CCR Special 11 — The Mattel Electronics Horse Race Analyzer

In the studio introduction for a March 1987 Computer Chronicles episode on computers and gambling, Stewart Cheifet showed Gary Kildall a hand-held, calculator-like device that claimed to help people pick winning race horses. Although Cheifet never identified the device by name, it was the Mattel Horse Race Analyzer, an odd footnote in the history of Mattel Electronics, which itself was a short-lived subsidiary of the famed Los Angeles-based toy company.

Stewart Cheifet holding the Mattel Electronics Horse Race Analyzer, a long rectangular-shaped calculator with an LCD screen on the left side, on the set of “Computer Chronicles” in 1987.

Indeed, by the time this episode aired, Mattel Electronics was long gone. The Horse Race Analyzer continued, however, under the ownership of one of the two men credited with creating the device. And as far as I can tell, it was sold, without the Mattel branding, up thru the mid-1990s.

Founders’ Scandal Helped Push Mattel Into Electronics Business

Mattel’s history began in 1945, when spouses Isadore and Ruth Handler and their friend Harold Matson formed a small business to build and sell picture frames. Matson was known as “Matt” and Isadore Handler went by his middle name of Elliott, so they named their company “Mattel” as a portmanteau.

Today, Mattel is best known for selling toys, notably its Barbie line of dolls, which debuted in 1959. After Barbie made Mattel the number-one toy company in the world, the Handlers took the company public in 1963. (Matson was long out of the picture by this point.) Then, in one of the great recurring storylines of 20th century business, Mattel proceeded to conglomerate, aggressively growing itself through acquisitions in unrelated industries before the entire scheme collapsed under its own weight.

In Mattel’s case, that collapse occurred between 1972 and 1975. As the company’s losses from their acquisitions started to mount, the U.S. Securities and Exchange Commission accused the Handlers and other senior Mattel executives of falsifying business records and financial statements. The Handlers were forced out of the company in 1975. Arthur Spear, who had been named president and CEO in 1973, was left to rebuild Mattel.

It was during this period that Spear had Mattel enter the consumer electronics business. Alex Smith, in his book They Create Worlds, explained that Ray Wagner, who reported to Spear and ran Mattel’s flagship toy division, decided to hire “new managers out of the traditional packaged good business who could jump-start product development and marketing.” One of these executives, Michael Katz, took over new product category marketing.

As Smith noted, games had traditionally been a “weak spot for Mattel,” and since pocket calculators “had become ubiquitous” during the mid-1970s, Katz thought it made sense for Mattel to develop a line of calculator-sized games utilizing LED technology. This led to Mattel releasing a number of games, mostly based on popular sports like American football and basketball, starting in late 1977 under the newly created brand name of “Mattel Electronics.”

The handheld games took off in 1978, with football and basketball selling over one million units each. Bolstered by this success, Mattel Electronics started developing a programmable home video game console to compete against the Atari VCS (a/k/a the Atari 2600.) In late 1978, Mattel Electronics became a full-fledged division within Mattel Toys headed by Jeffrey Rochlis, a former marketing executive with Nabisco.

Ex-Cosmetics Executive Sold Toymaker on Horse Racing Computer

During this time another former consumer products marketing executive, Evan W. “Bill” Mandel, started consulting for Mattel. Mandel spent most of his career at Revlon, the cosmetics company founded in 1932 by brothers Charles and Joseph Revson and Charles Lachman. (Revlon was Revson with an “L” substituted for the “S” to acknowledge Lachman.) Charles Revson effectively ran the company as a dictator until his death in 1975. During Revson’s tenure, Revlon was notorious for its high rate of management turnover, as the CEO routinely bullied, humiliated, and fired subordinates who challenged his authority.

This included Mandel, who at one point was considered Revson’s heir apparent. Mandel started with the company in 1955 as an advertising manager. By the late 1960s he was the company’s executive vice president in charge of marketing. But working for Revson took an enormous toll. Andrew Tobias, who wrote the definitive book on Revson’s tenure at Revlon after his death, said that Mandel “would start shaking on Sunday afternoons in preparation for Monday-morning management meetings, and actually passed out at one such meeting, it became so heated.”

Revson eventually shuffled Mandel off to a make-work position as president of Revlon Development Corporation. Once Mandel’s stock options at the parent company vested, he left Revlon for good in January 1972. At the age of 45 he was a multi-millionaire and essentially retired from business, but he continued to work as a consultant based in New York.

I don’t know why Mattel hired Mandel specifically as a consultant. As Alex Smith and other Mattel historians have pointed out, Art Spear brought in a number of executives with experience in consumer products while rebuilding Mattel’s finances following the Handlers’ departure. And based on contemporary accounts, Mandel was widely considered one of the country’s top marketing minds. Indeed, Mattel and Revlon were among the first companies to embrace television sponsorship in the 1950s as a new form of direct-to-consumer marketing. So no doubt Mandel was long familiar with Mattel and vice-versa.

In any case, Mandel was the person who convinced Mattel Electronics–and Jeffrey Rochlis in particular–to expand its growing handheld line into a device targeting people who liked to bet on horse races. Maurice Zolotow, a longtime Hollywood biographer, wrote the only detailed account I could find of Mandel’s work with Mattel in his 1983 book, Confessions of a Racetrack Fiend. Zolotow described Mandel as a “passionate horseplayer” who often took an hour to analyze a race before placing a bet. When the first wave of microcomputers came out in the late 1970s, Mandel purchased an Apple II and started toying with ways to have the computer help with his analysis.

At some point in late 1977 or early 1978, Mandel pitched the idea of a dedicated race handicapping computer to Mattel Electronics. Rochlis approved the project. To create the actual handicapping program, Mandel worked with another Bill: Dr. William L. Quirin, an associate professor of mathematics and computer science at Adelphi University in New York City. Quirin had made a name for himself as an expert in using computers to analyze race handicapping. His first of several books on the subject, Winning at the Races: Computer Discoveries in Thoroughbred Handicapping came out in 1979.

Mandel was already familiar with Quirin’s work and the two men developed a handicapping algorithm that would fit on a single microprocessor–specifically, a TMS1100 microprocessor from Texas Instruments. The 4-bit CPU came with 2 KB of on-board ROM and just 64 bytes of RAM. It was a popular choice in the late 1970s for handheld calculators and various home electronic games, such as Milton Bradley’s Simon.

Mandel and Quirin shared authorship on the eventual patent filed for what they described as an “electronic horse race analyzer.” A separate patent was filed by Mandel and Mattel designer Ted Mayer, who created the case design for the device. (Mayer is probably best known for his work on the original “Masters of the Universe” toy line for Mattel.) The finished product was dubbed the Mattel Horse Race Analyzer.

Who Knew That a Horse Race Could Be So Complicated?

So how did the Mattel Horse Race Analyzer work? The basic idea was for the user to enter information about an upcoming race using data from the Daily Racing Form (DRF), a popular tabloid newspaper that covered the sport. The user would start by entering some basic information about the race, such as the amount of the purse (prize money), distance, and the starting position of each horse. For each horse entered, the user would then need to input a variety of statistics, including the days since their last race, the number of races since there last layoff of at least 30 days, their two-year performance record, and so on.

As you can imagine, this process could be quite tedious. A number of reviews pointed out it could take upwards of 20 minutes to enter all of the necessary information. This made the Analyzer somewhat impractical to use at the track, particularly since there was typically only about 20 minutes between races.

Once all the data was entered, the Analyzer then used Bill Quirin’s algorithm to produce “ratings” for the top four horses. These ratings purported to measure the projected speed of each horse relative to the others entered in the race. The horses were displayed one-by-one starting with the horse predicted to finish first. After the player cycled through all of the horses, the Analyzer’s memory would clear.

Mattel Executive Probably Oversold This One

The earliest mention I found in the press for the Horse Race Analyzer was in the January 14, 1980, Chicago Tribune. At the winter Consumer Electronics Show that month, Mattel announced the Analyzer would soon be available for around $100. Maurice Zolotow later wrote that Mattel only showed a a “mockup” at CES and no actual demonstration units were made available.

A few weeks later, Mattel Electronics president Jeffrey Rochlis told the Fort Worth Star-Telegram that the Horse Race Analyzer had been tested on more than 500 races–and the computer selected the top-three finishers correctly 57 percent at the time. Rochlis even claimed that if bettors followed the device’s picks they would see a net return of about 15 or 16 percent, which he noted was a better rate of return than you could get from U.S. treasury bills. Another Mattel executive was quick to point out, however, that the company was not promoting the Horse Race Analyzer as a “get-rich-quick scheme” or gambling device, the latter of which was illegal in Texas.

Still, Rochlis continued to make claims that people would make money using the Horse Race Analyzer. He told Wall Street Journal reporter Steve Sansweet in June 1980, “The significance of this product is that it works, that you win money on it,” and that “greed is the key word” when it came to Mattel’s marketing of its newest electronic handheld product.

Mattel started advertising the Horse Race Analyzer around July 1980, which suggests that’s when the product officially released. The advertising emphasized the supposed “accuracy” of the Analyzer’s recommendations (see below).

As previously noted, Jeffrey Rochlis claimed the Analyzer picked the top-three finishers 57 percent of the time. This figure was rounded up to 60 percent in the early advertising. But Mattel latter focused on a 90-percent accuracy rate in selecting the top-four horses, a figure supposedly backed by real-world testing conducted by the independent National Consumer Testing Institute. Longtime Washington Post horse racing writer Andrew Beyer called BS on this in an August 1980 column:

Mattel expected to get results that would be the basis for a powerful advertising campaign. But when the Institute delivered its results, informed sources say the Mattel people were surprised and chagrined. They had to be. In 1,000 races, the Analyzer had selected 221 winners.

Considering that public favorites win 33 percent of the time, picking 22 percent isn’t much of an achievement. It would be a tolerable result only if the Analyzer were selecting a lot of long-priced winners. But people who have used the device say it gives them mostly short-priced horses.

The most significant statistic from the 1,000-race test would be the net profit or loss from a $2 win bet on all the Analyzer’s selections. The fact that Mattel never cites such a figure presumably is significant. Instead, advertisements trumpet such meaningless statistics as the fact that one out of three highest-rated horses finishes first, second or third 90 percent of the time. Probably every threadbare tout-sheet seller outside the gates of Timonium could make the same claim.

Indeed, one unidentified handicapper succinctly told Knight-Ridder’s Maryjean Wall in June 1981, “Any experienced bettor can eliminate a field down to four live horses, so what’s the point?”

Analyzer Failed to Predict Drop In Its Own Purse

My guess is that most experienced horse bettors saw little point in the Mattel Horse Race Analyzer. I could not locate any sales figures for the device. But going through newspaper advertisements from the time, you can see how quickly prices plummeted as Mattel clearly over-estimated consumer demand for what was essentially a niche calculator.

In the earliest ads from July 1980, Mattel promoted the availability of the Analyzer for $100 at high-end department stores like Gimbels and Jordan Marsh. This was basically an introductory price. By the 1980 holiday season, retailers and mail-order catalogs were advertising units for between $125 and $130.

But by April 1981, the Los Angeles-based department store chain The Broadway started selling the Analyzer for $79.99, a nearly one-third cut from its peak holiday pricing. By mid-1981, Mattel cut its suggested retail price back to $100. It’s unlikely any retailer was stll charging that much. The Broadway lowered its price to just $29.99 by October 1981. And I found at least one store giving away a Horse Race Analyzer for free with the purchase of a Texas Instruments TI-99/4A computer. (Retailers were also desperate to get rid of TI machines thanks to a disastrous home computer price war between Texas Instruments and Jack Tramiel’s Commodore International.)

Aside from lack of demand, another factor that likely worked against Mattel was the presence of competition. Yes, there were other horse race calculators on the market. And at least two products predated the Mattel device’s mid-1980 launch.

The first was the Kel-Co Class Computer, which was manufactured by New York-based APF Electronics. The Kel-Co actually traced it roots back to 1969, when A. Stuart Kelsey produced his first horse race calculator in the form of a slide rule. In 1977, businessman Armond Cannella started working with Kelsey to convert the manual calculator into a microprocessor-based device. The finished Kel-Co Class Computer used the exact same TMS1100 CPU as the Mattel Horse Race Analyzer. And like Mattel’s product, it initially retailed for around $100.

Then there was the RaceTrack Computer distributed by the California-based Starshine Group. (Starshine was likely an importer of discount electronics from Asia.) This device first came out sometime in 1978 and had at least two subsequent models, the Racetrack Computer II and III, respectively. The Starshine units didn’t look much different than an ordinary hand-held calculator. And the advertisements touted how simple it was to use–just 5 steps as opposed to around 30 for the Horse Race Analyzer. Most notably, the RaceTrack Computer retailed for between $29.95 and $39.95 when Mattel’s product launched.

Video Game Crash Ended Mattel Electronics, But the Horse Race Analyzer Lived to Run Again

At the end of 1982, the wheels started to come off the home video game market in North America. Although this crash largely revolved around the Warner Communications subsidiary Atari, Inc., other companies including Mattel were quickly caught up in the collapse. After numerous delays, Mattel Electronics had finally released its own home console–the Intellivision–in 1979 and enjoyed some modest success. But by late 1982 the Intellivision was technologically surpassed by rival toy company Coleco Industries, Inc., which had released its own console, the ColecoVision.

By the middle of 1983, Mattel Electronics was circling the drain. That July, the division laid off 260 people–about 15 percent of its staff–after the parent company reported a first-half loss of more than $100 million, most of that due to the slowdown in the electronics business. A month later, Mattel laid off another 400 employees. In February 1984, Mattel shuttered the electronics division altogether. The Intellivision was sold off to one of Mattel’s marketing executives, Terrance Valeski, who kept the console alive, primarily as a mail order product, until around 1990.

As it turned out, the Horse Race Analyzer had a similar fate. Only in this case, it was co-creator Bill Mandel who purchased the rights to the product back from Mattel. I couldn’t find any details on exactly when this happened or how much Mandel paid. But we do know that in January 1983, Mandel created a company in Connecticut called Advanced Handicapping Technologies, Inc. (ATHI). It was through ATHI that Mandel continued manufacturing and marketing the Horse Race Analyzer, with references to “Mattel Electronics” on the housing replaced with the more generic “Handicapping Computer.”

From this point forward, it appeared that ATHI sold the Analyzer mostly through the Sharper Image catalog. I found one listing in the 1988 catalog (see below), which priced the Analyzer at $69. Other news reports confirmed the device sold for between $50 and $70 at various times. Sharper Image also used to run retail stores, and I found a couple of newspaper articles referencing sales of the Analyzer in those. But as far as I can tell, the Analyzer was never sold in any other mass-market retailer after Mattel exited the picture.

An entry for the “Horse Race Analyzer” in the 1988 Sharper Image catalog.

I also don’t know exactly when the Horse Race Analyzer left the market for good. The last newspaper reference I could find to the Sharper Image sales was from August 1996. The U.S. Patent and Trademark Office canceled the “Horse Race Analyzer” trademark in January 2003. Later that year, in December 2003, Bill Mandel died at the age of 77. ATHI was dissolved the following November.

Notes from the Random Access File