If you’re not a fan of the video game Fortnite, or a long-time user of livestreaming platform Twitch, the name Ninja is likely new to you. Offline, he goes by Richard Tyler Blevins. But within the gameplay-streaming subculture, Ninja needs no introduction.
In the late 2010s, he was by far the world’s most popular Twitch-streamer, with more than 15 million followers. Ninja’s outsized profile has had enormous positive spillover effects for his preferred platform. However, this extreme popularity has also resulted in Twitch being disproportionately dependent upon a rare set of superstars like Ninja. The mixed blessings of superstardom for digital platforms are the subject of a recent working paper co-authored by Brad Greenwood, professor of information systems and operations management at Costello College of Business at George Mason University.
“For us, Ninja was a useful empirical device because he was pledging to stream for 12 hours a day without any vacations or other interruptions. We wanted to know what happened when he went on a honeymoon or to the Super Bowl to be promoted,” says Greenwood. Given this predictability, his sudden, short disappearances gave the researchers a unique opportunity to study the reaction of other streamers and their ability to capture the unmet demand from viewers.
Greenwood contrasts this study with his earlier research that examined the repercussions of Ninja leaving Twitch entirely to join a rival platform, asking, “Does temporary absence send a different signal from the dampening effect on content liquidity we see when Ninja gets poached entirely?”
Greenwood and co-authors, Dominik Gutt of RWTH Aachen University and Jens Forderer of University of Mannheim, used data from Sullygnome, a website that tracks data on Twitch channels, to identify the days on which Ninja had not streamed during the period January 2018–July 2019. They codified 80 absences as strictly unexpected—announced less than 24 hours in advance and mostly due to “family obligations, business travel, exhaustion, and sickness.”
With Twitchtracker and Sullygnome, they found that channels were consistently more likely to supply content on the days of Ninja’s absence, to the tune of a 4.1 percent relative increase in the likelihood of a channel streaming Fortnite content on the days of his absence compared to the days when he actively streamed.
In other words, when Ninja unexpectedly went dark, lesser-known livestreamers producing similar Fortnite content reallocated their content supply. “In case of the temporary absence, we saw a standard product market competition reaction: the creators responded to unexpected absences by changing when they stream and how much content they provision,” Greenwood says.
His paper concludes that short-term, unexpected superstar absences positively affect platforms by encouraging other channels to fill the void. “Temporary versus permanent absenteeism sends a different signal to the market. It creates a demand vacuum that is filled by other content creators. The platform will rebalance its liquidity after the moments of unexpected dearth of content,” Greenwood explains.
Unlike employees, streamers are independent, and platform managers have limited control over their decisions on when to stream. “For platform managers, the problem lies in the fact that content creators are subject to the same stuff that all employees are subject to, like getting sick or dealing with family issues,” Greenwood says.
He argues that the key to success for platform managers lies in fostering healthy digital ecosystems. Platform managers can diversify away from dependence on superstars by leveraging unexpected absences as opportunities to groom or spotlight up-and-comers. They could also use recommendation algorithms to direct viewers toward alternative content creators in lieu of superstars.
These takeaways are also applicable to other digital platforms where supply must meet demand in real time. Greenwood uses the example of ride-sharing services: “Uber proactively texts drivers to tell them where surge is. The more drivers they have on the platform operating at any given time, the more liquid their market is, which means less waiting time for potential riders. If the riders see that the wait is 25 minutes, they're immediately going to check another platform.” Maintaining a vibrant community able to absorb shocks and continue producing services and content can help navigate unpredictable liquidity, he concludes.