Case: 13-12733   Date Filed: 04/02/2014   Page: 1 of 18




                                                            [DO NOT PUBLISH]


            IN THE UNITED STATES COURT OF APPEALS

                    FOR THE ELEVENTH CIRCUIT
                      ________________________

                             No. 13-12733
                       ________________________

                D.C. Docket No. 3:10-cv-00191-WKW-WC


WALTER BUSSEY, et al.,

                                              Plaintiffs,

DOLLIE WILLIAMS,
DAVID M. PALMER, II,
ROBIN PAIGE,

                                              Plaintiffs - Appellees,

versus

MACON COUNTY GREYHOUND PARK, INC.,
d.b.a. Quincy’s 777,
d.b.a. Victoryland, et al.,

                                               Defendants,

MULTIMEDIA GAMES, INC.,
IGT,
BALLY GAMING, INC.,

                                              Defendants – Appellants.
               Case: 13-12733       Date Filed: 04/02/2014      Page: 2 of 18


                              ________________________

                      Appeal from the United States District Court
                          for the Middle District of Alabama
                            ________________________

                                       (April 2, 2014)


Before HULL and BLACK, Circuit Judges, and SMITH, * District Judge.

PER CURIAM:

       Plaintiffs-appellees Dollie Williams, David M. Palmer, II, and Robin Paige

(collectively, “appellees”) filed a putative class action against the following

defendants: (1) Macon County Greyhound Park, Inc. d/b/a “Victoryland” and

“Quincy’s 777,” a dog racing facility and gaming park in Macon County, Alabama

(“Victoryland”); (2) Milton E. McGregor, the President and Chief Operating

Officer of Victoryland; (3) Multimedia Games, Inc.; (4) IGT; and (5) Bally

Gaming, Inc. (“Bally”). The last three defendants (collectively, “the

Manufacturers”) are the manufacturers, owners, and operators of electronic bingo

machines that were in use at Victoryland during the time period relevant to this

case. Appellees, all of whom had played on the bingo machines, asserted that




       *
        Honorable C. Lynwood Smith, Jr., United States District Judge for the Northern District
of Alabama, sitting by designation.
                                            2
                 Case: 13-12733         Date Filed: 04/02/2014         Page: 3 of 18


operation of the machines constituted illegal gambling activity. 1 They sought

recovery of all money they lost while playing on the machines during the six

months preceding the filing of the lawsuit, pursuant to an Alabama statute

providing that:

              All contracts founded in whole or in part on a gambling
        consideration are void. Any person who has paid any money or
        delivered any thing of value lost upon any game or wager may recover
        such money, thing, or its value by an action commenced within six
        months from the time of such payment or delivery.

Ala. Code § 8-1-150(a) (1975).

        The district court granted appellees’ motion for class certification,

designating Dollie Williams, David M. Palmer, II, and Robin Paige as class

representatives, and certifying a class of: “All persons who, at any time during the

period beginning September 4, 2009 through and including February 1, 2010,

while using their Q-Club cards, lost money or value playing electronic ‘bingo’ at

Macon County Greyhound Park, commonly known as Victoryland.” All

defendants petitioned for permission to appeal pursuant to Federal Rule of Civil

Procedure 23(f), but this court granted the petition only as to the Manufacturers.



        1
          See Ala. Const., Art. IV, § 65 (“The legislature shall have no power to authorize
lotteries or gift enterprises for any purposes, and shall pass laws to prohibit the sale in this state
of lottery or gift enterprise tickets, or tickets in any scheme in the nature of a lottery; and all acts,
or parts of acts heretofore passed by the legislature of this state, authorizing a lottery or lotteries,
and all acts amendatory thereof, or supplemental thereto, are hereby avoided.”)


                                                    3
              Case: 13-12733    Date Filed: 04/02/2014    Page: 4 of 18


After review of the record and the briefs of the parties, and having the benefit of

oral argument, we reverse.

                                I. BACKGROUND

A.    Operation of the Bingo Machines and “Q-Cards”

      In order to play electronic “bingo” at Victoryland, a player must first insert

some thing of value — e.g., cash, a “house card” tied to an account number (but

not a specific person), or a ticket representing credits obtained from earlier play —

into a bingo terminal. A player could not play unless at least one other player

already was playing on a different machine, because players effectively played

against each other. A single game lasted only a few seconds, and one player won

each game.

      Sometimes, a player would also insert a “Q-Card” into the machine before

playing. A Q-Card was a loyalty card that allowed players to earn bonuses and

perks, while in turn providing Victoryland the ability to track demographic

information about players for marketing purposes. Each Q-Card was associated

with a specific individual and a specific player number. Even so, Q-Card use was

optional — not all players possessed Q-Cards, and not all players who had such

cards were required to use them each time they played electronic bingo. In fact,

not every bingo machine at Victoryland was equipped with a Q-Card reader.



                                          4
              Case: 13-12733     Date Filed: 04/02/2014   Page: 5 of 18


Moreover, a Q-Card did not represent value. A player could not use the card to

add value to the machine, or to extract any winnings.

      Approximately 270,000 Victoryland customers had Q-cards between

September of 2009 and March of 2010, but a majority of those persons did not use

the cards while playing bingo. In fact, two of the class representatives — Palmer

and Williams — admitted during deposition that they did not always use their Q-

Cards when playing bingo. Additionally, because use of the cards did not require

photo identification or a secret PIN number, a person other than the individual to

whom a Q-Card was registered could play using that card. Indeed, it was common

for Q-Card holders to loan their cards to other players, in order for the card holder

to more quickly accumulate points toward bonuses and other perks. A player also

might use another player’s card if the other player accidentally left the card in the

machine after completing a game. Thus, there was no way for Victoryland to

ensure that the person using a Q-Card at any given time was the same person to

whom the card was registered.

B.    Tracking Players’ Wins and Losses

      To track players’ wins and losses, Victoryland used a computer software

program called the Advantage Tracking System (“ATS”). The only way to tie

win/loss data to a particular individual was if that individual used his Q-Card while



                                           5
                 Case: 13-12733        Date Filed: 04/02/2014          Page: 6 of 18


playing. The ATS system generated win/loss reports based on each player’s

“session” of play. A session began when the player inserted his or her Q-Card into

the machine, and it ended when the player removed the Q-Card. Thus, a session

could consist of a single bingo game, or it could consist of many games. There is

no evidence that the ATS system was equipped to track wins and losses on a game-

by-game, as opposed to a session-by-session, basis.2


       2
         Plaintiffs rely heavily upon a statement made by Victoryland’s corporate representative
suggesting that it is possible that the ATS system could produce game-level data. The entire
exchange during the representative’s deposition is as follows:

               Q. Okay. So, [the report from the ATS system] would have that level of
       detail [about the player’s wins and losses] at least on a session basis?

                 A. I’m — I’m not following your question.

              Q. Okay. The information you just described would exist at least on a
       session basis?

              A. I don’t remember if it’s on a session basis or on an itemized play or
       entry basis. I just don’t recall.

                 Q. Okay. What do you mean by “itemized play basis”?

             A. That every time he purchased cards, it shows that or if it — or if it
       accumulates it. I just don’t remember.

                 Q. So, in other words, kind of a game-by-game basis?

                 A. Uh-huh, yes.

                 Q. Okay. So it’s possible that it may even have game-by-game level
       detail?

                 A. Yes. I just — I don’t know that. I don’t recall.



                                                  6
                Case: 13-12733         Date Filed: 04/02/2014       Page: 7 of 18


       For each session played by each individual, the ATS system generated a

report identifying the machine on which the session was played, the beginning and

ending time of the session, the total number of seconds the session lasted, and the

total number of games played during the session (but not the length of each game,

or the amount won or lost during each game). The report also contained financial

data, including a column each for Purchases, Prizes, Jackpots, and Wins. The

Purchases column indicated the total amount of purchases made on the machine

during that session. The Prizes column showed the total amount of winnings the

player received during the session. The Jackpot column showed any wins

exceeding the amount of $1,200. The “Win” column indicated the amount of the

win for the house and the loss for the player — i.e., the sum of all Purchases for the

session minus all Prizes for that session.




               Q. How would you go about finding that out?

               A. Look at the table.

Tr. doc. 218-1, at 80-81 (alterations and emphasis supplied). Later in the deposition, counsel for
the appellees asked the following question: “Prior to the break, Mr. Russell, we were discussing
the amount or type of data that may still be available on Macon County’s computer. And you
mentioned it should have the session level data at least, if not possibly game level data, correct?”
The Victoryland representative responded, “Yes, ma’am.” 

This testimony cannot be
fairly construed to mean that there is a reasonable possibility that the ATS system produces
game-level data. Rather, the testimony reflects that the representative simply did not recall
whether the data was produced on a game-level or session-level basis. He did, however, state
that he could find that answer by looking at the reports generated by the ATS system (“tables”),
which clearly include only session-level data. Additionally, the corporate representative for


                                                 7
               Case: 13-12733      Date Filed: 04/02/2014      Page: 8 of 18


       The ATS system also generated a report spanning the entire class period that

showed: (1) the name and address of each person who used a Q-Card while

playing bingo on one of defendants’ machines during the class period; (2) the total

amount wagered by each Q-Card user during the class period (while using their Q-

Cards); (3) the total amount won by each Q-Card user during the class period

(while using their Q-Cards); (4) the net loss to each Q-Card user (i.e., the net gain

to the house) during the class period (while using their Q-Cards); and (5) the

manufacturer on whose machine the wins or losses were incurred. There is also a

spreadsheet that shows, for each Q-Card user, the total amount lost (i.e., the total

amount won by the house) on each defendant’s machines (i.e., how much each user

lost on Bally machines, IGT machines, and Multimedia machines).

C.     Relationship Between Victoryland and the Other Defendants

       The Manufacturers leased their electronic bingo machines to Victoryland. A

corporate representative of IGT described the agreement between IGT and

Victoryland as follows:

              The agreement between IGT and VictoryLand provided that
       VictoryLand would pay IGT a percentage of the net win of each
       electronic bingo terminal calculated periodically.      If the rent
       calculated in this manner fell below a threshold amount, IGT would
       receive a guaranteed minimum rent. The rent calculation was made


Multimedia testified unequivocally that the company was “not able to determine on a game by
game level what the outcome of that game was.” Tr. doc. 218-4, at 20.


                                              8
                 Case: 13-12733     Date Filed: 04/02/2014   Page: 9 of 18


      by subtracting the amount paid out to all winners on a given terminal
      from the amount wagered by all players on that terminal and if this
      amount was positive, it would be multiplied by the agreed percentage.
      If the amount was negative or below a defined dollar amount,
      VictoryLand would pay a minimum rent. 3

Victoryland’s agreements with the other bingo machine manufacturers were

similar. In any event, the amount of the Manufacturers’ payments to Victoryland

depended upon the aggregated results of all activity on each machine over a given

time period, usually weekly. The payments did not depend upon any particular

player’s wins or losses, or on whether the players were using a Q-Card. The rent

payments were not connected in any way to data collected about Q-Card use.

                                II. STANDARD OF REVIEW

      A district court’s grant of class certification is reviewed for an abuse of

discretion. Vega v. T-Mobile USA, Inc., 

, 1264 (11th Cir. 2009)

(citing Klay v. Humana, Inc., 

, 1251 (11th Cir. 2004), abrogated in

part on other grounds by Bridge v. Phoenix Bond & Indem. Co., 

(2008)). “As long as the district court’s reasoning stays within the parameters of

Rule 23’s requirements for certification of a class, the district court decision will

not be disturbed.” Hines v. Widnall, 

, 1255 (11th Cir. 2003) (citing

Shroder v. Suburban Coastal Corp., 

, 1374 (11th Cir. 1984)).


      3
          Tr. doc. 224-1 ¶ 3.


                                             9
             Case: 13-12733      Date Filed: 04/02/2014    Page: 10 of 18


             A district court abuses its discretion if it applies an incorrect
      legal standard, follows improper procedures in making the
      determination, or makes findings of fact that are clearly erroneous. A
      district court may also abuse its discretion by applying the law in an
      unreasonable or incorrect manner. Finally, an abuse of discretion
      occurs if the district court imposes some harm, disadvantage, or
      restriction upon someone that is unnecessarily broad or does not result
      in any offsetting gain to anyone else or society at large. In making
      these assessments, we review the district court’s factual
      determinations for clear error, and its purely legal determinations de
      novo.

(quoting 

).

                                 III. DISCUSSION

      A plaintiff seeking class certification bears the burden of satisfying all

implicit and explicit requirements of Federal Rule of Civil Procedure 23. 

see also Valley Drug Co. v. Geneva Pharmaceuticals, Inc., 

,

1187-88 (11th Cir. 2003) (holding that the burden of proof to establish the

propriety of class certification rests with the advocate of the class, and the district

court has an independent obligation to evaluate whether the requirements of Rule

23 have been met).

      One threshold requirement is not mentioned in Rule 23, but is implicit in the

analysis: that is, the plaintiff must demonstrate that the proposed class is

“‘adequately defined and clearly ascertainable.’” Little v. T-Mobile USA, Inc., 

, 1304 (11th Cir. 2012) (quoting DeBremaecker v. Short, 

,



                                           10
                 Case: 13-12733         Date Filed: 04/02/2014            Page: 11 of 18


734 (5th Cir. 1970)).4 The trial court’s discussion of the “adequately defined and

clearly ascertainable” requirement was clear, concise, and supported by applicable

law, and it will be reiterated here:

       “An identifiable class exists if its members can be ascertained by
       reference to objective criteria.” Fogarazzo v. Lehman Bros., Inc., 

, 97 (S.D. N.Y. 2009). The analysis of the objective criteria
       also should be administratively feasible. “Administrative feasibility”
       means “that identifying class members is a manageable process that
       does not require much, if any, individual inquiry.” Newberg on Class
       Actions § 3.3 p. 164 (5th ed. 2012). Where a plaintiff satisfies this
       threshold issue, the district court then “conducts a rigorous analysis of
       the [R]ule 23 prerequisites.” Vega v. T-Mobile USA, Inc., 

, 1266 (11th Cir. 2009) (citation and internal quotation marks
       omitted). 5

       Additionally, the explicit language of Rule 23 requires proof that:

                 (1)     the [putative] class is so numerous that joinder of all
                         class members is impracticable;

                 (2)     there are questions of law or fact common to the class;

                 (3)     the claims or defenses of the representative parties are
                         typical of the claims or defenses of the class; and

                 (4)     the representative parties will fairly and adequately
                         protect the interests of the class.



       4
         The Eleventh Circuit adopted as binding precedent all decisions of the former Fifth
Circuit handed down prior to the close of business on September 30, 1981 in Bonner v. City of
Prichard, 

, 1209 (11th Cir. 1981) (en banc).
       5
           Tr. doc. 241, at 5 (alteration in district court’s opinion).



                                                    11
                 Case: 13-12733   Date Filed: 04/02/2014   Page: 12 of 18


Fed. R. Civ. P. 23(a) (alteration supplied). These requirements are commonly

referred to as numerosity, commonality, typicality, and adequacy.

      In addition to satisfying all of the requirements of Rule 23(a), a party

seeking class certification must show that the action is maintainable under at least

one of the three sub-sections of Rule 23(b). Here, appellees sought certification

under Rule 23(b)(3). Therefore, they were required to show “that the questions of

law or fact common to class members predominate over any questions affecting

only individual members, and that a class action is superior to other available

methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P.

23(b)(3).

      We find that the district court abused its discretion in concluding that the

proposed class was adequately defined and clearly ascertainable, and that issues

common to all class members would predominate over issues affecting only

individual members.

A.    Ascertainability

      The district court found that the objective criteria for identifying members of

the defined class could “be analyzed in a manageable and administratively feasible

way, namely, by reference to Victoryland’s records.” 6 We agree that the reports


      6
          

                                           12
              Case: 13-12733     Date Filed: 04/02/2014     Page: 13 of 18


generated by defendants’ ATS system provide sufficient data to identify all

individuals who, while using their Q-Club cards, suffered net losses at the session

level. The problem is that the district court’s class definition is not restricted to

individuals who suffered session-level losses. As currently defined, the class also

could encompass individuals who suffered losses at the game level, but not at the

session level. Appellees have not provided any indication that they have, or even

that they can obtain, data about losses at the game level. Accordingly, the class

definition must be revised to include only those players who, while using their Q-

Club cards, suffered net losses at the session level.

B.    Predominance

             A class may be certified only if “there are questions of law or
      fact common to the class.” Fed. R. Civ. P. 23(a)(2). “Under the Rule
      23(a)(2) commonality requirement, a class action must involve issues
      that are susceptible to class-wide proof.” Murray v. Auslander, 

, 811 (11th Cir. 2001). Commonality requires “that there be
      at least one issue whose resolution will affect all or a significant
      number of the putative class members.” Stewart v. Winter, 

, 335 (5th Cir. 1982) (footnote omitted).

Williams v. Mohawk Industries, Inc., 

, 1355 (11th Cir. 2009). The

Supreme Court has emphasized that, “for purposes of Rule 23(a)(2) “‘[e]ven a

single [common] question’” will do . . . .” Wal-Mart Stores, Inc. v. Dukes, – U.S. –

, 

, 2556 (2011) (alterations in original).




                                           13
            Case: 13-12733     Date Filed: 04/02/2014    Page: 14 of 18


             The same analytical principles govern Rule 23(b). If anything,
      Rule 23(b)(3)’s predominance criterion is even more demanding than
      Rule 23(a). Amchem Products, Inc. v. Windsor, 

, 623-
      624, 

, 

(1997). Rule 23(b)(3), as an
      “‘adventuresome innovation,’” is designed for situations “‘in which
      “class-action treatment is not as clearly called for.”’” 

, at – , 131 S. Ct., at 2558 (quoting 

-
      615, 

). That explains Congress’s addition of
      procedural safeguards for (b)(3) class members beyond those provided
      for (b)(1) or (b)(2) class members (e.g., an opportunity to opt out), and
      the court’s duty to take a “‘close look’” at whether common questions
      predominate over individual ones. 

.

Comcast Corp. v. Behrend, – U.S. – , 

, 1432 (2013). Additionally,

      [t]o obtain Rule 23(b)(3) class certification “the issues in the class
      action that are subject to generalized proof and thus applicable to the
      class as a whole, must predominate over those issues that are subject
      only to individualized proof.” Kerr v. City of West Palm Beach, 

, 1557-58 (11th Cir. 1989) (quotations and citation omitted).
      “Common issues of fact and law predominate if they ‘ha[ve] a direct
      impact on every class member’s effort to establish liability and on
      every class member’s entitlement to injunctive and monetary relief.’”
      

(quoting Ingram v. Coca-Cola Co., 

, 699 (N.D. Ga. 2001)). On the other hand, common issues will
      not predominate over individual questions if, “as a practical matter,
      the resolution of [an] overarching common issue breaks down into an
      unmanageable variety of individual legal and factual issues.”
      Andrews v. Am. Tel. & Tel. Co., 

, 1023 (11th Cir. 1996).
      Certification is inappropriate if the “plaintiffs must still introduce a
      great deal of individualized proof or argue a number of individualized
      legal points to establish most or all of the elements of their individual
      claims.” 

(citing Perez v. Metabolife Int’l, Inc.,
      

, 273 (S.D. Fla. 2003)). The predominance inquiry
      requires an examination of “‘the claims, defenses, relevant facts, and
      applicable substantive law,’ . . . to assess the degree to which
      resolution of the classwide issues will further each individual class



                                         14
                Case: 13-12733         Date Filed: 04/02/2014   Page: 15 of 18


      member’s claim against the defendant.” 

(quoting Castano
      v. Am. Tobacco Co., 

, 744 (5th Cir. 1996)).

Babineau v. Federal Express Corp., 

, 1191 (11th Cir. 2009) (first

alteration supplied, other alterations in original).

      The Manufacturers challenged the district court’s finding of predominance

only insofar as it concerned damages. The district court found:

             Plaintiffs also must prove that they lost money and the amount
      of those losses. Admittedly, the damages calculations will involve
      some individualized inquiries, as Defendants devote much ink to
      pointing out, but not to the extent that those inquiries predominate
      over the common questions with respect to liability. Generally, “the
      presence of individualized damages issues does not prevent a finding
      that the common issues in the case predominate.” Allapattah 

(collecting cases); see also Sacred Heart Health
      Sys., Inc. v. Humana Military Healthcare Servs., 

, 1178
      (11th Cir. 2010) (“The presence of individualized damages issues
      does not prevent” certification of a Rule 23(b)(3) class.). Here, the Q-
      Club card data stored in the IGT Advantage System will reduce the
      number of individual factual inquiries to a significant degree . . . .
      Although Defendants highlight the shortcomings of the data, those
      shortcomings present issues for sifting at the merits stage, not the
      class certification stage. 7

      That finding was an abuse of discretion, in light of the Supreme Court’s

decision in Comcast Corp. v. Behrend, – U.S. – , 

(2013), which

was handed down on March 27, 2013 — only two days before the district court’s

opinion, and long after the parties had briefed the class certification issue to the


      7
          

(internal citation omitted).


                                                   15
             Case: 13-12733     Date Filed: 04/02/2014     Page: 16 of 18


district court. In Comcast, the Supreme Court reiterated that class certification is

an evidentiary question, not just an analysis of the pleadings.

             Repeatedly, we have emphasized that it “‘may be necessary for
      the court to probe behind the pleadings before coming to rest on the
      certification question,’ and that certification is proper only if ‘the trial
      court is satisfied, after a rigorous analysis, that the prerequisites of
      Rule 23(a) have been satisfied.’” [Wal-Mart 

] (quoting General Telephone Co. of Southwest v. Falcon, 

, 160-161, 

, 

(1982)). Such
      an analysis will frequently entail “overlap with the merits of the
      plaintiff’s underlying claim.” 564 U.S., at 

. That
      is so because the “‘class determination generally involves
      considerations that are enmeshed in the factual and legal issues
      comprising the plaintiff’s cause of action.’” 

at 160, 

).

(alteration and emphasis supplied).

      The district court acknowledged the Comcast decision, but it nonetheless

failed to conduct the “rigorous analysis” required by that decision — instead

deferring resolution of important questions bearing on the class certification

analysis to the merits stage of the case. That failure cannot be overlooked, because

the “shortcomings in the data” are significant and bear directly on the issue of

predominance. As discussed above, appellees did not identify, either in their briefs

or at oral argument, any method for quantifying their losses at the game level, as

opposed to the session level, even though recovery of game-level losses is what

they have requested in this case. They also have not identified any definite method



                                          16
               Case: 13-12733    Date Filed: 04/02/2014    Page: 17 of 18


for allocating their damages among the Manufacturers and Victoryland. In other

words, for each dollar lost by a class member on a machine owned by Bally, for

example, appellees have offered no method for determining how many cents were

pocketed by Victoryland, and how many were pocketed by Bally. The money was

collected in the first instance by Victoryland, and then distributed among the

Manufacturers according to their respective leases with Victoryland. The amount

due under the lease was calculated based on the net player loss, by all players, on

each machine, over a certain period of time, usually weekly. It remains unclear,

therefore, how the district court would ever be able to allocate liability and

damages among the various defendants. See 

(concluding that plaintiffs’

expert’s model for damages “falls far short of establishing that damages are

capable of measurement on a classwide basis,” and finding that “[q]uestions of

individual damage calculations will inevitably overwhelm questions common to

the class”).

                                 IV. CONCLUSION

      For the foregoing reasons, we conclude that the district court abused its

discretion in certifying the plaintiff class. The district court’s order certifying a

class is therefore reversed, and the case is remanded for the district court to modify

the class definition to include: “All persons who, at any time during the period



                                           17
              Case: 13-12733       Date Filed: 04/02/2014     Page: 18 of 18


beginning September 4, 2009 through and including February 1, 2010, while using

their Q-Club cards, suffered a net loss of money or value playing electronic

‘bingo,’ as shown by session-level data obtained from defendants’ Advantage

Tracking System (ATS) computers at the Macon County Greyhound Park,

commonly known as Victoryland.” The district court is further directed, on

remand, to conduct the “rigorous analysis” required by the Supreme Court’s

Comcast decision regarding whether calculation of the class members’ damages

would necessitate such individual inquiry that individual issues would predominate

over common ones.8

       REVERSED AND REMANDED.




       8
         To the extent necessary, the district court may wish to allow some discovery on the
damages issue. In any event, supplemental briefing on the damages issue appears required
because plaintiffs must show a model establishing that “damages are capable of measurement on
a classwide basis.” Comcast, 569 U.S. at 

.


                                             18