (N and V, CAS 2010/A/2266); • intentionally losing a first football match and arranging a draw for the second, while abnormal betting patterns were detected for those matches, i.e. ten times the usual amount of money for such a match (Pobeda, CAS 2009/A/1920); • deliberate underplay on the field by cricket players (deliberate "no ball" and playing the maiden over), exactly as agreed in a fix which had been arranged with the players' agent, who was promised GBP 140,000 for the fix (Asif, CAS 2011/A/2362 and Butt, CAS 2011/A/2364); • deliberate poor play and consequent loss of a tennis match on receipt of EUR 15,000 (de la Rica, CAS 2014/A/3467) and proven responsibility for a loss through actions on the field for payout from betting gains (Sammut, CAS 2013/A/3062); and • deliberate mistakes (in casu, not proven) by a goalkeeper allowing the opposite team to score two goals in the context of highly abnormal betting patterns (Labuts, CAS 2018/A/6075).
It is thus apparent that the CAS panels have taken generally the same broad perspective on match-fixing as international legislators, both ratione materiae and ratione personae, in an effort to effectively tackle the multi-faceted phenomenon of competition manipulation, which may manifest itself in many different forms, ultimately aiming at ensuring a fixed result or event, and thereby eliminating the element of uncertainty which should be inherent to any sports competition.

Selected evidentiary issues
Evidentiary issues are obviously paramount in the fight against match-fixing, 247 given the very nature of this offence and its broad scope, both ratione materiae and ratione personae. In the following lines, we will focus on a selection of such issues, notably standard of proof and admissible means of evidence, as extracted from the CAS jurisprudence. 248 In the jurisprudence briefly described above, the CAS panels have made it clear that nothing is worse than creating a false positive and issuing sanctions to an innocent person or club. Apart from the recurring question of the burden of the proof, 249 the panels focus primarily on issues of the standard of proof, 250 as well as admissibility and means of gathering evidence.

Standard of proof
As far as evidence is concerned, what is most important is not the evidence itself, but that it is capable of proving the facts sufficiently in the eyes of the panel. Indeed, in order to be convinced, the panels apply a high standard of proof. Even if a criminal standard is considered inapplicable in the context of private associations, the panels need to be comfortably satisfied that a conduct of a type which undermines the basic premise of fairness upon which all sporting contests are premised actually happened, i.e. at a level greater than a mere balance of probability, but less than proof beyond reasonable doubt. 251 Considering the different nature of criminal procedures and disciplinary ones, relative lack of investigatory powers of sports governing bodies compared to those of a criminal prosecutors, incentive to conceal such evidence 252 (and consequent similarity to doping 253 ) and within the context of seriousness of allegations and importance of fighting corruption, 254 CAS has repeatedly dismissed the application of the "beyond reasonable doubt" 255 standard, to apply the criteria of the "comfortable satisfaction". The latter is much 247 See also Rigozzi and Quinn (2014). 248 These issues were analysed immediately after Pobeda in Barak and Koolard (2014) as well as, after, by Palermo and Williams (2018, p. 17). 249 On this subject, see, for example Kaplow (2012). 250 On this subject, see, for example Davis (1994). 251 About the comfortable satisfaction standard of proof, see, for example, Davis (2012), Deakes (2014), Duttig |(2018). 252 Notably stated first in O (para 21) and cited by numerous panels thereafter. 253 See, for example, O (citing Pobeda) paras 20-21 and Fenerbahçe, paras 278-279. 254 See for example Besiktas, paras 108-112. 255 On this subject, see, for example, Newman (1993). more flexible than the former but still more stringent than a simple balance of probabilities. As it has been repeatedly stated, the "comfortable satisfaction" standard therefore lies somewhere in between beyond reasonable doubt and a mere balance of probability. 256 Panels themselves have observed the standard to be inconsistent and difficult to follow. 257 This is also because, prior to applying a comfortable satisfaction standard, a panel must first look at the concerned federation's guidelines for an expressly chosen standard of proof ("preponderance of probability" in tennis, for example 258 ). Parties (clubs or athletes) are presumed to have "consented" to such standard contractually. 259 Factors to judge appropriateness of such standard include how close it is to comfortable satisfaction (for example, held suitably proximate in the case of tennis) or whether it contravenes public policy. CAS itself, however, lacks authority to harmonize such standard. 260 This has now been confirmed in Labuts, noting that where the applicable rules are silent on the standard and the parties do not agree on a common standard, the CAS is to apply the standard used in prior awards. 261 The fact remains that given the seriousness of an allegation of manipulation of a sports competition, the comfortable satisfaction criteria are considered fulfilled if there is a high degree of confidence in the quality of the evidence, and if the latter has been collected legitimately. 262 This indicates, therefore, that the standard lies closer to the test of beyond a reasonable doubt than to that of a mere balance of probability. 263

Means of evidence
As has already been written, unless the applicable sport regulations contain specific evidentiary rules, CAS panels determine evidentiary issues according to the procedural rules adopted by each panel. 264 In making determinations on evidentiary issues, the CAS arbitrators are in no way bound to apply the procedural rules that would be applicable in a Swiss court, i.e. evidence not otherwise admissible might be considered admissible and relied on. 265 Their freedom is only limited by the obligation to make sure that the contemplated procedural rule complies with the necessity for equal treatment of the parties and the right of both parties to be heard (under Art. 182(3) of the Swiss Private International Law Act), and is consistent with "procedural public policy". 266 In practice, in the match-fixing cases handled by the CAS, multiple types of evidence were used to try to reach the standard of proof described above. Across these awards, among the most frequently put forward, 267 one finds: • suspicious betting patterns and conducts provided by betting operators; • written evidence (rarely present); • witness statements and confessions; • expert evidence analysing betting patterns and on field behaviour; • wiretaps, intercepted and taped phone or Skype conversations (sometimes with a coded language), their frequency and the point of time they were made; • video recordings of on-field behaviour and interactions among parties; • evidence of unneeded, private and suspicious meetings; • evidence from parallel criminal investigations and connected decisions; • rumours and anonymous letters; • parallel circumstances (notings in match reports, potential financial benefit or a club's poor financial condition); and • coincidence between what was agreed between fixers and what happened on the field, result of the games, on field performances.
While most of these means of proof are not very different from those used in other types of sport offences, there is one specific to the manipulation: suspicious betting patterns, 256 See, for example, Sivasspor at para. 131-132 and Skënderbeu at para. 64; see also, Rigozzi and Quinn (2014, p. 24). 257 See Fenerbahçe at paras 275-277. 258 Köllerer at paras 6 and 29, and Savic at paras 8.4 and 8.6. 259 Köllerer at paras 38 and 40 (player consent) and Skënderbeu at para 64 (clubs' consent in UEFA's admission forms). 260 Köllerer and Skënderbeu, as above; see also, Vanakorn at paras 92-93. 261 Labuts at para 46. 262 Köllerer at para 62, Vanakorn at paras 94 and 98. 263 As also concluded by Rigozzi and Quinn (2014, p. 27). 264 See, for example, Rigozzi and Quinn (2014, p. 4). 265 See SFT decisions concerning Ukrainian match-fixing (including club Metalist) A. v. Football Federation of Ukraine ("FFU"), 4A_448/2013and X. v. FFU, 4A_362/2013, both decisions of 27 March 2014and-see also, von Segesser and Leimbacher (2014. 266 See, for example, Rigozzi and Quinn (2014, p. 4); see also Metalist at paras 8.10-8.11, Köllerer at para 29 and Vanakorn at paras 92-93. The SFT, in an appeal from the award in Metalist confirmed the ability to use such evidence-see Beffa and Ducrey (2015, pp. 122-123). 267 See also Palermo and Williams (2018, p. 20) where the CAS awards' observations per specific type of evidence have been elaborated. often based on a Betting Fraud Detection System ("BFDS") report. 268 Such reports are founded on algorithms and mathematical analysis that are able to quantitatively indicate if the betting patterns of a particular sports competition present a statistically significant deviation from the expected model. Illogical betting behaviour is thus detected even before the sports competition takes place and an alert is triggered. 269 These alerts, based on quantitative data and analysis, are then put to the test of the qualitative assessment that observers will be able to make of the course of the sporting competition. If the on-field action confirms the abnormal betting patterns and cannot be explained other than by the participants' willingness to satisfy the bettors, it will be considered-with a high degree of certainty-that the competition has been manipulated. In order not to create false positives, an assessment of the on-field situation by three experts is generally required. On the contrary, if the quantitative analyses of abnormal betting patterns and the qualitative observations do not strongly match (or do not match at all), it will be considered that the BFDS has malfunctioned and the case will be dismissed. 270 Such reports have been considered "valuable" evidence, particularly when specific individual responsibility can be attributed through it. 271 Yet, they are not definite by themselves, with no regulations governing them, requiring support of qualitative expert analysis 272 and corroborating evidence such as on-field conduct or that of betting operators. 273 Importantly, this application was confirmed most recently in Labuts, where the Panel almost solely relied on these reports and expert interpretation thereof to conclude a match was manipulated, 274 but countered this evidence with analysis of other on-field behaviour to conclude that the appellant goalkeeper was not himself primarily responsible for the fix. 275

Sanctions
Here too, there are some questions of principle that the panels must resolve before deciding on the merits of a sanction. 276 First, the principle of nullum crimen, nulla poena sine lege scripta et certa 277 (no sanction may be imposed unless there is an express provision describing in sufficient clarity and specificity, not only the misconduct but also the applicable sanction) is considered to be fundamental; and second, the question of the ne bis in idem principle 278 is applied with the difficulty that some decisions are disciplinary ones and other administrative or civil ones.
The CAS awards on manipulation of sports competitions mainly concern football, but also tennis, skiing, bridge and cricket. As previously mentioned, they affect various factors such as clubs, staff members, referees and so forth. The principal sanctions with which the CAS has dealt in these awards are: • bans, ineligibility for some competitions and suspension (of a person, of a club, of a referee; for one year, for a couple of years, and even life-long); • fines; • disqualification of a player; • game annulment or result cancellation; and • point deductions and relegation.
Some cases also involved criminal sanctions imposed by the concerned State (prison and/or fine), but the CAS has nothing to do with them. We briefly note here, in passing, that such sanctions are provided in national regulations and are extremely disparate. To give just a few examples, fixing the result of a match may be sanctioned, in Australia, Greece or Poland, by imprisonment up to 10 years, whereas in Denmark the maximum sanction is of 1 year. 279 In each case, and contingent on express specific of a criterion for sanctioning, if any, under the applicable regulations, 280 the CAS panels have to determine if the sanction imposed by the respective governing body is adequate, not illegal, not contrary to public policy, appropriate to the level of guilt and the gravity of the violation, 281 and proportionate-which means that it has to be reasonably required in search of a justifiable aim. 282 In addition, it has to be nondiscriminatory in the sense that an equal treatment has to be given to similar situations. 283 A panel ordinarily defers to the sanction imposed by a federation given their expertise in governing their sport and objectives unless the sanction is found, based on the above criteria, "evidently and grossly disproportionate". 284 Thus, where the sanction is a life-time ban, the CAS generally considers that the fact of adding a fine to it makes the whole sanction disproportionate because the ban already has a financial effect by affecting the future earnings of the condemned, while also considering that where integrity was sought to be protected, harming individual privacy and a sports person's development was justified. 285 Life-time bans were also mitigated in absence of evidence of extent of actual involvement in the fix. 286 Exemplary life-time bans were also considered suitable given a sport's peculiar susceptibility to fixing. 287 Still, as regards proportionality, an alleviation of the sports sanction cannot be obtained by alleging that a prison sentence has been pronounced in the same case, whereas that criminal sentence has been mitigated because a sports sanction would also be pronounced. This would be tantamount to benefiting twice from the same mitigating circumstance. 288 Finally, doping's scale of sanctions has also been used as guidance again. 289 Moreover, CAS awards steer clear of making a link between the standard of proof and the sanction. It is not the case, therefore, that a lower sanction would directly be warranted because a higher degree of evidence was not available. 290 The standard of evidence is always applicable the same way, being the need to prove guilt which leads to a sanction that flows from the degree of guilt (and not from the strength of the evidence). Yet, where evidence was inadequate for CAS's satisfaction, despite reversal of lower fora conviction, costs were imposed, as it was the suspicious behaviour which led to the proceedings in the first place and the Panel could not rule out wrongdoing. 291 All in all, in the field of manipulation of sports competitions, CAS awards confirm, to a very large extent, the sanctions imposed by the lower sporting bodies.

Conclusion
The review of these thirty CAS awards allows us to draw several conclusions and to identify a few open questions.
First, from a quantitative perspective, we found that while there was a peak of CAS cases in the years 2011-2014, there was a decline since 2015. Seeing the entry into force of the Macolin Convention in 2019, might trigger the conclusion that this instrument comes too late, since the phenomenon has been decreasing for several years now. Obviously, such a view would be tantamount to forgetting that the CAS awards only represent the tip of the iceberg. For its part, the Macolin Convention goes much further than the questions dealt with in the CAS awards, notably because it tends to combat the phenomenon on the wide scale of all sports at all levels, and not only top-level sport (Fig. 1).
Second, we observe that there is a very strong overrepresentation of cases related to football, with a smaller number of cases involving tennis and cricket.Nevertheless, the existence of match fixing cases in other sports, such as skiing and bridge, shows that the phenomenon is clearly not restricted to aparticular sport or category of sports (teams or individuals, for example) (Fig. 2).   Third, the ne bis in idem principle (double jeopardy) as applied by the CAS panels requires, in our opinion, a more in-depth consideration in future awards. Indeed, the panels are regularly confronted with cases in which administrative, disciplinary, penal and/or civil sanctions have been pronounced for the same case of manipulation of sports competition 292 and at different levels (national and international, courts and sports bodies, administrative or disciplinary). 293 Recurrently, the CAS panels conclude that there is no violation of the principle of ne bis in idem by considering that the various sanctions are not of the same nature and are therefore not likely to constitute double jeopardy, Lao Toyota being a notable exception. 294 Nevertheless, it remains worrisome to note that the same person may be subject to several sanctions for the same act on the sole basis of the difference in the legal nature of each of them. The result of such an interpretation of the ne bis in idem principle is all the more questionable when it leads to several sanctions that not only bear the same name (for example, "fine"), but that are exactly the same from the individual's perspective (since they all affect his/her patrimony, in the example here above), even if their legal nature is indeed different.
Sometimes-nevertheless only if the legal nature of both sanctions is the same-the CAS corrects this problem by considering that the double penalty becomes disproportionate. While considered necessary for setting an example, particularly in susceptible sports on the one hand, on the other, for example, a financial penalty in addition to a life-time ban, is considered excessive as the latter would, in any case, have an additional financial effect on a player by affecting his/her future earnings 295 .
The issue of life bans (isolated from any other sanctions) may be delicate in the high-performance sporting context, where an athlete's career only lasts for a few years; therefore, it can be argued that a ban for eight years (similar to the maximum ban applicable in doping offences) is efficient and deterrent enough. 296 Life-time bans have also been considered disproportionate based on the existence of mitigating factors, such as no proof of actual involvement in the fix and the effect on an athlete's career. 297 Fourth, the principle of legal certainty as apparent from the CAS jurisprudence is worth a more in-depth consideration. Generally, most CAS awards refer to prior awards on various aspects, including in awards related to matchfixing, and panels are unlikely to depart significantly from this practice. 298 Though not obligated to follow precedent, they tend to do so in the interest of legal certainty. 299 It has been concluded that this is closer to the civil law practice of "jurisprudence constante" (or non-deviation except when clear error or injustice shall result) than common law's stare decisis. 300 Across the studied match-fixing awards, starting with Benfica and Guimarães, CAS panels have referenced prior awards noting the importance of certainty, but chosen whether to deviate or not from them, per case.
This notably triggers a potential problem regarding the consistent interpretation of the same provisions (as seen in the awards involving UEFA regulations, notably in Fenerbahçe and Skënderbeu), standard of proof, and quantum of sanction (in cases arising having the same facts such as those of Asif and Butt, or within the same sport, such as for lifebans in tennis as in Savic, or for the same objective of deterrence given the value of preserving integrity). Also, panels have regularly borrowed from doping jurisprudence, to lend further consistency across CAS disciplinary sanctions; yet, at the same time, they note categorically that the strict degree of certainty as in criminal procedure is unrequired given the hybrid nature of proceedings as seen in Skënderbeu II. In Phnom Penh, for instance, the panel categorically noted prior awards on attribution of officials' acts to a club but chose to deviate from them.
Given the recurrence of match-fixing cases, the particular nature of this threat and the current absence of a harmonized approach at the level of the international sport regulators, it is thus probable that CAS jurisprudence will continue to develop in the coming years, thus setting new milestones for the contouring of the rather "amorphous" current framework.
Funding Open access funding provided by University of Neuchâtel. Open Access funding provided by Lib4RI -Library for the Research 292 Most recently in Lao Toyota where the different levels of applicability of the principle and exceptions were discussed. Previously, this question had arisen in awards such as Metalist, Skënderbeu, Trabzonspor, and Sivasspor; see also Palermo and Williams (2018, p. 15). 293 Even where the principle is not explicitly cited such as Asif and Butt. 294 Lao Toyota at paras 39-42 and 51 to 58. 295 As seen in Köllerer at paras 70-73 and Savic at paras 8.36-8.38 and 9.3. 296 See also White (2017) and Rigozzi and Quinn (2018, p. 106)where it is suggested that an approach considering proportionality per case for doping related sanctioning is advisable (CAS awards on manipulation having accepted/suggested applying doping jurisprudence to sanctioning-Fenerbahçe at paras 56-57 and 60). 297 Sammut at paras 179 and 180. 298 As has been seen in the awards summarized, but also other seminal awards (Canadian Olympic Committee and Beckie Scott v. IOC, CAS 2002/O/373, award dated 18 December 2003 at para 14) and widely opined before. See, for example, Koffman-Kohler (2007, p. 366). 299 Blackshaw (2009, p. 155). 300 See conclusion based on empirical study confirming past scholarship in Bergasel (2012).
Institutes within the ETH Domain: Eawag, Empa, PSI & WSL. This research has been made possible due to the support of the Swiss National Science Foundation (SNSF-100011_192497).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.