High roller: A WVU doctoral researcher watches the odds

By Micaela Morrissette, West Virginia University

Keno. Lotto. Hot Spots. Scratch-offs. Fantasy sports. Powerball.

Bingo!

West Virginia’s got ’em all.

“In West Virginia, pretty much everything is legal. We are the only state in the U.S. where you can still bet on greyhound racing. Isn’t that crazy? We have legal sports betting, we have venues with video gaming terminals, we have online lottery purchases. We have iGaming, which is online gambling, from poker to slots. My expertise is in iGaming, and it seems like every month West Virginia sets a new record for iGaming revenue.”

That’s Matt Mullis talking, a doctoral student researcher at the West Virginia University Center for Gaming Research and Development.

The WVU Center for Gaming Research emerged in the wake of the state legislature’s 2024 passage of the Responsible Gaming Research and Industry Development Act, which allows WVU to analyze transactional data on all gaming activities statewide, in partnership with the State Lottery Commission.

“Access to that information puts WVU on the cutting edge of gambling research,” Mullis said. He now surfs waves of data showing gambler behavior of every kind, from engagement on video lottery terminals to bets placed with mobile casinos.

Mullis has found that the average mobile slots bet in West Virginia is just one dollar.

But he has found up to 43 million mobile slots bets placed at a single operator’s slot machines in the space of a month.

And he has found that those 43 million bets were placed by only around 16,000 players.

Mullis has discovered that on days when Social Security payments happen — every second, third and fourth Wednesday — demand for Hot Spot slot machines increases. He said that suggests some part of the transfer payment is being channeled into gambling right when it arrives.

He has also learned that, despite repeated returns to the roulette wheel, he himself has no appetite for risk.

“I just went to an incredible gambling conference in Vegas,” he said. “It was pretty extravagant, hosted at the Bellagio, and I learned so much from people from various fields who had very different opinions. There were presentations on how to make money sports gambling and there were people speaking about the act of gambling as a mental health problem. But the most helpful advice I got there was, ‘Be in the market.’ I’ve learned a tremendous amount about gambler beliefs just by talking to gamblers or playing the games myself.”

So Mullis dutifully pulls out his phone every once in a while and opens Caesar Sports or another casino app and bets a dime or so — as research. He still hasn’t felt the gambler’s itch, though he emphasizes that’s not for lack of trying by operators like DraftKings and FanDuel, who entice players to chase a drug-like high by starting them off with “Free Bet” promotions and make it amazingly easy to link bank accounts to bettor accounts.

While humans have indulged in games of chance for more than 4,000 years, the mid-2020s are something special. Gambling activity is surging in the U.S., fueled by the rising legality and popularity of mobile gaming and online sports betting, including record numbers of wagers on the 2026 World Cup.

Marketing has also played a role, and Mullis pointed to one study showing that the average National Hockey League broadcast contained an average of three and a half references to gambling per minute.

“In one month, 25% of the bettors in our mobile slots data placed more than a thousand bets each,” Mullis said. “Our work at the Center for Gaming Research is largely about observing the changing behavior in this new market. By giving us this data for research, West Virginia is looking to understand the risks and protect consumers.”

Mullis can zoom out for a wide view of the data. For instance, he can see that the average West Virginia Hot Spot takes in, before taxes, about a thousand dollars a day. West Virginia taxes that revenue at about 50%, compared to the roughly 10% tax the state assesses on iGaming revenues. This makes Mullis curious about the effect of iGaming’s rising popularity on Hot Spots and, by extension, state tax coffers.

Or, he can zoom in, all the way to the level of the individual bet. He knows some outlier gamblers by their behavior, like a few “whales” who deposit hundreds of dollars into their iGaming accounts nearly every day, like clockwork.

But that doesn’t tell Mullis how betting exerts its hypnotic fascination, or why some are driven to gamble it all away.

“People enjoy the act of betting,” Mullis said. “But if someone wins and then we see that they bet more, are they betting more just because they have more money to bet with? Or are they betting more after they win because they feel like they’re hot, which implies players are incorrectly updating their beliefs? My research does suggest that past betting outcomes really matter, but even with the unbelievable access to data that we have, it’s hard to identify certain effects in this market, hard to understand what people’s intentions are. We still can’t read their minds.”

He might be able to change them, however. One down-the-road application that could come out of the WVU research is flagging problem iGaming behaviors, like re-funding an account three times in an hour, or wagering a thousand dollars in 30 minutes, that could merit the imposition of a short break by the app.

Mullis first cracked open the Pandora’s Box of statewide gambling data thanks to a question posed by Professor Brad Humphreys, director of the Center for Gambling Research and associate dean of academic affairs and research for the WVU John Chambers College of Business and Economics, who would become Mullis’ co-adviser.

“One of the first pieces of data that Brad and the Center got from the state were daily machine-level revenues from West Virginia video lottery terminals between 2024 and 2025,” Mullis said. “If you pointed out a machine in the state, we could tell you exactly how much money that machine got on a certain day.

“So for our Urban Economics class, Brad pulled that up and asked us if we had any ideas about what to do with the data. My hand shot up, and I said, ‘On days when there’s a big sporting event, does slot-machine gambling go down because, say, people are paying attention to the West Virginia game?’ He said, ‘All right, let’s find out.’ Ever since then, I’ve had access to an unbelievable amount of data through the Center for Gaming, and it’s so easy to ask questions no one has had the opportunity to ask before.”

Usually, Mullis explained, data representing individual bets comes from lab experiments or surveys, not the real world. That means behavior is likely skewed by the lab environment, and sample sizes are restricted.

But with the data rolling in from across West Virginia month after month, he has been able to show that, yes, people do hit the slot machines less when WVU or the Pittsburgh Steelers are playing.

Now Mullis’ research is headed down the rabbit hole of “loss chasing” in iGaming — when, in order to win their money back quickly, a gambler decides to increase wager size after losing a few smaller bets in a row, creating a snowball effect.

“Gambling is a losing proposition,” he said. “You should go into the market expecting to lose. And when it comes to the lottery, for instance, sometimes the people playing are the people who can’t afford to lose. Yet many states’ lotteries, like West Virginia’s, generate millions in funding for infrastructure and social programs. So there are moral questions that are complicated and interesting, and a lot of them I don’t think you can answer without the data we’re getting at WVU now.”