11-0800-15243-2

STATE OF MINNESOTA

OFFICE OF ADMINISTRATIVE HEARINGS

FOR THE GAMBLING CONTROL BOARD

In the Matter of the

Lawful Gambling License of

American Legion Post 303, Fridley

License Number 00001

FINDINGS OF FACT,

CONCLUSIONS OF LAW,

AND RECOMMENDATION

The above-entitled matter came on for hearing before Administrative Law Judge Barbara L. Neilson on June 4, 2003, at the Office of Administrative Hearings in Minneapolis, Minnesota. The hearing concluded on August 14, 2003. The OAH record remained open until October 27, 2003, for the receipt of the parties’ post-hearing submissions.

E. Joseph Newton, Assistant Attorney General, 445 Minnesota Street, Suite 1800, St. Paul, MN 55101-2134, appeared on behalf of the Minnesota Gambling Control Board. Jack E. Pierce, Attorney at Law, Pierce Law Firm, P.A., 4230 Central Avenue N.E., Minneapolis, Minnesota 55421, appeared on behalf of the American Legion Post 303, Fridley (“the Respondent” or “the Post”).

NOTICE

This report is a recommendation only and not a final decision. The final decision with respect to this matter will be made, after review of the record, by the Minnesota Gambling Control Board. The Board has the authority to adopt, reject, or modify any Findings of Fact, Conclusions, and Recommendations contained in this report. According to Minn. Stat. § 14.61, the Board’s final decision shall not be made until this Report has been made available to the parties to the proceeding for at least ten days and an opportunity has been given to each party adversely affected to file exceptions and present argument to the Board. Parties should contact the Minnesota Gambling Control Board, Suite 300 South, 1711 West County Road B, Roseville, MN 55113, telephone no. 651-639-4000, to determine the procedure for filing exceptions or presenting argument.

If the Board fails to issue a final decision within 90 days of the close of the record, this report will constitute the final agency decision under Minn. Stat. § 14.62, subd. 2a. The record closes upon the filing of exceptions to the report and the presentation of argument to the Board, or upon the expiration of the deadline for doing so. The Board must notify the parties and the Administrative Law Judge of the date on which the record closes.

STATEMENT OF ISSUE

Prior to the hearing in this matter, the Board’s motion for summary disposition on Count 1, paragraphs 18, 19, 22, and 24, Count II, and Count VII was granted. The Board was denied summary disposition on Count I, paragraphs 20, 21, and 23, and Count IV. As a result, Counts I, paragraphs 20, 21, and 23, Counts III, IV, V, and VI remained for consideration at the hearing. These remaining Counts were subsequently dropped by the Board during the hearing. The sole issue left for determination is relevant to the penalty to be imposed on the Post by the Board and involves the determination of the discrepancy between the amount of profit actually deposited by the Post into its gambling account relating to its tri-wheel paddlewheel game and the likely amount the Post would have deposited as profit in the absence of employee theft.

Based upon all of the proceedings herein, the Administrative Law Judge makes the following:

FINDINGS OF FACT

1. American Legion Post 303, located at 7365 Central Avenue Northeast in Fridley Minnesota, is an organization licensed by the Gambling Control Board to conduct lawful gambling and holds a premise permit. The Post’s gambling activities include a paddlewheel game referred to as a “tri-wheel.”[1]

2. Two of the Post’s employees have been criminally charged with theft of proceeds from the Post’s tri-wheel game. There is no evidence that any determination of how much money these employees stole has been made thus far in any other proceeding.

3. The tri-wheel is a mechanical, vertical wheel marked off into numbered sections which, when spun, uses a pointer to indicate the winning number. The table gives the player an option of betting on individual numbers, sections of numbers, or odd or even numbers. The wheel is broken into three concentric circles allowing for bets to be placed on any number corresponding to the numbers on the table. Players purchase chips from the cash bank cashier that in turn are used to buy tri-wheel tickets. The tickets come in a perforated sheet of five tickets that costs $5 per sheet or as a stub with only two tickets that costs $2 per stub. The tickets are used to place bets at the paddlewheel table. A sheet of five tickets may be folded together accordion-style and placed as a single $5 bet on one number, or the tickets may be separated and placed on separate numbers in multiple combinations of bets (i.e., one $3 bet and one $2 bet, one $4 bet and one $1 bet, or five $1 bets). Patrons mark their tickets on the back with a unique identifier and then place the tickets in slots on the betting table consisting of numbers and colors that correspond to the tri-wheel. The wheel has three rings on it (blue, orange, and yellow), and three numbers are selected with each spin. The paddlewheel operator keeps track of each spin of the wheel and records the spins with respect to the number selected by the spin and the given payout on that spin.[2]

4. Based upon probability theory and assuming all bets are placed on the tri-wheel’s blue numbers (the number with the least expected profit to the house), the tri-wheel game should yield a 12.5 percent profit or “hold.”[3]

5. During the four-year time period of January 1999 through December 2002, the Post earned an average of 5.28 percent in profits or “hold” on its tri-wheel game. The Post’s profits ranged from a low of –4.44 percent in April of 1999 to a high of 19.77 percent in June of 2002. The Post’s total gross receipts from tri-wheel activity during this four-year time period were $2,703,588; total prizes awarded were $2,560,811; and total net receipts from tri-wheel activity were $142,777.[4]

6. Based upon an expected hold of 12.5 percent, the Post’s profits on $2,703,588 in gross receipts during the time period of January 1999 through December 2002 would have been expected to be approximately $337,949. The amount of the deficit in the Post’s hold thus is approximately $195,172 (the expected hold of $337,949, minus the actual hold of $142,777). While this is the best estimate, it is not possible to be certain of the exact value of the deficit due to the fact that the tri-wheel is a random game.[5]

7. Based upon concerns expressed by the Board, an agent of the Minnesota Department of Public Safety conducted an investigation of the Post’s tri-wheel operation in January and February of 2002. Cameras were installed to aid in the investigation. The cameras made it possible to see the tri-wheel operations, the customers, the table activity, and the tri-wheel itself.[6]

8. At the request of Professor Sanford Weisberg, statistics professor at the University of Minnesota, Board Compliance Officer Gary Danger reviewed each tri-wheel ticket that had been played at the Post and reviewed documents showing each number that was selected on each spin of the tri-wheel at the Post between September 1 and December 9, 2001. Exhibit 4 was prepared based upon this review and shows the number of times each number on the blue wheel was selected during the 5,100 spins that occurred during that time period. Each number occurred on average about equally often. Professor Weisberg created Exhibit 5, a bar graph representation of the spin occurrences, based upon the information contained in Exhibit 4.[7]

9. Both parties’ experts agreed that the Post’s tri-wheel was fair (i.e. the mechanical operation of the wheel did not indicate a proclivity to favor one number over another).[8]

10. Because the wheel was fair, the Post’s tri-wheel game is ruled by the laws of probability, and the amount of profitability to the house that should be observed can be computed. However, because the tri-wheel is a random game, the amount of profitability (and thus, the amount of the deficit here between the Post’s expected profit and its actual profit) cannot be calculated with exactitude. In order to calculate the variability, it is necessary to know the size of the average bet.[9]

11. At the request of Professor Weisberg, Mr. Danger reviewed the Post’s records and videotapes of tri-wheel activity at the Post during the evening of January 18, 2002, in an attempt to determine the betting patterns of patrons. During the time frame of 5:00 p.m. to midnight on January 18, 2002, $4,713 was spent on bets at the Post. The total number of bets placed was 3,257. Patrons placed 2,523 $1 bets; 401 $2 bets; 138 $3 bets; 1 $4 bet; and 194 $5 bets during that time period. The average bet placed that day was $1.45. Variations in the size of the average bet affect the confidence interval calculations.[10]

12. Because the average bet size is uncertain, Professor Weisberg repeated the confidence limit calculations of several different average bet sizes between $1.00 and $5.00. The widest interval is obtained if it is assumed that the average bet size is about $3.50. In that instance, the 99 percent confidence interval for the deficit is between $129,708 and $260,635.[11]

13. The Board does not know the average bet that was placed on the Post’s tri-wheel during the time frame of July 1, 1997, through January 31, 2002, or the average bet placed during the time period of September 1, 2001, through December 9, 2001. [12]

14. Professor Weisberg demonstrated that, if no cheating or theft had occurred, the game was played according to the rules, only blue numbers were played, and the average bet was $1.45, the Post’s tri-wheel receipts for the period of January 1, 1999, through December 31, 2002, should have been in the range of $152,788 to $237,554. Professor Weisberg also calculated a “worse case” scenario assuming that the average bet was $3.50 rather than $1.45 and arrived at an interval of $129,708 to $260,635.[13]

15. If players did not play only the blue numbers on the tri-wheel, the interval estimate would be shifted upward. The estimates provided by Professor Weisberg were based on giving the greatest amount of benefit to the Post.[14]

16. Some other organizations with lawful gambling operations have also had paddlewheel profits occur at a lesser level than 12.5 percent. For example, the American Business Womens Chapter Lady Slipper had a 7.79 percent hold in fiscal 1998; a 7.5 percent hold in fiscal 1999; a 9.57 percent hold in fiscal 2000, and an 8.89 percent hold in fiscal 2001. That organization did, however, have a 13.62 percent hold in fiscal 2001. Although individuals at the American Business Womens organization were caught stealing, the Board has not commenced any proceeding against that organization. The Corcoran Jaycees had an 8.99 percent hold in fiscal 2000 and a 4.58 percent hold in fiscal 2001. The Board’s compliance officer had no knowledge that any action has been taken by the Board against the Corcoran Jaycees alleging insufficient internal controls.[15]

17. The Post’s hold in fiscal 1998 was 3.35 percent; its hold in fiscal 1999 was 2.43 percent; its hold in fiscal 2000 was 4.26 percent; its hold in fiscal 2001 was 6.13 percent; and its hold in fiscal 2002 was 6 percent. Except for fiscal 2001, when the Corcoran Jaycees had a lower hold than the Post, the Post consistently had the lowest hold of all the organizations reported on Ex. 10 between fiscal 1998 and fiscal 2002. That was significant to the Board in deciding to take the present action. Moreover, at the time that surveillance cameras were installed at the Post in January 2002, the Post had the lowest percentage of hold in the state.[16]

18. There is no evidence that any member of the American Legion gave consent to its employees to play the tri-wheel or purchase paddlewheel tickets. Although Mr. Danger is not specifically aware of the manner in which the theft of the Post’s tri-wheel proceeds was carried out, it appears that at times a losing ticket was marked as a winning ticket even though that number was not spun and assumes that employees pocketed the money. In addition, the theft may have been facilitated by the fact that employees counting money at the end of the evening had access to the money contained in the drop box money and one employee deposited money in a drawer rather than in the drop box. There is no evidence that the Post ever told its employees to engage in such Rule violations.[17]

19. The Post has complied with corrective action ordered by the Board in the past.[18]

20. On December 24, 2002, the Board served and filed a Notice of and Order for Pre-Hearing Conference alleging seven counts of gambling violations against Respondent.

21. The Board thereafter moved for partial summary disposition on Counts I, II, IV and VII based upon the Respondent’s admissions and the terms of a Consent Order entered into by the Board and the Post’s former gambling manager.

22. Prior to the hearing, the Administrative Law Judge granted summary disposition for the Board relating to Count I (paragraphs 18, 19, 22, and 24), as well as Count II and Count VII. Summary disposition was denied as to the other counts involved in the motion for summary disposition.[19] Paragraphs 20, 21, and 23 of Count I and the entirety of Counts III, IV, V, and VI proceeded to hearing.

23. During the course of the hearing, the remaining Counts were dropped by the Board. The evidence presented at the hearing focused solely upon the difference between the amount of profit the Post actually deposited in its lawful gambling account relating to the tri-wheel game and the amount that would have been deposited in the absence of employee theft.

Based upon the foregoing Findings of Fact, the Administrative Law Judge makes the following:

CONCLUSIONS

1. The Gambling Control Board issues lawful gambling licenses and permits. The Board has the authority to revoke or suspend licenses and permits, and censure licensees and assess civil penalties for a violation of gambling law and rule.[20]

2. The State of Minnesota Gambling Control Board and the Administrative Law Judge have authority to consider the alleged violation by the Respondent pursuant to Minn. Stat. §§ 14.50, 14.58, and 349.155, subd. 5, and Minnesota Rules part 5225.0880, subp.7.

3. The Respondent received timely and appropriate notice of the allegations against it and the time and place of the hearing.

4. The Board has complied with all relevant substantive and procedural requirements of statute and rule.

5. For the reasons discussed in the attached Memorandum which is incorporated by reference in the Conclusions, the Board has established by a preponderance of the evidence that the discrepancy between the amount that should have been deposited into the Post’s gambling account and the actual amount that was deposited is at least $129,700.

RECOMMENDATIONS

IT IS HEREBY RECOMMENDED:

That the Minnesota State Gambling Control Board find that the discrepancy between the amount that was actually deposited in to the lawful gambling account of American Legion Post 303 of Fridley and the amount that would have been deposited in the absence of employee theft is at least $129,700.

Dated: November 24,2003

_/s/ Barbara L. Neilson _________

BARBARA L. NEILSON

Administrative Law Judge

Reported: Tape Recorded (three tapes); No Transcript Prepared.

NOTICE

Pursuant to Minn. Stat. § 14.62, subd. 1, the agency is required to serve its final decision upon each party and the Administrative Law Judge by first class mail or as otherwise provided by law. If the Board fails to issue a final decision within 90 days of the close of the record under Minn. Stat. § 14.61, this report becomes a final decision. In order to comply with Minn. Stat. § 14.62, subd. 2a, the Board must then return the record to the Administrative Law Judge within 10 working days to allow the Judge to determine the discipline to be imposed.

MEMORANDUM

The Board has alleged that Post employees stole tri-wheel proceeds. Two employees have been criminally charged with the alleged theft. The Gambling Manager entered into a consent order in which he admitted that he failed to sufficiently supervise the conduct of lawful gambling at the Post. The Administrative Law Judge previously ruled in this case in connection with the Board’s motion for summary disposition that, based on the admissions of the Gambling Manager and the Post and the holding in In re the Lawful Gambling License of Henry Youth Hockey Association,[21] the Board was entitled to summary disposition concerning its allegation that the Post violated Minn. Stat. § 49.167, subd. 1, and Minn. R. 7861.0030, subp. 9, and 7861.0060, subp. 1(B), by failing to sufficiently supervise the conduct of lawful gambling at its premises. (The Administrative Law Judge also found that the Board was entitled to summary disposition regarding certain other counts). This contested case proceeded to hearing on the remaining counts. Midway through the hearing, the Board dismissed the remaining counts. The focus of the hearing thus was upon the discrepancy between the amount that should have been deposited into the Post’s gambling account and the actual amount that was deposited, which the Board believes is important to establish in order for the Board later to determine what, if any, penalty should be imposed on the Post.

Sanford Weisberg, statistics professor at the University of Minnesota, testified on behalf of the Board.[22] Professor Weisberg indicated that, based upon a total amount bet between January 1, 1999, and December 31, 2002, of $2,703,588, a total amount paid in prizes of $2,563,811, and an expected hold of 12.5 percent for the tri-wheel game, it would be expected that the Post would have earned profit in the amount of $337,949 on the game. The Post’s actual hold during the period in question was only $142,777. The difference between the actual hold and the expected hold is $195,172.[23] Although this is the best point estimate that Professor Weisberg could make, he admitted that he could not state the exact amount of the deficit in the Post’s gambling account with any degree of certainty. Because of this uncertainty, Professor Weisberg made assumptions favorable to the Post by assuming that all bets were placed on the numbers in the blue portion of the wheel and ignoring multiple bets by one person. Professor Weisberg testified that he was comfortable using an average bet size of $1.45 based upon one day’s worth of data, and, applying probability theory and standard error analysis, testified that he was 99 percent certain that the actual amount of the deficit fell within the interval of $152,788 to $237,554. However, Professor Weisberg acknowledged in his report that “the average bet size is uncertain” and he repeated the 99 percent confidence interval calculations assuming a variety of average bet sizes ($1.00, $2.00, $2.50, $3.00, $3.50, and $5.00). See Ex. 13.

Arkady Shemyakin, mathematics professor at the University of St. Thomas, testified on behalf of the Post. Although Professor Shemyakin found no errors in Professor Weisberg’s methodology and agreed that the wheel itself is fair, he disagreed with Professor Weisberg’s conclusions. Professor Shemyakin believes that the amount of $152,788 should be used as a fair point estimate for the amount of the deficit rather than the higher amount calculated by Professor Weisberg. Professor Shemyakin pointed out that there are two components of the deficit: one is a non-random component (such as a skewed wheel, systematic underreporting of the amount of bets, or overreporting of the amounts paid in prizes) and one is a random component (the house being “lucky” or “unlucky”). To sort out the random or chance component, it is necessary to come up with a confidence interval. He suggested using the $152,788 figure because it gives a more conservative estimate of the nonrandom part of the deficit.

Professor Shemyakin also disagreed with the assumptions made by Professor Weisberg relating to the average bet size and independence of individual bets. He testified that it is not statistically sound to use data from just one day within four years in determining the average size of bets, since the information provided in that one day might not be representative of the entire four-year period. In addition, Professor Shemyakin asserted that, while Professor Weisberg’s assumption of the independence of individual bets was a “healthy assumption,” it was arguable because there was not enough evidence to support either that there was a coordinated effort of people making bets or that bets were made in a totally independent manner. In addition, bets on a spin may not be exactly independent but may be influenced by the results of the previous spins. Professor Shemyakin asserts that it is reasonable to assume that some players’ bets are not independent and that they use similar strategies or demonstrate similar reaction to the previous spins. Instead of using the total number of bets and the average individual bet size in the calculation, Professor Shemyakin suggested basing the calculation on the total number of spins, the average bet per spin, and the assumptions of perfect linear dependence of individual bets and a constant size of bet per spin. According to Professor Shemyakin, the worst case scenario is completely coordinated betting where all individuals place their bets on the same number. Based upon the assumption of perfect dependence of all of the bets, Professor Shemyakin arrived at $43,126 as the lower bound of the deficit. He determined that the upper bound, based on the assumption of independence of individual bets, should be $129,708.[24] He believes that the real number is somewhere between these numbers, but has no idea where it is. However, he acknowledged that he does not have confidence that the actual amount that left the Post is between those two numbers because these are the lower confidence intervals.

After careful consideration of the competing approaches suggested by the parties’ experts, the Administrative Law Judge finds that Professor Weisberg’s analysis is based upon more reasonable assumptions, including that of independence of bets, than that of Professor Shemyakin. Dr. Weisberg’s analysis also appears to be fair because he gave the benefit of the doubt to the Post by looking only at the blue numbers (the use of the blue numbers is favorable to the Post because it yields a lesser profit to the house and arrives at a smaller deficit). The Judge believes that Professor Weisberg’s estimate of the point value of the deficit ($195,172) has a reasonable basis since it was derived by simply applying the 12.5 percent expected hold to the total amount bet, but does not recommend that the Board use this number as the exact amount of the deficit due to the random nature of the tri-wheel game and the fact that the actual hold for the house clearly may be lower than the expected hold. Therefore, as Professor Weisberg demonstrated, it is also important to take into consideration the variability in the amount of the deficit due to the size of the average bet.

When assessing the confidence intervals calculated by Professor Weisberg based upon various average bet sizes, the Administrative Law Judge is persuaded by Professor Shemyakin’s testimony that data from only one day of betting does not provide convincing evidence of the average bet size over a four-year period. As a result, the Judge recommends that, rather than assuming an average bet size of $1.45, the Board instead assume an average bet size of $3.50 and conclude that it is likely that the deficit is at least equal to the lowest number associated with the several intervals calculated by Professor Weisberg. The assumption of a $3.50 average bet yields the widest confidence interval and benefits the Post by reaching a lesser number for the lower bound of the interval. This seems appropriate given the uncertainty of the actual average bet size. Thus, it is recommended that the Board find that the amount of the deficit is at least $129,700, which corresponds to the lower bound of the confidence interval calculated by Professor Weisberg based on the assumption of an average bet size of $3.50. The use of the $129,700 figure is consistent with the testimony of both experts that the element of chance inherent in gambling makes it impossible to predict the hold with precision or pinpoint the exact amount of the deficit.[25] Under these circumstances, it is more appropriate to presume that the likely amount of the deficit is at least an amount that corresponds to the lowest bound of the widest interval calculated by Dr. Weisberg (that based upon an average bet size of $3.50). The Judge was not persuaded by the Post’s argument that the discrepancy in the account might be as low as $43,126, particularly since even Professor Shemyakin admitted that “one cannot seriously suggest that [the $43,126 figure] should be used” and acknowledged that “perfect linear dependence of individual bets “is a much less natural assumption than independence.”[26]

Thus, the Judge has recommended that the Board find that the amount of the deficit was at least $129,700. The Post asserted during the hearing that there was no proper basis for the Administrative Law Judge to infer that the rule violations found during the motion for summary disposition led to the stealing of that amount of money. The Post contended that the Board had not provided any evidence to make that leap. To the contrary, the Post asserted that it had no knowledge whatsoever of the theft, which was done in a manner not reasonably susceptible to detection. The Post also emphasized that it made the corrections required by the Board in every compliance review. These arguments should be considered by the Board in exercising its discretion to determine what, if any, penalty is appropriate.

B.L.N.

[2] Testimony of Danger; Ex. 6; see generally Minn. R. 7861.0100.

[4] Testimony of Danger; Corrected Ex. 1; Ex. 2; Danger’s Letter of Oct. 24, 2003.

[5] Testimony of Weisberg; Corrected Ex. 1; Ex. 13; Danger’s Letter of Oct. 24, 2003.

[8] Testimony of Weisberg and Shemyakin; Exs. 4-5.

[10] Testimony of Weisberg, Danger; Exs. 3, 13.

[11] Testimony of Weisberg; Ex. 13; Corrected Ex. 1; Danger’s letter of Oct. 24, 2003.

[13] Testimony of Weisberg; Ex. 13; Danger’s Letter of Oct. 24, 2003.

[19] Order on the Motion for Partial Summary Disposition dated May 23, 2003.

[20] Minn. Stat. § 349.151, subd. 4 & § 349.155, subd. 4 (2000)

[22] It should be noted that the numbers and time period reflected in Exhibit 9 to which Professor Weisberg testified on the first day of hearing vary from those reflected in Exhibit 13, which was prepared by Professor Weisberg during the interval between the first and second days of hearing and formed the basis for both Professors Shemyakin and Weisberg’s testimony during the second day of hearing. It was evident from the Board’s post-hearing brief that it is urging that the rationale set forth in Exhibit 13 be followed, as modified by Mr. Danger’s post-hearing letter.

[23] Prof. Weisberg asserted in Ex. 13 that the amount of the deficit was $198,172; however, Mr. Danger’s post-hearing letter identified errors in Ex. 1 and indicated that this number should be shifted downward slightly to $195,172.

[24] Professor Shemyakin actually testified to a figure of $142,708, but it seems evident that this number should be adjusted downward by $3,000 in accordance with Mr. Danger’s letter of Oct. 24, 2003, and the changes to Exs. 1 and 13 that are noted in that letter.

[25] In fact, Exhibit 10 shows that several gambling organizations have failed to achieve a 12.5 percent hold in recent years.