The dark side of India’s gaming boom

Mobile Premier League (MPL), the real-money gaming service that hosted the tournament, spent months ensuring their servers could handle this volume of traffic. They hired Viswanathan Anand as brand ambassador to promote the tournament and promised the winner a prize money of ₹5 lakh. Entry into the tournament would be free—all you needed was a mobile device and an internet connection—but 30,000 winners would walk away with some part of the total prize pool, exceeding ₹11 lakh.

V.S. Rathanvel, an 18-year-old International Master chess player from Coimbatore, beat one grandmaster to win the tournament. Grandmaster Ankit Rajpara came in second. Both invested their winnings in chess training.

In the months since the Chess Mahayudh, the real-money gaming platform MPL has grown an extraordinary 2,000% in gross merchandise value. And it isn’t just MPL; the real-money gaming industry —where you play games online for stakes —is booming.

According to a Deloitte report, games like online poker and rummy have been growing aggressively year-on-year for the past four years. The ongoing lockdown in response to the covid-19 pandemic has only given further fillip to the gaming industry. Varun Ganjoo, marketing director for PokerBaazi, reported that their site’s traffic has risen 25% in recent weeks, with transactions also rising by 20%.

Growth figures, in revenue and customers, range between 50-200%. (Revenue of online card games grew from ₹258 crores in 2013-2014 to ₹3,400 crore in 2017-2018.) Fantasy sports, on the other hand, has been growing exponentially because of Indian Premier League (IPL) and other premier league sports. Dream11 is by far the largest player in the fantasy market. In an interview with Microsoft’s Satya Nadella late February, Mukesh Ambani said he believes that gaming will be the next big industry in India. “It is very hard to imagine but gaming will be bigger than music, movies and TV shows put together," he said.

Industry experts widely cite three reasons for this peculiar rate of growth: cheaper smartphones, better internet connectivity, and most significantly, the digital wallet, which allows for immediate withdrawal of winnings and payment of entry fees.

But there’s a fourth tool now available to the gaming industry; one they’re particularly hesitant to talk about. It is your data: when you like to play, who you lose to and win against, how much money you spend, the tables at which you like to sit, and how you respond to changes in gameplay. Deployed effectively, this data could take a potentially addictive product and make it impossible to resist.

The legal vagueness

Gambling is a state subject in India, but the standard in most states is this: If a chance-based game is played with stakes, it is gambling. But if a skill-based game is played with stakes, it is not gambling, but real-money gaming (aka skill-based gaming). More skill, less chance is legal. Less skill, more chance is illegal. With few exceptions, including Sikkim, Goa and Nagaland, this is the norm.

Over the course of many court judgements, a standard has evolved industry-wide. A two-step test is used to determine whether a game is predominantly skill- or chance-based. Test one: A player must be able to choose to lose. He exercises some agency in winning. (Snakes and ladders would fail this test, but Ludo would pass.) Test two: A player’s performance must improve over time. Consistently and steadily. The second is a tough test, and one that troubles the gaming industry immeasurably, because in the real world, performances often plateau, or rise and fall cyclically. Rarely, if ever, does a performance have a steady, positive slope.

Playing a game for stakes is legal in most parts of India only if it passes both tests. It is then real-money gaming, not gambling. That is why a Punjab and Haryana high court ruling has adjudged that the version of fantasy football played on Dream11 (where you choose 11 players from across a league to beat the odds on the market) was akin to a business, requires skill, and was, therefore, legal. A Supreme Court ruling says that horse-racing is a skill-based game because one needs to know a thing or two about horses. Texas hold ’em poker is found to be skill-based in most states.

The game players

Sachin (name changed) is a vice-president of marketing at a startup that is valued at multiple billion dollars. He began playing poker 10 years ago, when he was in his mid-20s. He had visited London and seen people play poker seriously, that there was reading and learning one could do, and that it was a challenging sport. He began to play weekends, with friends. They would begin Friday nights and continue into Saturday mornings.

Five years later, he cut down on playing with friends and began to play online. “I found that online poker is the most efficient version of poker. I don’t have to get dressed. I don’t have to travel to the game. I don’t have to meet people. I can spend all that time studying and playing poker instead. This is more poker per unit time."

Sachin is unmarried. He plays poker 10-15 hours every weekend from his home, on his preferred platform, PokerBaazi.

“Sometimes luck works in your favour and sometimes it does not. But if you read, watch games on YouTube, brush up skills, over time, the luck will even out. Over time, you will be making money." He insisted he plays for entertainment, but believes it is important to beat the odds. “Some people have football. Some people have tennis. This is my passion."

The MPL office on the outer edges of Bellandur, Bengaluru, is strictly white wall and glass, sparse and clean. Inside, coders sit in long rows. Dibyajyoti Mainak, consultant general counsel and the man in charge of communications and legal function at MPL, said: “We have internal data on randomly selected players of all our games on our platform. We’ve graphed their performance over time, and we can see that their performance always has a positive slope".

Mainak pulls out a few graphs on his laptop. While the performance vs time slopes are always positive, some slopes are steeper than others. “Temple Run is a higher skill game than most," he demonstrated, “because it relies on muscle memory, and that improves with time."

“For a really good game," he continued, “ease of understanding has to be low, so that as many people can play it as possible. But ease of mastery has to be really high, to keep players playing the game."

A quiz, played for stakes, hits the sweet spot. The format is easy to understand by everyone, he said, but it is impossible to comprehensively master. MPL has hosted several big quizzes, including one on Karnataka on the occasion of Karnataka Rajyotsava and several on Bollywood.

But within these games, all players are not equal. The choices a player makes are stored. This data is used to categorize her first, then target her with offers, bonuses and promotions based on those categories. “There are players who play ₹2-games and there are players who play ₹200-games. There are players of lesser and greater skill. Based on how many games a person has played, and what tier of player he is, we can target the customers and give them free entry in relevant tournaments we’re promoting."

Sometimes, the platform will give discounts to specific players on entry fees, effectively changing the odds on the game. This is called a “seeded entry". With funding from Times Internet, Sequoia India and GoVentures among others, MPL can spend on winnings and entry fees for valuable customers, because it has cash to burn.

And if they can target effectively, they can retain customers effectively. Deepak Venkatramani, chief marketing officer at Nostragamus, another gaming platform, agrees with Mainak on most counts. He talks about how data is used to incentivize coming back every day. “I have found that if you allow a customer to spin a wheel for free, it will increase retention by 30%. Even if nobody wins anything out of it!" If the player comes back for a string of days, he forms a habit.

Like Mainak, Venkatramani agrees that quizzes are among the most aggressively growing category of games played with stakes.

Because of the data, the platforms know who their valuable customers are. Those customers can be targeted over others.

Research has demonstrated that what contributes to dopamine release in a game is not just the outcome, but whether the outcome is more positive than expected. “If people lose a bunch and that lowers their expectations, that will increase how happy they are when they finally do win," said Rob Rutlege, a neuroscientist at University College London, in an interview with the BBC in 2016. This information can be used to deploy bonuses to customers—unexpected earnings delivered to high-value customers after the game is over, or even during the gameplay.

There are some old-school ways of targeting too: MPL has created Whatsapp groups for the “most active and visible" women gamers. This was an effort to build a community of “vulnerable" players. These women have since built their own self-sustaining network of tournaments online.

Other tricks include: sound triggers to reward performance. Colours like red on the screen, believed to be more arousing. The well-timed notification. Yet much of how data is used to customize user experience is unknown, because nobody in the industry is willing to talk about it.

The regulatory wilderness

Yudhishthir (in Mahabharat) had one weakness, and it was gambling. What was worse: Shakuni knew this weakness. Now imagine, for a moment, that Shakuni could do what he did at scale: identify all the Yudhishthirs in the world, and egg them towards higher and higher stakes.

Let’s complicate this universe further. Imagine that the Yudhishthirs have a little device in their pocket through which they can gamble whenever they want. On the Metro. At the dentist’s. During lunch break. And, with advanced tools in behavioural technology, Shakuni knows the triggers that work. He knows when to deploy them for maximum impact. He can reach Yudhishthir any time he likes. Here, Yudhishthirs don’t stand a chance.

As of now, the online gaming industry, where the instrument of gaming is also the device that tracks your behaviour, exists in the regulatory wilderness. The law only checks for skill and chance, but the true danger of addiction lies elsewhere. As data science improves, gaming platforms will be able to look for patterns of play. They will design incentives, offers, discounts, notifications, the chance to win a gold coin or ₹20,000, and will test to see which ones are more effective in producing the behaviour they seek; which ones improve “stickiness".

They will be able to measure frustration, release and regret. Since most of the gaming platforms interface with an external wallet, the wallet will also have data on how much a player spends on gaming. The wallet will see the frequency and volume of deposits and withdrawals over time. They will be able to understand the broader context of the gamers’ financial habits.

The revenue model of all real-money gaming platforms is that they take a cut, varying from 10-25%, of the total stakes in gameplay. Which means it is in the interest of the platform to raise the stakes of play. This is also to say that in a casino, the house may lose because of bad luck, but in real-money gaming, the house can never lose.

There are, of course, no perfect solutions in sight. Under the General Data Protection Regulation (GDPR), gaming operators in Europe have to explain what data they collect and how they use it. Because of the economic costs of this, many well-loved games have had to shut down. Any new regulation in India would come down heavily on a fledgeling industry.

But the laws must evolve, because this old industry has been transformed by the tools of surveillance capitalism. It’s not just that this house always wins, it’s also that this house knows who you are and how you play.

Sneha Vakharia is an independent journalist based in Bengaluru