A probe into a mysterious software glitch that affected some 7,000 horse race wagering machines around the country is focusing on how long the machines' producer knew about the problem without alerting the public.

In a leaked e-mail dated May 15, a top California horse racing official suggests New York-based gaming vendor Scientific Games may have kept quiet about the malfunction for months.

On Wednesday, however, investigators said they haven't confirmed that.

The flaw in the machines' "quick-pick" function, discovered by a Bay Meadows customer betting on the Kentucky Derby, has triggered scrutiny of the nation's $15 billion horse betting industry. The question now is whether it was an innocent computer programming mistake or something more serious.

The e-mail obtained by MediaNews this week shows some officials suspect the latter.

"We have contacted Sci Games and they apparently became aware of this problem in February, but failed to disclose it to its customers, or certainly to us in California," wrote Richard Shapiro, chairman of the California Horse Racing Board, in the message to other racing officials.

"This is a breach of the public trust," he added.

Shapiro downplayed those statements Wednesday, saying they aren't based on firsthand knowledge.

"That is not something I know for sure," he said. "I was told by somebody that this situation may have existed back in February and may have been brought to

Advertisement
(Scientific Games') attention."

Scientific Games spokesman Tom Hodgkins said it would be premature to comment on the timeline of events until the company completes its own internal inquiry.

Meanwhile, he said, Scientific Games has instructed all its tellers to disable the quick-pick function on its BetJet machines. For an unknown period of time, the machines had apparently been excluding the last horse in each race from their presumably random, computer-generated selections.

Hodgkins said the company can't tell how many bettors were affected or how much money was involved because it doesn't specifically track quick-pick wagers.

Richard "Bon" Smith, who is leading the California Horse Racing Board's investigation, said Scientific Games officials were already aware of the glitch when he first contacted them on May 7. Whether they had known about it for days or months remains unclear, he said, though he's hoping to get some answers soon.

It took an unusual circumstance for the error to come to light.

Officials say the unidentified Bay Meadows bettor placed 1,300 quick-pick superfecta bets on the Kentucky Derby. That's an uncommonly high number, especially since most bettors prefer to make their own selections rather than leaving them to chance.

A winning ticket — with all four of the top finishers picked correctly, in order — would have paid $29,368.90.

Going through his betting slips after the race, however, the bettor found that Kentucky Derby winner Big Brown wasn't included on a single one.

Bay Meadows stewards alerted the California Horse Racing Board of the anomaly May 7. Two days later, the board ordered a stop to all quick-pick wagering in the state.

The May 15 e-mail from Shapiro to Ed Martin, president of Associated Racing Commissioners International, was intended to spread the word to other racing organizations around the country.

Still, the issue wasn't made public until this week. On Tuesday, state Sen. Leland Yee reacted by calling on the state auditor to investigate. He wants to know the scope of the problem and why it took at least two weeks to become public.

E-mail Will Oremus at woremus@dailynewsgroup.com.