Maths & Betting: The Pari-Mutuel And The Numbers Racket

In the immediate post-Prohibition era, Mr. Arthur Flegenheimer, also known as Dutch Schultz, experienced a sudden downturn in profits from his beer-supplying activities.

In order to introduce some liquidity into his business empire he called on the mathematical expertise of a Mr. Otto Biederman, aka Otto Berman aka Dutch Otto aka Abba Dabba aka Avisack.

Mr. Biederman himself preferred the appellation Abba Dabba, which he particularly liked for its ‘spooky ring’.

Abba Dabba was something of a gambling man and a renowned handicapper of racehorses. He is reputed to have correctly assessed the winners of 28 races out of 29 in one particularly hot streak.

During the early thirties, Abba Dabba had dreamed up a scheme which Dutch now called upon to be implemented, this scheme being a numbers racket. Abba Dabba’s strategy was as follows:

In the numbers rackets of the time, winners were indicated by the 3rd digit in the pari-mutuel totals for the first, third and seventh races at a given racetrack. For example, if the pari-mutuel totals for races one, three and seven were $212.90, $435.68 and $682.50, the winning number would be 252.

Abba Dabba would go to the specific track whose totals were being used, wherever this might be, and call in the figures he obtained before the seventh race. A Schultz employee in New York would read back to him the numbers which had been most popular in that day’s sales for the numbers game in Harlem.

As the story goes, Abba Dabba then opened his form books and charts and in a few moments calculated and placed a number of bets on the seventh race which were designed to prevent the final digit of the day’s most popular number from coming up.

I am curious as to how Abba Dabba would even approach this problem, let alone successfully rig the numbers, and I would be grateful if someone could shed some light on the matter.

I have used the exact pari-mutuel figures given by Tom Clark in his book The World Of Damon Runyon. This is because I don’t completely understand them. They seem on the low side for race totals, but I am assuming they are just that, i.e. the total amount of money bet on the pari-mutuel on a given race at a given track.
Footnotes:

  1. Abba Dabba Berman earned the sum of $10,000 per week from Schultz between 1933 and 1935, during which time he doubled Dutch’s profits in the numbers game.

  2. Abba Dabba appears in seven Runyon stories as the horse player Regret. He was a personal friend of Runyon. He died along with Schultz and two other gang members in a bloodbath at the Palace Chop House in Newark, NJ in 1935.

  3. Both Berman and Schultz are major characters in Billy Bathgate (1991, Robert Benton) played by Steven Hill and Dustin Hoffman respectively.

This site, Kill the Dutchman!, has further information onf this subject. According to the author, the totals given are not the amounts wagered but the total of the parimutuel payouts. The first number is the total of from races one through three. The second uses one through five, while the third uses one through seven.

I’m assuming that all win, place, and show payouts were totalled to generate the policy numbers. If that’s the case, determining the payouts is not too hard. (The amazing part is the rapidity with which he did it). Upon determining the dangerous combinations he likely placed a series of “show” bets. The show pool is typically the smallest and least volatile, which makes it easy to manipulate precisely.

The article mentions one advantage he did have, though. Because Dutch owned the track, he was able to past-post his wagers.

Thanks FortMarcy, you are a true champion of the turf. 1970 wasn’t it?

I am somewhat clearer about the task facing Abba Dabba on his visits to the track, and more than somewhat in awe of his mathematical skills.

The rest of the piece is equally fascinating.