Fulvio Fantoni, Claudio Nunes v. European Bridge League (EBL)

Arbitration CAS 2016/A/4783 Fulvio Fantoni & Claudio Nunes v. European Bridge League (EBL), award of 9 January 2018

Panel: Mr Lucas Anderes (Switzerland), President; Mr José Juan Pintó (Spain); Prof. Ulrich Haas (Germany)

Bridge Match-Fixing Article 56 CAS Code Tempus regit actum Article 73(B) (2) of the Laws of Duplicate Bridge “Comfortable satisfaction”

1. Documents which one of the parties to CAS proceedings tries to introduce as evidence into the proceedings from the very beginning of the CAS proceedings do not qualify as “new” exhibits in the meaning of Article R56 of the CAS Code.

2. As a general rule, transitional or inter-temporal issues are governed by the principle of “tempus regit actum”, according to which any deed should be regulated in accordance with the law in force at the time it occurred. As a consequence, procedural actions, such as the filing of an appeal, should be done in compliance with rules and time limits in force when they are performed.

3. According to the wording of Article 73(B) (2) of the Laws of Duplicate Bridge, an illicit behaviour punishable under those rules is composed of two elements: first, the players prearrange a (illicit) method of communication and second, this method is effectively used for exchanging information.

4. The standard of proof of “comfortable satisfaction” has been defined by CAS jurisprudence by comparison, declaring that it is greater than a mere balance of probability but less than proof beyond a reasonable doubt. In particular, CAS jurisprudence has clearly established that to reach this comfortable satisfaction, the panel should have in mind “the seriousness of allegation which is made”. It follows from the above that this standard of proof is a kind of sliding scale, based on the allegations at stake: the more serious the allegation and its consequences, the higher certainty (level of proof) the panel would require to be “comfortably satisfied”.

I. PARTIES

1. Messrs Fulvio Fantoni and Claudio Nunes (hereinafter, the “Appellants” or the “Players”) are professional duplicate bridge players affiliated to the Fédération Monégasque de Bridge and to the Federazione Italiana Gioco Bridge.

2. The European Bridge League (hereinafter, the “Respondent” or the “EBL”) is an association with its registered office in Lausanne, Switzerland, and is the governing body for organising professional bridge tournaments in Europe, in particular, the 2014 EBL European Championships that took place in Opatija, Croatia from 21 June 2014 to 1 July 2014 (hereinafter, the “Competition”).

II. FACTUAL BACKGROUND

3. A summary of the most relevant facts and the background giving rise to the present dispute will be developed based on the parties’ written submissions, the evidence filed with these submissions, and the statements made by the parties and the evidence taken at the hearing held in the present case. Additional facts may be set out, where relevant, in connection with the legal discussion which follows. The Panel refers in its Award only to the submissions and evidence it considers necessary to explain its reasoning. The Panel, however, has considered all the factual allegations, legal arguments, and evidence submitted by the parties in the present proceedings.

4. The Players participated in the Competition representing Monaco, as a pair, in the duplicate bridge tournament. During the Competition, the Players participated in 13 matches against several teams.

5. After the Competition, Ms. Maaijke Mevius (hereinafter, Ms. Mevius), an amateur Dutch player, analyzed 6 of the 13 matches of the Monaco team and, in her opinion, considered that the Appellants were placing the opening lead in a non-natural manner. She then formed a hypothesis for a “signalled meaning”, which was that when the Appellants had a “defensive” position, they placed the lead card vertically to indicate that they had “unseen high honors” (i.e. A, K, Q) in that suit. Otherwise, the lead card was placed horizontally.

6. In September 2015, Ms. Mevius sent her observations to Mr. Kit Woosley (a professional Bridge player) in order to test her hypothesis.

7. Mr. Woosley published his analysis in a website and developed the following two hypothesis taking into account the placement of the cards (emphasis added):

“I examined the following hypothesis regarding the way Fantoni-Nunes placed their opening lead on the table: • The card is placed vertically when: ○ The led suit contains an unseen high honor (ace, king, or queen) OR ○ The led card is a side-suit singleton in a trump contract

• Otherwise, the card is placed horizontally”.

For the development of such hypotheses he analysed the information from 10 out of the 13 matches of the Competition and concluded that “[o]n the 82 of the 85 hands [leads], this simple hypothesis correctly predicted orientation of the opening lead”.

8. After these accusations, the EBL appointed a committee to investigate the performance of the Appellants during the Competition (hereinafter, the “Investigation Committee”).

III. THE INVESTIGATION PHASE AND THE DISCIPLINARY PROCEEDINGS BEFORE THE EBL

9. On 25 September 2015, the Investigation Committee sent a letter to the Players which, in its relevant parts, reads as follows (emphasis added):

“[t]he European Bridge League (EBL) has initiated investigation of allegations that you have consistently exchanged information about cards held using illicit prearranged methods (…). This is considered by the Laws of Duplicate Bridge § 73 B.2 as “the gravest possible offense”. (…). The EBL has established an Investigation Committee (…). Our mandate is to assemble the available evidence, present it to you for comment, and then have it analyzed by an international expert panel. (…). As first step we hereby provide you with the enclosed evidence which has been collated”.

Furthermore, the Investigation Committee attached a document that contained, inter alia, information regarding the following:

➢ The position (vertical-horizontal) of the “defensive” cards (opening leads and third hands) played in 12 games of the Competition (in total 323 “defensive” cards) and the “holding” information of each of these cards. ➢ Two hypothesis of the alleged exchange of information, which was explained as follows (emphasis added): “When defending, the way Fantoni-Nunes orient their cards the first time they lead a suit communicates encouragement or discouragement in the suit. A vertical position of the card typically indicates an unseen top honor (A/K/Q) or a singleton which can lead to a ruff in a suit contract. A horizontal position of the card signals discouragement. This either denies an unseen top honor or a singleton, or signals that the defender would not like this suit continued”. (For ease of comprehension, the Panel will refer to the hypothesis in which vertical leads mean high honors, as the “High Honor Hypothesis” or the “Code”; and, to the hypothesis in which vertical leads entail a “singleton”, as the “Singleton Hypothesis”. For the hypothesis that includes the High Honor Hypothesis and the Singleton Hypothesis the Panel will refer to it as the “Combined Hypothesis”). ➢ The first findings of the Investigation Committee which read as follows: “This signalling method is used both by the partner who leads the suit and by the partner who follows the suit (third hand). (…). These signals are 97.17% accurate on lead, 94% at trick 1,

84.69% after trick 1 and 90.97% accurate overall”.

10. On 20 November 2015, the Appellants denied the Investigation Committee’s accusations and attached two experts’ reports:

➢ Mario Colombo’s report dated 18 November 2015 (hereinafter, M. Colombo) that, inter alia, contained the following (emphasis added): “(…) 4. (…) In order to define what is Vertical and what is Horizontal, an unquestionable classification should be used. Besides any consideration on the fact that some cards do not keep same orientation across time in video (meaning that they could be considered temporarily horizontal and temporarily vertical), physics clearly states that it is not possible to calculate the exact angle of an object through the 3D to 2D transformation of video unless are known shooting distance, orientation angle, focal length and optical aberration of the lens. 5. Some cards have been classified ambiguous; therefore the choice of the one who wrote the report was simply to exclude them not counting these cards in the sample. This is absolutely arbitrary and statistically unacceptable because of three reasons: a. If a real code exists, it has to be applicable to all cases, without discarding any of them according to marginal utility. An ambiguous card being neither clearly horizontal nor clearly vertical implies: i. either the definition of a clear rule for ambiguous cards ii. or the definition of the rule as a dichotomy (Vertical vs. Non Vertical, Horizontal vs. Non Horizontal). (…). Therefore, (…) the conclusion is that the alleged code is just a coincidental opinion and not scientific evidence, since it is based on a restricted sample of deals, ambiguous, biased and not unique, while several unbiased and unambiguous patterns based on the same restricted sample of deals were found, suitable to match data at 100%”.

➢ Angel Mateo Salvatierra’s report dated 19 November 2015 (hereinafter, Salvatierra) that, inter alia, stated the following (emphasis added): “(…) I should state that the alleged signal based on the angle with which the card is placed on the table raises serious doubts, of which I would highlight the following: A) It is unclear what accuracy and adjustment the card needs to have in order to be considered as vertical or horizontal, which introduces doubtful positions as fits for the alleged signal, which cannot be clearly rejected, thereby increasing the level of deviation of the results. B) In some cases, the card is initially placed in one position but this subsequently changes, without it being clear which position determines the signal. C) All players place cards on the table in a certain position that, with the exception of the diagonals, will inevitably be closer to the vertical or horizontal position. If a player also usually places the card vertically, this would also apply to the reference cards in that suit on more than 55% of occasions (…).

E) It should be noted that looking for the probability of a fit to a certain code or pattern in a specific number of hands and cards played is not the same as looking for an alleged code that matches the cards played and the hands in question. F) (…) the existence –or otherwise– of a code cannot be determined by simply applying the range of fits that is determined subsequently, since one cannot refer to a code when the same signal could fit other different codes. G) It has been pointed out to me that even opening leads in singleton suits have been discounted in order to increase the number of fits, when apparently this could have a high degree of significance on the game if it is used to transmit unauthorized information at the same time.

(…) I have no reason to doubt it that among these other patterns there are some that could be much more relevant in terms of transmitting unauthorized information. It seems logical, since the first card played certainly fits an authorized information code and, according to what I have been told, in at least 28 cards (of 106 leads - 26.41%) the card placed in a vertical or horizontal position could only transmit a single type of authorized information to the exclusion of any other information that could be considered as unauthorized. (…) I have been able to observe a high percentage of fits (26.41%) in relation to these 28 cards, which can only lead to the conclusion that a simple fit, in itself, means nothing. (…). 4. CONCLUSIONS (…) cards defined as ambiguous must be included in the sample and their exclusion only leads to a reduction in the representativeness and reliability of the sample”.

11. Given the Appellants’ objection, the Investigation Committee appointed the statistical experts Mr. Nicolas Hammond, Prof. Greg Lawler and Prof. Peter Buchen (hereinafter, “EBL Experts”) whom, in their three respective reports, concluded that the probability of a random placement of the cards was almost impossible. A brief summary of these reports is listed below:

➢ Nicholas Hammond’s report dated 30 December 2015 (hereinafter, “Hammond 1”) which, in its relevant parts, reads as follows (emphasis added): “(…) In my review I include all 13 matches – 114 boards when they are on lead. (…). It can be difficult to determine with some leads if they are horizontal or vertical. I chose between vertical or horizontal even though some leads were clearly diagonal leads – I chose the closest to the two options. (…). My rule was to write down if it was closer to horizontal or closer to vertical from the players’ perspective. There are 114 videos with Mr. Fantoni/Mr. Nunes on lead. 1 is a pass out, 1 is ignored. Of the 112 remaining leads, 7 are singleton leads, 105 are not. Statistics/Probability (Singleton Leads) For the singleton leads 6 of 7 leads matched the hypothesis. There does not appear to be sufficient data on singleton leads - it is possible that this is “observational bias”. (…).

With only 7 singleton leads, one of which is a false positive, there is not sufficient data to agree with the hypothesis beyond a reasonable doubt. Statistics/Probability (Horizontal/Vertical Leads) (…) We will take all 112 leads, including the singleton leads as this benefits Mr. Fantoni/Mr. Nunes. Of these 112 leads, there are 47 vertical leads (I am including the singleton leads). (…). Of these 112 leads, there are 45 leads with an unseen higher honor. I am going to ignore the additional hypothesis of singleton leads (these benefits Mr. Fantoni/Mr. Nunes). On the 47 vertical leads, 42 of them have a high honor. (…). The mathematical equivalent is an urn containing 112 colored balls. 45 of these balls are blue (contain a high honor), 67 of them are red (do not contain a high honor). We are going to take 47 balls (all of the vertical leads) out of the urn without replacing any balls. What are the odds that we will select 42 blue balls (high honor) and 5 red balls (no high honor)? (…). Using the well-known formula for selecting colored balls from an urn, we see that the probability of selecting 42 blue balls, 4 red balls from an urn containing 45 blue balls, 67 red balls is 0.000000000000000000001480447211. Expressed another way this is 1 in 675,471,568,771,037,000,000. This number was generating using the most favorable results for Mr. Fantoni/Mr. Nunes – including the singleton leads, ignoring the singleton hypothesis. The red balls represent the false positives – the one documented vertical lead with false positive and 4 additional singleton leads. (…). Other Tournaments I looked at one video (…) from the Bermuda Bowl 2013 as an example. Mr. Fantoni/Mr. Nunes were on lead 9 times. (…). For all the 9 leads (none were from a singleton) the orientation of the lead matched the hypothesis. There are other videos from this tournament. If there is any doubt then it appears that these videos can be used as additional tests of the hypotheses. (…). Conclusion The hypothesis of singleton leads is not proven beyond a reasonable doubt. There is 1 false positive out of 7 examples. With the hypothesis of vertical leads showing a high honor, I did everything possible to make the numbers more favorable to Mr. Fantoni/Mr. Nunes by including the singleton leads, rejecting the singleton hypothesis. The odds are 1 in 675,471,568,771,037,000,000 of their leads being random and matching the hypothesis. (…). The hypothesis of vertical leads showing a high honor is shown beyond any reasonable doubt”.

➢ Peter Buchen’s report dated 31 December 2015 (hereinafter, “Buchen 1”), which in its relevant parts reads as follows (emphasis added): “(…) While it is not possible with statistical analysis alone to conclude with absolute certainty the existence of illegal signalling, the data may provide an indication of such to varying degrees of confidence. The analysis

depends wholly on the data provided and involves no bridge judgement. Thus the conclusions drawn from the analysis are completely unbiased with respect to bridge considerations. However it should be made clear that errors or bias in the data can affect the results of the analysis. […]. The data involved 85 deals in which 82 matches with the hypothesized signal were observed. The number of matches observed is particularly high, so even without calculating the precise statistics, one might reasonably guess that the probability of obtaining that many matches would be quite small if the FN leading mannerisms were in fact random. (…). (…). It is a practical certainty that the hypothesized signal encoded by leading a card vertically or horizontally is determined by the signaller’s holding in the given suit. (…) this conclusion is independent of bridge considerations but depends strongly on the accuracy of the data input into the statistical analysis”.

➢ Greg Lawler’s undated report (hereinafter, “Lawler”) which, inter alia, reads as follows (emphasis added): “I will assume that there were 85 signals given with 82 matches. There was a slight bias for horizontal leads (the estimated value of pH is 45/85 = .529 ···). To be very conservative we will choose q = .6, that is, we claim that the true probability of getting a horizontal play is between .4 and .6. This is less than 10-14 (it is so small that there is no reason to give the actual value). This is a very straightforward mathematical computation, but it must be understood in context of the particular situation. My knowledge of F-N is entirely from information on the Bridge Winners site and my personal observation of the videos. All of my observations were after the code was “cracked” by Maaijke Mevius and others and tabulated and checked by Kit Woolsey. Indeed, like most bridge players, the first I heard about this accusation was in Mr. Woolsey’s article. I do not know how many matches were viewed by Ms. Mevius and the others in cracking the code. However, even if all of them were used, if we said that the number of possible codes viewed (the M of the previous section) is at most a million, then chance that any of these codes would fit so well is less than 10-8. For this reason, I agree with the statement. (…). Beyond any reasonable doubt, the actions of F-N were not random and there was signalling. I am not saying that we need to determine the entire code that F-N were using beyond a reasonable doubt. Indeed, I agree with Mr. Hammond that the code for leads from singletons has not been proved significantly. I also would discard the “diagonal” leads that could well have a meaning that we do not know”.

➢ Nicholas Hammond’s report dated 14 January 2016 (hereinafter, “Hammond 2”) which explained the method he used to determine the orientation of the cards.

12. Given the outcome of the EBL Experts’ reports, the Investigation Committee recommended that the EBL initiate a disciplinary procedure against the Appellants.

13. Consequently, the EBL constituted a Disciplinary Committee composed of Mr. Jurica Caric (as President), Mr. Po Sundelin and Mr. Rex Anderson (hereinafter, the “Disciplinary Committee”).

14. On 15 April 2016, the EBL submitted its formal allegation against the Players before the Disciplinary Committee (hereinafter, the “Formal Allegation”), which can be summarized as follows:

➢ Hammond 1 analysed the 114 leads of the 13 matches of the Competition, however, two cards were excluded from the sample: the first one due to a multiple placement of Fantoni and the other one because it was a “pass out”. Taking into account the remaining 112 leads, Hammond 1 tested the Singleton Hypothesis and the High Honor Hypothesis.

o The Singleton Hypothesis meant that the card is placed vertically “when the led [sic] card is a side-suit singleton in a trump contract”.

▪ From the 7 singletons played in the 112 leads, 6 confirmed the hypothesis and 1 is a false positive. Notwithstanding the latter, Hammond 1 discarded this hypothesis as he maintained that the evidence “was not sufficiently robust since there is one false positive out of seven leads” and the standard of proof he applied was “beyond reasonable doubt”.

o The High Honors Hypothesis meant that “the card is placed vertically when the led (sic) suit contains an unseen high honor (Ace, King or Queen)”; otherwise, the card is placed horizontally.

• “Without considering the seven singleton leads, there are 105 remaining leads: 62 were horizontal and 43 were vertical. Amongst these 105 leads, expert Hammond found three horizontal false positive (i.e. these suits contained an unseen high honor) and two vertical false positive (these suits did not contain an unseen high honor)”.
• Hammond 1 concluded that the High Honors Hypothesis was confirmed since the “odds are 1 in 675,471,568,771,037,000,000 of their leads being random and matching the hypothesis”. Lawler and Buchen are in basic agreement with the calculation. Therefore, the Code exists.

➢ Hammond 2 has proven that the positions of the analysed leads are correct and although there may be differences, these are minor and do not affect the outcome of the study.

➢ Taking into account the aforementioned, the EBL concluded that the Players have used a pre-arranged method of communication infringing Article 3.1 of the EBL Disciplinary Code and Law 73 of the Laws of Duplicate Bridge. Considering that the Players’ illicit behavior is described as the gravest offense possible in the Laws of Duplicate Bridge, the EBL recommended that the Players were sanctioned with (i) a lifetime ban from competing in the EBL tournaments as a pair, (ii) an individual ban from competing in the EBL tournaments for period of no less than 12 years, and iii) a fine of CHF 25,000 for each of the Players.

15. On 17 June 2016, the Appellants objected to the Formal Allegation and filed the following expert reports before the Disciplinary Committee:

• Miguel Angel Canela’s report dated 1 June 2016 (hereinafter, “Canela”) that, inter alia, stated the following (emphasis added):

“This hypothesis was supported by a data set covering 323 cards (…) but the new data set includes only openings leads. (…). The new data set contains 114 observations, but Mr Hammond drops 2 of them (one is pass out and in the other one Mr Fantoni changed the initial orientation of the card), leaving us with 112 observations. Not only the data are changed, but also the hypothesis, which now associates the vertical position of the card to the unseen honours, dropping the singleton. (…). Requisites for valid inference (…). The first requisite is that the hypothesis has to be developed independently of the data used to test it, either from previous data or from a theory. Data always confirm hypotheses that have been extracted from the same data. In the academic jargon, we call this “fishing for a hypothesis”. (…).

Mr. Buchen’s report (…). It may seem that p = 0.5 is an obvious choice, but it is not so. It would be if the players were always placing cards either horizontally or vertically, with equal probability. None of these things is true. First, players do not put cards only vertically or horizontally. They put them with a certain angle (discussed later). Some of these cards are close to vertical, some close to horizontal, and some in between. This last case has been excluded from Mr Buchen’s data by a previous sample selection, or forcing the cards to be vertical or horizontal. So he is careful warning the reader about the data. (…).

Mr. Hammonds report (…). Mr Hammond suggests that Messrs Fantoni and Nunes could exclude the cards for which they disagree on the orientation. But the problem is not about disagreement on the orientation. It is that it is unclear that there is an orientation. Moreover, this would be a further sample selection, and accepting that there is a code. Because it turns out that the potential code is not a proper code, but an intermittent code. This, together with the fact that after forcing the cards into the vertical/horizontal dichotomy, some of cards do not fit the presumed code, leaves us with a code such that:

• sometimes one player cheats the other player,
• sometimes includes ambiguous cards.

I do not see how such a code could work.

An alternative code To evade the “testing on the same sample” question, Mr Lawler says in the last page of his report that, even if other codes may be found, the probability of finding a code that fits as well as the proposed one is

less than 1 over 100 million. Since I came interested in this assertion, the defense gave me the following code: the card is placed vertically when

• The lead suit contains an ace and a 4 or
• The lead suit contains a king or
• The lead suit is made of five cards in the suit without a jack or
• Nunes leads with a singleton And horizontally otherwise. (…).

Conclusion I have pointed out the inconsistencies in the evidence presented against Messrs Fantoni and Nunes, which I summarize as follows:

• The circular logic in testing a hypothesis on the same data used to develop it, and the arguments used by the experts to evade this point.
• The changes in the hypothesis and the data.
• The fact that the selection of the data led to a biased data set.
• The arbitrariness of classifying cards as vertical or horizontal even in dubious cases”.

• Carlo Colombo’s report dated 17 June 2016 (hereinafter, “C. Colombo”): This report basically established (i) the correct angles of the cards when the Players placed them on the table and (ii) that the classification of the cards should have been horizontal, vertical and diagonal. The expert concluded that “the number of fits can be significantly reduced by virtually any sensible classification system. It is worth noting that fit reduction percentages could be even dramatically higher that those reported in the examples, if other classification criteria and parameters were used”.

16. On 18 July 2016, the Disciplinary Committee decided the following (hereinafter, the “Appealed Decision”):

“(…)

1. Messrs. Fulvio Fantoni and Claudio Nunes are banned from participating in EBL events or activities as individual players for a period of five (5) years.

2. Messrs Fulvio Fantoni and Claudio Nunes are banned from participating in EBL events or activities as a pair playing together for life.

3. Messrs Fulvio Fantoni and Claudio Nunes shall bear the costs of the present proceedings including the costs incurred by the EBL in the investigation and prosecution phase in the amount of 20.000 EUR each.

4. All other motions or prayers for relief are dismissed”.

17. On 25 August 2016, the EBL notified the grounds of the Appealed Decision to the Appellants, which can be summarized as follows:

• The EBL has the burden of proof and the applicable standard is “comfortable satisfaction”.

• The Investigation Committee analysed the leads of the 13 matches of the Competition; however, the Disciplinary Committee agrees with the Appellants that data of the matches used to discover the Code should not be used to verify the Code’s existence. Thus, the only considered data are the 64 lead cards in the matches against Bulgaria [11 boards], Denmark [8 boards], Germany [8 boards], Iceland [8 boards], Ireland [10 boards], Israel [11 boards] and Romania [9 boards].

• After analyzing the videos and the experts’ written and oral evidence related to mathematics and probability, it is concluded that the Players violated Article 3 of the EBL Disciplinary Code when they exchanged information through a pre-arranged method of communication, also forbidden by Article 73 of the Laws of Duplicate Bridge.

• In relation to the Appellants’ claim that the Code has not been used in all situations and that the allegations should therefore be dropped, the Disciplinary Committee is of the opinion that not using the Code may sometimes be used to protect the Code and that cannot constitute the basis for the dismissal of the allegations.

• The Commission considers that the statistics expert opinions provided by the EBL are convincing and that they support the conclusions.

• Considering that the Appellants’ conduct contravenes the spirit of bridge and the essence of any sporting competition and that the length of a career in bridge is longer than other sports, the Players are sanctioned with (i) an individual ban for a period of 5 years from any EBL activity, (ii) a life-time ban playing as a pair in any EBL activity and (iii) with a fine of CHF 20.000 for each player.

IV. PROCEEDINGS BEFORE THE COURT OF ARBITRATION FOR SPORT

18. On 15 September 2016, the Appellants filed the relevant Statement of Appeal against the Appealed Decision before the Court of Arbitration for Sport (“CAS”), pursuant to Articles R47 and R48 of the Code of Sports-related Arbitration (the “CAS Code”).

19. In the same Statement of Appeal, the Appellants appointed Mr. José Juan Pintó as arbitrator and requested the CAS to order the EBL “to produce the full case file concerning the Appellants”.

20. On 20 September 2016, the EBL rejected the Appellants’ request regarding the production of the case file by stating that it had already been presented during the disciplinary proceedings and that the production of documents should be made in the appeal brief.

21. On 27 September 2016, the EBL informed the CAS Court Office that it appointed Dr. Ulrich

Haas as arbitrator.

22. On 17 October 2016, the Appellants filed their Appeal Brief, pursuant to Article R51 of the CAS Code.

23. On 24 October 2016, in accordance with Article R54 of the CAS Code, the CAS Court Office informed the parties that the Panel appointed to settle the present dispute had been constituted as follows:

President: Mr. Lucas Anderes, attorney-at-law in Zurich, Switzerland. Arbitrators: Mr. José Juan Pintó, attorney-at-law in Barcelona, Spain. Mr. Ulrich Haas, professor of law in Zurich, Switzerland.

24. On 1 December 2016, the EBL filed its Answer, pursuant to Article R55 of the CAS Code.

25. On 2 December 2016, the CAS Court Office advised the parties that, pursuant to Article R56 of the CAS Code, they were no longer authorized to supplement or amend their requests or their arguments, produce new exhibits, or specify further evidence on which they intended to rely.

26. On 14 December 2016, the CAS Court Office informed the parties that, pursuant to Article R57 of the CAS Code, the Panel had decided to hold a hearing in the present procedure. Moreover, the Appellants were invited to inform the CAS Court Office, whether they maintained their request to produce the “full case file” or not.

27. On 15 December 2016, the CAS Court Office advised the parties that Mr. Roberto Nájera Reyes, attorney-at-law in Barcelona, Spain would assist the Panel as Ad Hoc Clerk.

28. On 19 December 2016, the Appellants informed the CAS Court Office that (i) they objected to the Respondent’s proposed testimonies of Ms. Mevius and Mr. Bas Drijver (bridge expert); (ii) they requested the Panel to exclude all the arguments regarding “bridge logics” brought by the Respondent; (iii) they requested the Panel to exclude a third report of Nicholas Hammond (hereinafter “Hammond 3”) filed in the Answer to the Appeal since, in their view, the information therein was available before the Appealed Decision; and, (iv) they maintained their request to produce the first instance file.

29. On 21 December 2016, the Respondent filed its comments on the Appellants’ objections with the CAS. Regarding the production of the first instance file, the Respondent stated that the Players already had it in their possession but “[i]f the Players wish to produce any elements of the case file that have not already been produced, the EBL [would] not object”. The Respondent also objected to the Appellants’ request to exclude the testimonies of Ms. Mevius and Mr. Drijver, the arguments regarding “bridge logics” and Hammond 3.

30. On 19 January 2017, the CAS Court Office informed the parties that the Panel had decided: (i) to dismiss the Appellant’s objection to the testimonies of Ms. Mevius and Mr. Drijver; (ii) to invite the Appellants to name an experienced bridge player in order to provide testimony on

Mr. Drijver’s statements; (iii) to reject the Appellants’ request regarding the exclusion of the “bridge logics” arguments brought by the Respondent; (iv) to deny the Appellants’ request regarding the exclusion of Hammond 3; (iv) to order the Respondent to produce the EBL file and deliver it directly to the Appellants who, in turn, had 10 days to submit the documents (which were not yet part of the CAS file) on which they intended to rely.

31. On 27 January 2017, the CAS Court Office informed the parties that the hearing for the present case was set on 26 and 27 April 2017.

32. On 6 February 2017, the Appellants appointed Mr. Carlos Fernandez Sanchiz as bridge expert in order to provide testimony on Mr. Drijver’s statements.

33. On 23 February 2017, the Appellants filed before the CAS Court Office the documents from the first instance file that they considered relevant for the present case.

34. On 6 March 2017, the Respondent objected to the documents filed by the Appellants since, in its view, all of them were in their possession before these CAS proceedings and no exceptional circumstances were met for admitting the documents.

35. On 23 March 2017, the CAS Court Office informed the parties that the Panel had decided to admit the documents filed by the Appellants and the reasons for this decision would be given in the present award (see infra para. 100)

36. On 25 and 26 April 2017, the parties returned a duly signed copy of the Order of Procedure.

37. The hearing of the present procedure took place in Lausanne, Switzerland, on 26 April 2017. The Appellants were represented by their lawyers Mr. Jorge Ibarrola and Ms. Natalie St. Cyr Clarke. The Respondent was represented by Mr. Eric Laurant, President of the Investigation Committee, and by its external lawyers, Messrs. Ross Wenzel and Nicolas Zbinden. In addition, Mr. José Luis Andrade, counsel to the CAS and Mr. Roberto Nájera Reyes, Ad Hoc Clerk, assisted the Panel at the hearing.

38. At the outset of the hearing, both parties confirmed that they had no objections as to the composition of the Panel and did not object to the jurisdiction of the CAS.

39. After the respective opening statements, the following persons were examined by the Panel in order of appearance:

➢ Witness testimony:

• Maaijke Mevius (via tele-conference).

➢ Bridge expert conference:

• Mr. Bas Drijver (via tele-conference); and,

• Mr. Carlos Fernández Sanchiz (in person), who was assisted by a translator, Ms. Silvia Borja Pardo.

➢ Statistical and Data expert conference:

• Mr. Nicholas Hammond (in person).
• Ms. Marjolaine Viret (in person).
• Prof. Kenneth Regan (via video-conference).
• Prof. Greg Lawler (via video-conference).
• Prof. Peter Buchen (via video-conference).

40. During the hearing, the parties had the opportunity to present their case, to submit their arguments and to answer the questions posed by the Panel. At the end of the hearing, all the parties expressly declared that they did not have any objection with respect to the procedure and that their right to be heard had been fully respected.

V. SUMMARY OF THE PARTIES’ SUBMISSIONS

41. The following summary of the parties’ positions is illustrative only and does not necessarily comprise each and every contention put forward by the parties. The Panel, however, has carefully considered, for the purposes of the legal analysis which follows, all the submissions made by the parties, even if there is no specific reference to those submissions in the following summary.

A. The Appellants

42. In their Appeal Brief, the Appellants attached two new expert reports:

➢ Marjolaine Viret’s report dated 17 October 2016 (hereinafter, “Viret”) that, inter alia, stated the following:

“8. […] In particular, these expert reports analised the probability of observing the data assuming the hypothesis of randomness, but their conclusions express the probability that the players were cheating given the data observed. This represents a fallacy of the transposed conditional (also commonly known as prosecutor’s fallacy). (…). 13. (…) Statistical evidence can contribute valuable information to a case in court, but this is subject to: The validity of the methodology and of data collection; The relevance of the statistical evidence for the case and the manner it is communicated to the panel. (…). 23. Incorporating the statistical evidence into the overall evidentiary framework of the case is typically the

final step in the panel’s weighing of the evidence. (…). 31. (…) Experts can only make statements with respect to evidence [E] they have been provided with. Only the legal panel is in a position to assess the probability of the hypothesis [H] at stake in the case. This is because only the panel has an overall view of the evidence on the record necessary to appropriately weigh that evidence. 34. (…) The panel cannot simply endorse the expert’s conclusions without assessing whether those conclusions are soundly rooted in the expert’s analysis. (…)”. ➢ Kenneth Regan’s report dated 17 October 2016 (hereinafter, “Regan”) that, inter alia, stated the following:

“(…) (a) The data-taking may have primary bias in the classification of data items, so that the match rate m observed in the accusation should be lowered. (b) There may be unaccounted elective bias in how the statistical tests were constructed - especially when key elements of the construction were determined after the fact. (c) The null-hypothesis modelling may need adjustments that entail a statistical test of lower sensitivity than the one described, requiring a calculation of Pr(R | H) that gives less extreme results”.

43. The Appellants’ submissions, in essence, may be summarized as follows:

The formal invalidity of the Appealed Decision

44. Mr. Jurica Caric was appointed as the President of the Disciplinary Committee which issued the Appealed Decision. However, Mr. Caric is also a member of the Executive Committee of the EBL and, according to the EBL Statutes (2016), none of the members of the Executive Committee can be part of the Disciplinary Committee. Thus, the Disciplinary Commission was chaired by a member who was not allowed to be part of it and therefore the Appealed Decision shall be annulled.

Burden and Standard of Proof

45. It is undisputed between the parties that the EBL carries the burden of proof to establish, to a standard of “comfortable satisfaction”, that the Appellants committed an illicit act. However, the CAS jurisprudence has established that when the accusations are serious, “the comfortable satisfactions standard may not be much different from the standard of beyond reasonable doubt” (CAS

46. In other words, the facts and arguments raised by the EBL shall be considered to be proven if the Panel’s comfortable satisfaction reaches the level of “beyond any reasonable doubt”. Otherwise, the Panel must rule in favour of the Appellants.

47. Furthermore, the arbitrators of this case – and not the experts – are the ones that shall assess whether the standard of proof is reached or not and, thus, the Panel shall disregard any expert opinion that concludes that the Code is proven “beyond reasonable doubt” or similar.

The statistical evidence by itself cannot prove the illicit conduct

48. Although statistics can be a valid piece of evidence they cannot be the only piece of evidence when assessing the culpability of the accused. Statistics show presumptions and suspicions; however, these cannot be enough to prove a fact to the comfortable satisfaction of the Panel. Statistics must be accompanied by other kinds of evidence such as motives, alibis, testimonies, etc., which have not been presented in this case.

49. It is true that the panel of the case CAS 2015/A/4351 sanctioned the appellants based on “suspicious conducts” that resulted from statistical studies but this was only because the disciplinary code of the Lithuanian federation allowed it to do so. In casu, the EBL regulations do not permit sanctioning based on suspicions and, for this reason only, the Appealed Decision shall be annulled as there is no direct evidence proving the alleged Code.

The inexistence of the Code and the lack of reliability of the EBL’s evidence

50. The EBL has invented a Code that simply does not work because the cards are not always placed in a vertical or horizontal way and, thus, the leads positioned diagonally would only mislead the partner about the alleged given information. Furthermore, there are several plays where the alleged Code does not make sense and even in some others the Players act against the information they allegedly get from the Code. Lastly, the Slavinsky Code cannot be reconciled with the alleged Code, since the former considers the “Jack” as a high honor and the latter does not.

51. The Disciplinary Committee concluded that the Appellants cheated by only trusting on the statistical analysis of the EBL Experts which has been seriously challenged by M. Colombo, Salvatierra, Canela, C. Colombo. These expert reports show that:

i) There are several mistakes in the hypothesis and the method to calculate the coincidences was incorrect:

• The cards are not always placed in a vertical or horizontal way, and the cards positioned diagonally were excluded from the statistical studies. Thus, the classification (vertical-horizontal) is subjective, arbitrary and unfair.
• Salvatierra and C. Colombo established that if a proper and accurate measure of calculation is used the coincidence percentage drops from 90.64% to 83.9%. If the sample size were raised, the percentage would be much lower.
• The games played against France and Holland were excluded from the formal allegation without motive. According to M. Colombo and C. Colombo, if these games were included in the analysis the coincidence percentage would also be lower.
• Canela and Salvatierra have established that the method used by the EBL is incorrect since it is based on the probability of causality falling far below 100%; the evidence supporting an accusation of culpability shall be based on probability as close to a certainty or 100%, if possible.

• Canela considered that the EBL provided fabricated evidence based on tests with erroneous and manipulated data.
• M. Colombo considered that the EBL’s observations are subject to errors and those margin errors should have been taken into consideration in the calculations.
• Lawler stated that we cannot talk about randomness when the Appellants’ natural position for placing their cards and further kind of behavior has not been previously analyzed.
• Furthermore, Regan stated that the EBL classified the cards in a biased way, the statistical tests were construed with key elements determined after the facts, and the hypothesis need to be adjusted which would provide less extreme results.

ii) The vertical-horizontal classification is merely subjective and unreliable;

• Canela rejected the subjective classification of the cards as horizontal and vertical because the Players also placed cards in a diagonal way; therefore, this “black-white” classification is completely arbitrary, superficial and unfair.

• C. Colombo stated that depending on the selected video-frame a card may have several orientations because the leads did not remain in the same position throughout the play.

iii) The arbitrary omission of part of the sample;

• Several games and cards were excluded from the data sample which results in bias and subjective information. Canela considered that there have to be observations of a random sample for valid conclusions.
• Salvatierra questioned the reliability of the sample because the “ambiguous” cards (the ones that could not be categorized as vertical-horizontal) were arbitrarily excluded from the sample. This exclusion of cards will always result in a higher percentage of “fits” of the alleged Code. For example, “taken to the extreme, if 322 cards out of 323 are considered to be ambiguous with only one fit, this formula would conclude that the percentage of fits is 100%”. A contrario sensu, while the greater the sample is, the lower Code coincidences would appear. Even Buchen (EBL expert) agrees that the statistical results “depend[s] strongly on the accuracy of the data input into the statistical analysis”. Therefore, the most logical step is to include all the “ambiguous” cards in the studies for a reliable conclusion.
• The EBL has made a significant effort to raise the percentage of coincidences when it has arbitrarily excluded more than 30 cards from the studies. Moreover, the Investigation Committee discarded the Singleton Hypothesis in the middle of the proceedings with the single purpose of raising the coincidence percentage.

iv) The hypothesis was tested with the same data used to develop it

• The information used to develop the hypothesis was the same used to confirm it. Canela considers there is a “circular logic” (also known as “fishing the hypothesis”) because “data always confirm hypotheses that have been extracted from the same data”.
• The correct method would have been to demonstrate the hypothesis with new information (“fresh data”) and to consider all the defensive cards in the sample (not only the leads). Canela has stated that the EBL has repeatedly tried to “fish a hypothesis”.
• According to Canela the EBL experts “calculate the probability of obtaining the observed events by chance, find a small value, and conclude that these events did not happen by chance”. Canela even proposes an alternative code comparable to Hammond’s code showing that “a hypothesis extracted from a data set can be proved on the same data beyond any reasonable doubt (…)”. In other words, there can be different accusations against the Players using the same correlation vertical-horizontal.

v) The value of the statistical analysis:

• Viret has argued that the experts should normally assess the “probability to observe the evidence assuming a hypothesis” and the deciding bodies should look for the “probability that a hypothesis is true, based on the overall evidence”.
• According to Viret, the Disciplinary Committee has concluded from the statistical evidence to what is known in criminal law as the “prosecutor’s fallacy” i.e “the fallacy of equating (1) the probability of the presence of the evidence, assuming innocence, to (2) the probability of innocence, assuming presence of the evidence”. A simple example of this would be to say that “there is a probability of 1:1mio to observe that data if the accused is innocent, therefore the probability that the accused is innocent is 1:1mio. This conclusion while intuitive appealing, can be extremely deceptive and is logically and mathematically wrong. (…) the statement is often directly followed by shift in the object of probability: therefore, there is an inversely high probability (999’999:1mio) that the accused is guilty”.
• CAS panels confronted with complex scientific evidence have repeatedly insisted on their autonomy from the views of the experts appointed by the parties, in particular in connection with Athlete Biological Passport cases: even if the panels cannot be required to completely “step into the specialist’s shoes”, “it cannot abdicate its adjudicative role” either.
• Viret stated that the EBL experts included in their conclusions that the Appellants “cheated beyond reasonable doubt”. However, the Panel members – and not the experts – are those who shall give value to the evidence taking into account the entire file.
• Thus, Viret considered that the EBL experts; (i) failed to provide the probability of the ultimate issue (probability of the hypothesis given the evidence);(ii) failed to provide the Panel with the necessary data to estimate that probability; (iii) confused

the probabilities analysed with those contained in the conclusions; (iv) used inconsistent and imprecise wording.

52. In light of the foregoing and as the famous saying states “[t]here are three kinds of lies: lies, damned lies, and statistics” or as Sir Winston Churchill said, “I only believe in statistics that I doctored myself”, the Panel shall conclude that the statistical evidence brought to these proceedings is biased, flawed and thus not reliable, and the sanctions imposed on the Appellants shall be annulled.

Disproportionality of the sanctions

53. If the Panel ultimately considers that the Appellants have breached the regulations, it shall consider that the repeated CAS jurisprudence has established that “to be proportionate, the sanction must not exceed that which is reasonably required in the search of the justifiable aim” (CAS 2005/A/976 & 986). Hence, the sanction should not go beyond what is required to discourage the misconduct from occurring in the future.

54. It is evident that the sanction imposed in the Appealed Decision is disproportionate and shall be reduced, particularly, because the Appellants have always played as a pair in duplicate bridge and the sanction infringes their possibility to obtain profitable gains for their activities as professionals.

55. The Appellants made the following requests for relief:

“Fulvio Fantoni and Claudio Nunes therefore request that the Court of Arbitration for Sport rules as follows: i. This appeal is upheld. ii. The Decision issued on 18 July 2016 by the Disciplinary Commission of the European Bridge League is annulled.

Alternatively iii. The Decision issued on 18 July 2016 by the Disciplinary Commission of the European Bridge League is set aside. iv. Fulvio Fantoni and Claudio Nunes have not committed any violation of the rules of the European Bridge League and shall not be imposed any sanction.

In any event v. The European Bridge League shall bear all the arbitration costs, if any, and shall be ordered to reimburse Fulvio Fantoni and Claudio Nunes all advances of costs paid to the CAS, included the minimum CAS Court Office fee of CHF 1000. vi. The European Bridge League shall be ordered to pay Fulvio Fantoni and Claudio Nunes a contribution towards the legal and/or other costs incurred in the framework of these proceedings in an amount to be determined at the discretion of the Panel”.

B. The Respondent

56. In its Answer to the Appeal, the EBL attached two new expert reports:

➢ Peter Buchen’s report dated 29 November 2016 (hereinafter, “Buchen 2”); and,

➢ Nicholas Hammond’s report dated 20 November 2016 (i.e. Hammond 3) which in summary contained (i) an explanation of the games that were used by Ms. Mevius and Mr. Woosley to formulate their respective hypotheses, (ii) a new statistical analysis based on “fresh data” which concluded that the proportion of “coincidences” of the Code were exceptionally low on a random basis (iii) and a basic agreement with the measurements of the cards angles calculated by C. Colombo.

57. The Respondent’s submissions, in essence, may be summarized as follows:

The formal validity of the Appealed Decision

58. The Appellants have attempted to annul the Appealed Decision based on the fact that Mr. Jurica Caric – member of the EBL Executive Committee – was the President of the Disciplinary Committee. Firstly the restriction that the members of the Disciplinary Committee may not be members of any EBL authorities does not apply to the President of such committee but only to the other two persons that form part of it. This is crystal clear from the wording of Article 47 of the previous EBL Statutes which read as follows: “The Disciplinary Commission consists of a president elected by, and from, the members of the Executive Committee, and two other persons who are not members of any of the EBL or WBF Authorities”.

59. Furthermore, according to the principle tempus regitum actum, the Disciplinary Committee was appointed on 29 January 2016 and therefore the applicable provision to the case at stake is the abovementioned Article 47 and not the rules mentioned by the Appellants. The appointment was validly made in accordance with the rules which were in force at the time.

60. In any case, if the EBL committed any procedural flaws during the first instance proceedings, the de novo power of the CAS can “heal” them.

Burden and Standard of proof

61. The Appellants are being accused of using the Code to pass valuable information regarding the hand of the player on opening lead. If proven, the Code would constitute a breach of Article 73B of the Laws of Duplicate Bridge and a reprehensible conduct for purposes of the Disciplinary Code.

62. The EBL accepts that it bears the burden of establishing the violation and that the standard of proof is “comfortable satisfaction”. However, unlike the Appellants’ interpretation, the Panel shall observe that the “comfortable satisfaction” standard is “the standard greater than a balance of probability but less than beyond a reasonable doubt” as it has been consistently considered by the CAS case law.

The discovery of the Code and the formal allegation of the EBL

63. The Panel shall firstly understand that the margins between world-class players are so tight that even a slim advantage can be determining. In light of the foregoing, the laws of the game strictly prohibit the exchange of unauthorized information between players and the EBL is trying to fight these types of conducts. Evidence of this is that besides the Appellants also other pair of players have been investigated:

• Fisher/Schwartz (Israel): They were found guilty and punished with the same sanctions as the Players and they did not appeal their decision.

• Piekarek/Smirnov (Germany): They admitted their wrongdoing at the outset of the investigation and explained that they felt that the only way to compete against other cheating couples was to equalize the conditions. Although they did not mention all the “cheating” couples, they specified that the Appellants were one of them. This pair was sanctioned with a 4-year individual suspension and with a lifetime prohibition to play together.

• Balicki/Zmudzinki (Poland): They are waiting for their decision to be rendered.

64. In the case at stake, Ms. Mevius was the first person to formulate a hypothesis when she noticed the non-natural manner in which the Players placed their “opening leads” from the available videos of the Competition. The hypothesis for the “signal” (the vertical-horizontal position) was formed from the match Monaco v. Russia alone. Furthermore, after reviewing the matches against Poland, the Netherlands and England she discovered the “signaled meaning” (i.e. that the vertical placement indicated an unseen high honor). Lastly, Ms. Mevius checked this hypothesis (vertical = High Honors) by analysing the matches against Poland (already used to discover the meaning), Norway and Sweden (not used for the development of the hypothesis). In short, Ms. Mevius used 6 matches of the Competition for the creation and testing of the Hypothesis High Honors.

65. Later on, Ms. Mevius sent her discoveries to Mr. Woolsey without disclosing the games she had used to develop her hypothesis. Based on Ms. Mevius’ findings, Mr. Woosley then developed the Combined Hypothesis for which he used the data of ten games (Monaco against Bulgaria, Denmark, England, Iceland, Norway, Poland, Romania, Russia and Sweden); he did not include the other remaining games of the Competition because he could not find them on the internet. The conclusions of Woosley were that “[o]n 82 of the 85 hands, this simple hypothesis correctly predicted the orientation of the opening lead”.

66. Following the recommendations of several National Bridge associations, the EBL opened an investigation with respect to these cheating allegations and, on 25 September 2015, the Investigation Committee sent the initial findings (based on 323 “defensive cards”) to the Appellants for their respective comments. However, this was not a formal allegation.

67. Once the Appellants denied the cheating allegations, the Investigation Committee consulted the EBL Experts who made their respective reports by observing only the (114) leads played in the

Competition. Contrary to what the Appellants have alleged, the exclusion of the “third hands” was not with the aim to increase the level of coincidences but with the sole purpose of facilitating the job of the experts since they are volunteers and are not being paid for their work and time.

68. The Formal Allegation filed by the EBL before the Disciplinary Committee stated that the Players “when defending and on opening lead, placed the card vertically to indicate an unseen high honor (A,K,Q) in that suit, and otherwise placed the card horizontally” (i.e. the High Honor Hypothesis). The EBL dropped the Singleton Hypothesis because, according to Hammond 1, there was not sufficient information to demonstrate it. This Formal Allegation has not been changed and is confirmed in these CAS proceedings.

69. In light of the foregoing and having the “whole picture”, the Panel shall disregard the entire Appellants’ arguments with respect to the alleged “arbitrary exclusion of games and cards” and that the EBL changed the hypothesis in the middle of the proceedings.

The value of the Code and the statistical evidence

70. On a “defensive” position bridge players need to make assumptions from the bidding and play about the location of honor cards of the “declarer’s” and “partner’s” hands and, therefore, exchanging unauthorized information about high honors in the led suit is extremely useful and gives the defenders a significant advantage over their opponents.

71. Furthermore, there are legal means of conveying information such as, for example, the Slavinsky code which is used by the Appellants and resolves two of four possibilities of the game. However, using the illicit Code combined with the Slavisnky code would resolve the four possibilities at once. This is why the Panel shall conclude that the Code has great and significant value from the bridge perspective such as has been testified to by Mr. Drijver.

72. The existence of the Code is primarily established through statistics and, in fact, it has been demonstrated by statistical means that the “signal” (vertical or horizontal card orientation) cannot match the “signaled meaning” (high honors in the relevant suit or the lack of them) on a random basis. In short, the odds that the pattern observed is a coincidence (the signal with the meaning), i.e. by chance, are 1 in not hundreds, thousands, millions or billions, but 1 in many many trillions.

73. In particular, Hammond 1 examined the Code in the 13 matches played by the Appellants at the Competition, which amounts to 114 boards where the Players were defending and on opening lead. From these 114 boards, two were excluded: one because it was a pass-out (no opening lead was made) and the other one because Fantoni changed the card position several times. From the 112 remaining leads (including the singletons but ignoring the Singleton Hyphotesis in order to improve the Players’ probabilities) 47 were vertical (42 with high honors) and 65 were horizontal (3 with high honors).

112 Opening Leads (including singletons but Position ignoring the Singleton Hypothesis in order to Horizontal (65) Vertical (47) improve the Players probabilities) Unseen high honors Yes 3 42 (A,K,Q) No 62 5

Hammond 1 explains that the mathematical equivalence of the aforementioned is an urn containing 112 coloured balls: 47 of these balls are blue (containing a high honor) and 65 are red (not containing a high honor). If 47 balls are taken from the urn, what would be the odds to take 42 blues (high honors) and 5 reds (not high honor)? The answer would be 1 in 675,471,568,771,037,000,000; in practical terms, zero.

74. Without considering the 7 singletons, there are 105 remaining leads: 43 verticals (42 with high honors) and 62 horizontals (3 high honors).

105 Opening Leads (excluding singleton leads) Position Horizontal (62) Vertical (43) Unseen high honors Yes 3 42 (A,K,Q) No 59 1

The probability of coincidence between the signal and signalled meaning by chance is 1 in 668,049,083,838,778,000,000,000.

75. Hammond 1 also tested the Code with the videos of the 2013 Bermuda Bowl that took place 10 months before the Competition: all of the 9 opening leads matched with the Code.

76. Lawler agrees in general terms with Hammond 1, since he took into account Woosley´s information (82 “matches” in 85 cards) and concluded that the probability of arriving at this result assuming randomness is less than 10-14. Lawler concluded that all mathematical models show that “the chance that the plays were random is small that it can be treated as virtually impossible”.

77. Buchen 1 also analyzed Woosley’s information (82 “matches” in 85 cards) and concluded that “[i]t is a practical certainty that the hypothesized signal encoded by leading a card vertically or horizontally is determined by the signaller’s holding in the given suit”.

78. These cheating allegations are being proven with statistics that demonstrate the significant correlation between the signal and the signalled meaning and that the probability of observing the relevant data based on random card placement is effectively zero. The Panel shall therefore consider that the Appellants conduct was deliberate.

79. Furthermore, “in view of the nature of the cheating, the EBL is effectively in a position of Beweisnotstand. The cheating allegation cannot be proven by direct means (in the absence of admission from the Appellants) and, in these circumstances, it was on the Appellants to “substantiate and explain in detail why it deems the facts submitted by the other party to be wrong” (see para. 255 of CAS 2011/A/2384-2386). The Appellants have failed to offer an alternative explanation (other than cheating) for the non-random placement of the cards”.

Regarding the objections raised by the Appellants

80. The Appellants have not materially challenged the statistical approaches or the mathematical calculations from the EBL Experts. Even Regan accepts that the “balls in the urn” model from Hammond 1 and the binomial model from Lawler and Bunchen 1 are valid with no meaningful differences.

81. The majority of Appellants’ objections are based on Salvatierra and M. Colombo reports that analysed the initial data and were filed prior Hammond 1, Buchen 1 and Lawler reports. In other words, the Appellants have objected to data and arguments that were never brought before the Disciplinary Committee and, in any case, are redundant and do not modify the EBL Experts’ opinions.

82. Canela proposed another four codes that could match with the positioning of the cards. However, these codes (i) have no valuable information, (ii) could mean one of 4 options (iii) the codes function differently depending on whether Nunes or Fantoni lead and (iv) the information provided has no Bridge logic.

83. Regarding the classification of the cards as horizontal or vertical, Hammond 2 explained the method he used for such classification. There is almost no disagreement between Hammond 2 and C. Colombo. There is discrepancy only in relation to one card and the exclusion of this card makes no material difference to the calculations. It is true that C. Colombo developed a computer program to estimate the cards angles in a more precise way. The EBL does not disagree with C. Colombo’s data and consequently it is willing to accept the proposed angles. Although and even taking C. Colombo’s figures there are very few cards close to the 45 degree boundary (i.e. diagonal position) and the majority are towards the extremes 0 or 90 degrees (i.e. horizontal or vertical).

Range of Angles Number of cards (Total 113) 0-10 45 10-20 12 20-30 7 30-40 3 40-50 0 50-60 3 60-70 2 70-80 12 80-90 29

In light of the above:

• If it is considered a horizontal position between 0 and 30 degrees range and a vertical position between 60 and 90 degrees range, the probability of observing the data assuming random card placements would be 1 in 290, 280,522, 559,451,000,000,000.

• If it is considered a horizontal position between 0 and 20 degrees range and a vertical position between 70 and 90 degrees range, the probability of observing the data assuming random card placements would be 1 in 2,555,746,121,701,220,000,000.

• If it is considered a horizontal position between 0 and 10 degrees range and a vertical position between 80 and 90 degrees range, the probability of observing the data assuming random card placements would be 1 in 18,580,162,080,940,000.

In conclusion, even accepting the angles proposed by C. Colombo, the likelihood of observing the number of “matches” based on random basis is still zero.

84. Regarding the Appellants’ allegation that the EBL is “fishing a hypothesis” because it did not use “fresh data”, it is completely false. The EBL has verified the Code against many more matches as shown below:

➢ If the 6 games used by Ms. Mevius for developing the High Honor Hypothesis are excluded, the 7 remaining games contain 63 leads from which 4 cards shall be excluded: 1 for multiple positioning changes and 3 for being singletons. From the remaining 59 leads there are 27 verticals (all with High Honors) and 32 horizontals (none with High Honors), meaning that the hypothesis matches the cheating allegation.

➢ Furthermore, the Code has been verified with the 2013 Bermuda Bowl data on which 22 available leads were analyzed. From this 22 leads two cards were excluded: one for being a singleton and the other one because there was a twist in the card positioning. Hammond 3 considered that there were 14 horizontal leads (1 with High Honors) and 6 vertical leads (all with High Honors).

➢ If all “fresh data” is combined (with the 7 Competition games not analyzed by Ms. Mevius plus the Bermuda Bowl data) there are 33 vertical leads (all with High Honors) and 46 horizontal leads (1 with High Honors). The probability of this being random is 1 in 550,685,243,114,369,000,000.

For the above, the Panel shall conclude that the EBL was not “fishing a hypothesis” because the Code has also been proven with “fresh data”.

85. Lastly, the few mismatches (or “false positives”) can be explained as a precautionary conduct of the Players in order to avoid the detection of the illicit Code.

Other elements that prove the existence of the Code

86. The Appellants have contended that this case is only based on statistical evidence, however, this is entirely false since there are other elements that prove the cheating of the Players, such as the following:

• The Code is valuable and effective from a Bridge perspective, even more so if the Players use the Slavinsky system.

• The Code is used by the two Players, i.e. the card position and its meaning coincide independently of which player is on lead.

• Both players placed the cards in a range close to 0 or 90 degrees (i.e. in a horizontal or vertical position).

• The Code was not only used in the Competition but also in the Bermuda Bowl.

• The American Contract Bridge League has also sanctioned the Players for using the same Code in the Spingold 2015.

• The Appellants have not presented any alternative reason to explain their card positioning.

• The players Piekarek/Schwarz have identified the Appellants as “dirty pairs”.
• If the card position is not random then it is clear that is deliberate.

87. Moreover, the EBL rejects the contention that statistical evidence cannot, on its own, suffice to establish a disciplinary violation to the comfortable satisfaction of a panel.

The proportionality of the sanction

88. The sanction imposed by the Disciplinary Committee shall be considered proportional taking into account the following:

• The Appellants’ cheating conduct is considered one of most serious offenses in bridge regulations and the Guidelines of the Word Bridge Federation (WBF) recommend a “life ban from playing with that partner and a 10 years suspension from participation in Zonal and /or National events”.

• The EBL asked for a 12-year individual ban and the Disciplinary Committee reduce it down to a 5 year-ban.

• The bridge career is significantly larger in comparison with other sports; indeed, players are able to perform at the elite level into their seventies.

• Other players have received similar sanctions as the one applicable to the Players.

• A severe sanction is necessary to punish the Players and deter others from any form of cheating that jeopardizes the essence of the game and the very values of the sport itself.

89. The Respondent made the following requests for relief:

“…

I. The Appeal filed by the Appellants is dismissed.

II. The Appellants shall bear all the arbitration costs.

III. The Respondent is granted a significant contribution to its legal and other costs incurred in connection with these proceedings, such contribution to be ordered jointly and severally against the Players”.

VI. JURISDICTION

90. Article R47 of the CAS Code reads as follows:

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

91. Article G (4)(a) of the EBL By-laws states the following:

“a) Any decision made by the Disciplinary Commission may be submitted exclusively by way of appeal to the Court of Arbitration for Sport (CAS) in Lausanne, Switzerland, which will resolve the dispute definitively in accordance with the Code of Sports‐related Arbitration”.

92. The jurisdiction of the CAS, which has not been disputed by any party, arises out of Article G (4) of the EBL By-laws, in connection with the abovementioned Article R47 of the CAS Code and is confirmed by the signing of the Order of Procedure by the parties.

93. Therefore, the Panel holds that the CAS has jurisdiction to rule on this case.

VII. ADMISSIBILITY

94. According to Article R49 of the CAS Code, “[i]n the absence of a time limit set in the statutes or regulations of the federation, association or sports-related body concerned, or in a previous agreement, the time limit for appeal shall be twenty-one days from the receipt of the decision appealed against (…)”.

95. The EBL By-laws provide a deadline for an appeal to the CAS in Article G(4)(b) which reads as follows:

“b) The time limit for appeal is twenty‐one days after the receipt of the decision of the Disciplinary Commission”.

96. The grounds of the Appealed Decision were communicated to the Appellants on 25 August 2016, and their Statement of Appeal was filed on 15 September 2016, i.e. within the time limit required by both the EBL By-Laws and Article R49 of the CAS Code.

97. Consequently, the appeal filed by the Appellants is admissible.

VIII. APPLICABLE LAW

98. Article R58 of the CAS Code reads as follows:

“The Panel shall decide the dispute according to the applicable regulations and, subsidiarly, to 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 that the Panel deems appropriate. In the latter case, the Panel shall give reasons for its decision”.

99. Based on the aforementioned article, the Panel finds that it must resolve the present dispute in accordance with the EBL regulations (more specifically, the Disciplinary Code and the WBF Laws of Duplicate Contract Bridge adopted by the EBL) as it has been sustained by both parties.

IX. PROCEDURAL ISSUE – THE ADMISSIBILITY OF THE DOCUMENTS FILED BY THE

APPELLANTS FROM THE EBL FILE

100. The Appellants filed several documents after being provided with the EBL file. The Respondent objected to the admission of these documents on file because, in its view, (i) all of the documents were in the possession of the Appellants since the beginning of the CAS proceedings, (ii) nothing prevented the Appellants from producing them in the Appeal Brief and (iii) pursuant to Article R56 of the CAS Code there were no exceptional circumstances to admit such “new” documents.

101. On 23 March 2017, the Panel admitted the documents filed by the Appellants for the following reasons:

➢ The Panel firstly noted that the Appellants did not try to introduce “new” exhibits in terms of Article R56 of the CAS Code as alleged by the Respondent. Instead, from the very beginning of these CAS proceedings they tried to introduce the whole EBL file as evidence. In their Statement of Appeal, they requested the production of the file pursuant to Articles R44.3 and R57 of the CAS Code.

➢ Furthermore, the Panel was only constituted on 26 November 2016. Thus, the Panel could not decide on the Appellant’s evidentiary request before the filing of the relevant Appeal Brief (17 October 2016) and, consequently, once the Panel was constituted it inquired on 14 December 2017 with the Appellants whether they maintained their request for the production of the EBL file.

➢ Moreover, once the Appellants confirmed that they upheld their request, the Respondent stated that:

“There is nothing material that is not already part of the CAS file.

If the Players wish to produce any elements of the case file that have not already been produced the EBL will not object. However, a wholesale production of a voluminous case file of which the vast majority (including all material elements) has already been produced would not be helpful to the parties or the Panel”.

In the Panel’s view, the Respondent, thus, consented to the production of documents from the first instance file as long as they were not already in the CAS file.

➢ Given the parties’ positions, on 19 January 2017, the Panel ordered the EBL to produce the “full case file” and, in order to avoid irrelevant or repeated documents, it was also ordered that the file was delivered directly to the Appellants who, in turn, had 10 days to present any document that were not yet in the CAS file and that were considered relevant for the case.

➢ Taking into account the aforementioned, the Panel reviewed the documents ultimately filed by the Appellants and concluded that, indeed, none of them were yet in the CAS case file, i.e. complying with the CAS’ directions in connection with the EBL position.

➢ Therefore, the Panel is of the opinion that the Respondent’s objection to the admission of these documents was in contradiction to what the latter initially had agreed to and therefore rejected the Respondent’s objection.

X. MERITS

102. According to R57 of the CAS Code, “[t]he 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”.

103. In the Panel’s view, the main issues to be resolved in the present matter are the following questions:

A. Was the Disciplinary Committee duly constituted for conducting the disciplinary

proceedings against the Appellants; and (in case this question is answered in the negative) what are the consequences thereof?

B. Was there Inappropriate Communication between the Appellants; and (in case this

question is to be answered in the affirmative) what are the proper sanctions (if any) to be imposed on the Appellants?

A. The constitution of the Disciplinary Committee

104. The Appellants have requested that the Panel annul the Appealed Decision because, in its view and according to rule G.3 (a) of the current EBL By-laws, none of the members of the Executive Committee can be part of the Disciplinary Committee and, in casu, Mr. Jurica Caric (member of the Executive Committee) chaired the Disciplinary Committee. The Respondent disagrees with

the above and has submitted that the rule invoked by the Appellants is not applicable to the case at hand, since it was adopted on 24 June 2016 only, i.e after the constitution of the Disciplinary Committee. Thus – according to the Respondent – it is the “old” rule that applies to the case at hand, i.e. Article 47 of the (previous) EBL Statutes. According thereto the Disciplinary Committee has been constituted correctly.

105. The Panel agrees with the Respondent. The Disciplinary Committee was indeed constituted before the enforcement of the new EBL By-laws. The view held by Panel is backed by the Appellants’ submissions in the first instance proceedings (dated 5 April 2016 and 17 June 2016), which were specifically directed to each of the members of the Disciplinary Committee.

106. Furthermore, the Panel has noted that the current EBL By-laws do not provide any transitional regulations and that, therefore, in compliance with consistent CAS jurisprudence the procedural actions (such as the constitution of the deciding body) shall be governed by the principle tempus regit actum. Ad exemplum, in CAS 2004/A/635 it was stated the following:

“The Panel observes that, as a general rule, transitional or inter-temporal issues are governed by the principle “tempus regit actum”, holding that any deed should be regulated in accordance with the law in force at the time it occurred. As a consequence, procedural actions, such as the filing of an appeal, should be done in compliance with rules and time limits in force when they are performed” (emphasis added).

107. In light of the above, the Panel shall analyse (the former) Article 47 of the EBL Statutes which ruled on the constitution of the Disciplinary Committee. The provision reads as follows:

“The Disciplinary Commission consists of a president elected by, and from, the members of the Executive Committee, and two other persons who are not members of any of the EBL or WBF authorities”.

108. Given the wording of this article, the Panel concludes that the Disciplinary Committee was duly constituted for conducting the disciplinary proceedings against the Appellants since it was compulsory that an executive member, such as Mr. Caric, chair the disciplinary body.

109. The Panel thus finds that the objections by the Appellants with regard to the constitution of the Disciplinary Committee shall be rejected.

B. The existence of Inappropriate Communication

110. The EBL has accused the Appellants of infringing Article 3 of the EBL Disciplinary Code and Law 73 (B) of the laws of Duplicate Contract Bridge, which read as follows:

• Article 3 of the EBL Disciplinary Code:

“3. Reprehensible Conduct (…) serious infringement of the EBL Status or Regulations; (…) illicit actions or behaviour affecting the proper running of a competition or its results”.

• Law 73 (B) of Laws of Duplicate Contract Bridge:

“B. Inappropriate Communication between Partners 1. Partners shall not communicate by means such as the manner in which calls or plays are made, extraneous remarks or gestures, questions asked or not asked, or alerts and explanations given or not given. 2. The gravest possible offence is for a partnership to exchange information through prearranged methods of communication other than those sanctioned by these Laws”.

111. Considering the wording of Article 73(B) (2) of the Laws of Duplicate Bridge, the Panel firstly concludes that the illicit behaviour is composed of two elements: the first one, is that the players prearrange a (illicit) method of communication and, the second one, is that this method is effectively used for exchanging information. Therefore, in view of the majority of the Panel, the main component in order to establish the rule violation is the existence of the prearranged method of communication.

112. This reasoning is not only confirmed with the wording of the rule but also with the following elements that were submitted during these proceedings, which are explained here below;

➢ In point 49 of the Appealed Decision, the EBL stated that: “The [Disciplinary] Commission therefore confirms that the Players have breached the (…) regulations, by using the code, which is considered as a prearranged method of communication”.

➢ The Guidelines of the World Bridge Federation (considered for establishing the sanction on the Players) state the following:

“Collusive cheating – pre-arranged methods of arranging unauthorised information by a partnership. (…)”.

➢ In para. 54 of its Answer to the Appeal the Respondent stated that “[t]he Players are accused of using a code to pass valuable information” and “[i]f proven, (…) the code would constitute a breach of article 73 B of the Laws of Duplicate Bridge and a reprehensible conduct for the purposes of the EBL Disciplinary Code”.

The Panel considers that the Respondent itself accepts that the Code (and the valuable information) has to be proven for establishing the infraction of Law 73 B.

➢ In para. 147 of its Answer to the Appeal the EBL submitted “that the Panel must be comfortably satisfied that the Players have exchanged illicit information pursuant to a pre- arranged code”.

➢ The Panel notes that the Respondent is not trying to prove that the Players simply exchanged information (e.g. through a question or an alert) in a given hand or few number of hands but, instead, the EBL is trying to prove that the Players systematically

used a communication code that gave illicit information during the whole matches of the Competition.

In light of the above, and in order to establish whether the Players should indeed be sanctioned or not, the Panel has to be convinced of the existence of the “prearranged method of communication” (i.e. the alleged Code).

113. When assessing the aforementioned issue, the Panel shall take into account; (i) that the burden of proof relied on the EBL (as it has been accepted by the latter); and, (ii) that both parties have accepted that the applicable standard of proof is “comfortable satisfaction”. Regarding the – slightly – different interpretations of “comfortable satisfaction” raised by the parties, the Panel considers that both points of view can be consolidated with the description given by the panel of the case CAS 2014/A/3628, which reads as follows:

“The “comfortable satisfaction” standard of proof has been developed by the CAS jurisprudence (i.e. CAS it by comparison, declaring that it is greater than a mere balance of probability but less than proof beyond a reasonable doubt. In particular, the CAS jurisprudence has clearly established that to reach this comfortable satisfaction, the Panel should have in mind “the seriousness of allegation which is made” (i.e. CAS 2005/A/908, CAS 2009/1920). It follows from the above that this standard of proof is then a kind of sliding scale, based on the allegations at stake: the more serious the allegation and its consequences, the higher certainty (level of proof) the Panel would require to be “comfortable satisfied”.

In short, given that the allegations – and consequences – against the Players are indeed serious, the Panel concludes that in order to confirm the sanction on the Appellants, it has to be comfortably satisfied (and with a high level of certainty) that the Appellants exchanged information by using the Code.

114. According to the Respondent, “[t]he cheating allegation before the EBL DC was that the Players, when defending and on opening lead, placed the card vertically to indicate an unseen high honour (Ace, King or Queen) in that suit and otherwise placed the card horizontally”. Furthermore, the Respondent sustains that the Code is of great value because any additional (illicit) information makes a decisive difference and that “the alleged code is particularly useful for players using Slavinsky leads”. In other words, the Panel identifies that the alleged pre-arranged method of communication used by the Players has the following elements:

➢ The Players use it when they are defending and in (all) the opening leads; ➢ The Players placed the opening leads either in a vertical or horizontal position; ➢ The vertical leads indicate an unseen High Honour (i.e. Ace, King, Queen) in the suit; ➢ The horizontal leads indicate that there are no unseen High Honours in the suit; ➢ The Code is of great value from a bridge perspective, moreover when it is combined with the Slavinsky code.

115. Taking into account the elements of the Code, and given the evidence brought to these proceedings, the majority of the Panel finds that there are several inconsistencies in the alleged pre-arranged method of communication identified by the Respondent, which are explained here below:

(i) The positioning of the cards: It has been proven that the Players did not place the cards solely in a horizontal or vertical way but also in a diagonal way. This is even accepted by Hammond 1 as follows: “[i]t can be difficult to determine with some leads if they are horizontal or vertical. I chose between vertical or horizontal even though some leads were clearly diagonal leads – I chose the closest to the two options”.

In this regard, the Respondent itself has presented a chart explaining on which angle boundary the leads were placed (see supra para. 83) and, indeed, there could be at least 15 cards (placed between the 20 and 70 degree boundary) that could be considered diagonal.

The majority of the Panel considers that this is problematic for the functioning of the Code because the Player placing the diagonal leads would only mislead his partner about the information intended to give.

(ii) “False positives”: A second inconsistency in the EBL’s accusation is that there are several cards that are “false positives” meaning that when the leads are vertical the suit does not contain a high honour and when they are horizontal these do contain a high honour. The Respondent itself has stated that in 112 lead cards there were 8 “false positive” cards.

Again, the majority of the Panel considers that this is a problem for the proper functioning of the Code since the Player on lead would be giving the wrong information to his partner.

(iii) The value of the Code: In the view of the majority of the Panel, the Respondent brought the “value” (or the effectiveness) as an important element to prove the existence of the Code, giving the impression that it could be used and be effective in all (or in the majority of) the leads. In fact, the Respondent stated that the Code was used when defending and in (all) the opening leads and that it was extremely useful from a Bridge perspective giving significant advantage over other pairs. Furthermore, the EBL called Mr. Drijver to the hearing in order to testify about the value of the alleged Code and other Bridge related matters. Mr Drijver indeed stated that “it is the best Code ever and (…) that it was amazingly useful”.

However, on the counter position, the Appellants presented documentations which stated that in 40 hands the alleged Code would not make sense for the game (since the alleged given information was already available) and furthermore, in 9 hands the Players responded in opposition to the alleged given information by the Code. In this regard, Mr. Fernandez also expressed at the hearing that he reviewed more than 50 hands thoroughly and considered that in 21 hands the Code could not be applied and in 5

hands not even a middle-level player would play them in the case that the Players had the alleged information provided by the Code (he identified board 21 (against England), board 29 (against Denmark), board 23 (against Ireland), board 11 (against Israel), and board 31 (against England)). After being invited to respond Mr. Fernandez’s assertions, Mr. Drijver stated that he only reviewed between 8 and 10 hands (out of 114) and said that if the Players were playing against the Code he could “only guess there could be Bridge reasons”, that sometimes “out of the bidding or other circumstances there is no reason to give your partner information”, and that maybe this was in order to avoid detection.

Having heard the testimonies, the majority of the Panel finds that Mr Fernandez’s statements were more consistent and convincing than the explanations (and presumptions) of Mr. Drijver which, in turn, were found quite contradictory to his previous statements about the “amazing effectiveness” of the Code. In short, the majority of the Panel is not convinced that the Code hypothesised by the Respondent provides the information asserted by the EBL.

(iv) Combination with the Slavinsky Code: the EBL stated that the Players combined the Code with the Slavinsky code and that this combination would resolve four possibilities of the game at once. In support of this argument, Mr. Drijver stated at the hearing that “the Slavinsky leads on itself are not that useful but with the unauthorized information is unbeatable”. The Players contested at the hearing that, if the Code was true, reconciling the two codes would only cause problems since the Slavinsky code also considered the “Jack” as a high honour whereas the Code does not. This was proven with Mr. Fernandez’ oral statements which confirmed the aforementioned. The Respondent did not rebuke this assertion and Mr. Drijver, after being questioned how the Code could be reconciled with the Slavinsky code, only stated that he was not an expert in the Slavinsky system.

In light of the aforementioned, Mr. Drijver’s statements were found quite contradictory and inconsistent, and the majority of the Panel indeed agrees with the Appellants’ view that the combination of the two codes would – again – give contradictory information to the partner and more “false positives” of the Code could be found (i.e. horizontal leads that do contain a “Jack”).

116. Given the aforementioned inconsistencies, the majority of the Panel cannot be convinced with comfortable satisfaction of the existence of the Code (or at least of the one asserted by the Respondent) because, if true, (i) it gives ambiguous information through diagonal cards and “false positives”, (ii) it goes against the other code used by the Players, and (iii) the Players sometimes could not apply it or played against it, and taking into consideration the own words of the Respondent that “even a slim advantage can be a determining one”, the majority of the Panel cannot conclude a patent advantage with a Code that has so many contradictions.

117. Furthermore, the Panel has also noted that Prof. Lawler stated at the hearing (and in his report) that “it is possible that we do not know the entire Code, we are only considering part of the Code”. This was seconded by Prof. Buchen who stated also at the hearing that “we do not have the whole story because the original observations even went down to second hand leads and third hand leads and we are now been looking at opening leads”. The majority of the Panel follows from these statements that the (EBL) experts

themselves find the analysis of the Code to be incomplete and considers that it cannot take the risk of sanctioning the Players based on an alleged partial code (with inconsistencies) because there is a significant risk that the Code does not exist at all or that the subsequent hypothesis, if any, would be completely different than the one at stake.

118. How can the Panel be comfortably satisfied in condemning the Players if the Code is contradictory? It shall be recalled, that the Respondent took the responsibility (and the burden) of proving the “prearranged method of communication” which, in the eyes of the majority of the Panel, has not been done to the required standard of proof and, therefore, without knowing whether the Players used a code or not there is no adequate certainty of what the exchanged information is.

119. Regarding the statistical evidence brought to these proceedings, the Panel is of the opinion that the statistical results depend strongly on the data introduced for making such calculations as it was stated in several of the experts’ reports. In particular, Buchen 1 established the following:

“(…) While it is not possible with statistical analysis alone to conclude with absolute certainty the existence of illegal signaling, the data may provide an indication of such to varying degrees of confidence. The analysis depends wholly on the data provided and involves no bridge judgement. Thus the conclusions drawn from the analysis are completely unbiased with respect to bridge considerations. However it should be made clear that errors or bias in the data can affect the results of the analysis.

(…) this conclusion is independent of bridge considerations but depends strongly on the accuracy of the data input into the statistical analysis” (emphasis added).

120. Taking into account the aforementioned, the majority of the Panel does not feel comfortable in relying on the calculations brought by the EBL because they were based on a (partial) hypothesis which did not take into account the existence of inconsistencies considered by the majority of this Panel when making the calculations. In addition to this, given the position of the parties, the following is also noted from the statistical studies:

➢ Exclusion of diagonal cards: It was already concluded that there were some cards that did not fit into the vertical or horizontal position (i.e. diagonal cards). In this regard, the Respondent sustained (based on Hammond 3) that even considering a horizontal position in a range between 0 and 10 degrees and a vertical position in a range between 80 and 90 degrees, the probabilities of coincidences of the Code assuming randomness would still be effectively zero (see supra para. 83).

To this effect, the Panel has reviewed Hammond 3 and considers that it cannot rely on the statistical numbers contained therein because it excluded all the “ambiguous” (or diagonal) cards for making the calculations. For example, for the calculations considering horizontal (0 to 10 degrees) and vertical (80 to 90 degrees) Hammond 3 only considered 69 cards of 104 opening leads; in other words, it excluded 35 cards that did not fit into the vertical or the horizontal range. The majority of the Panel considers that this approach is incorrect, firstly because the Code explained by the Respondent applied in (all) the opening leads of the Competition (not only in some of them), and

secondly because, logically, the exclusion of cards (that do not fit the Code) only raise the level of coincidence. Therefore, for a fair and appropriate conclusion, the calculations should include the “non-fits” of the Code (i.e. the cards that are not positioned in a horizontal or vertical way) and, moreover, to reconcile the numbers with the Slavinsky code that could raise more “false positives”. ➢ The inconsistencies on the “fresh data”: There is a common understanding between the parties that the hypothesis could not be tested with the same information used to discover it. In this regard, the Panel notes that the Respondent took into account as “fresh data” the 7 matches of the Competition not studied by Ms. Mevius and the cards played in the Bermuda Bowl (i.e. 20 cards). As a first remark, the Panel has reviewed the videos of the Bermuda Bowl and, given the low quality of the videos, it is very difficult to state in first place which cards were placed in a horizontal or vertical way. As a second remark, it should be said that it is unknown how many of these cards could be considered diagonal (since there is no angle measurement of the cards), how many could be false positives (including the Slavinsky “Jacks”), or how many plays could be considered non-sense for the game. Therefore, the majority of the Panel considers that the calculations based on “fresh data” are not reliable either.

121. With regard to the other elements brought forward by the EBL to prove the existence of the cheating allegations, the majority of the Panel finds that they are not effective either. In particular, the sanction imposed on the Appellants by the American Contract Bridge League was not based on the Code (as stated by the EBL) but on a more complex code that combined “noises” made by the Players and “spacing bidding cards”. The same is considered of Piekarek/Smirnov’s statements when accusing the Players of being one of the “dirty pairs” in bridge, because this is only hearsay based on suspicions with respect to a different cheating code in a 2007 competition. This only confirms that the cheating hypotheses against the Appellants vary considerably.

122. Lastly, the EBL also raised that it would be in a position of Beweisnotstand since, in its view, it was proven that the cards were not randomly placed, the cheating allegations could not be proven by direct means and, therefore, the Players should provide an alternative explanation of the non-random placement of the cards. As it has been stated above, the majority of the Panel does not consider that the EBL proved that the cards were not randomly placed because such (apparent) conclusion was based on unreliable statistics. Therefore, this argument is simply dismissed.

123. Taking all of the above into consideration, the majority of the Panel concludes that the exchange of information through the Code has not been proven to its comfortable satisfaction and rules that the appeal filed by the Players shall be upheld. Such conclusion does not mean that the Players are innocent of any wrongdoing, it only means that the EBL did not manage to prove to the comfortable satisfaction of the majority of the Panel that the Players committed an infraction of the EBL Rules. The Panel realizes how difficult it is for the EBL to establish the existence of a “code” between players, given the multiple possibilities of potential signs (moves, gestures, sounds, etc…). However, sanctions cannot be imposed on the basis of incomplete evidence. The present decision does not mean that the system of control provided by the EBL

Rules is invalid or that it cannot be used again. This decision is based on the evidence provided to the Panel in this particular case. Future investigations by the EBL, based on more consistent and reliable data, may lead to a different outcome than in the present matter.

ON THESE GROUNDS

The Court of Arbitration for Sport rules that:

1. The appeal filed by Messrs. Fulvio Fantoni & Claudio Nunes against the decision rendered by the Disciplinary Committee of the European Bridge League (EBL) on 18 July 2016 is upheld.

2. The decision rendered by the Disciplinary Committee of the European Bridge League (EBL) is set aside.

(…)

5. All other motions or prayers for relief are dismissed.