Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 23, 2014
arrowPress Releases
October 23, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

How to make your games better to stream, according to Day[9]
How to make your games better to stream, according to Day[9]
November 5, 2013 | By Christian Nutt

November 5, 2013 | By Christian Nutt
More: Console/PC, Indie, Art, Design, Business/Marketing, GDC Next

Sean Plott, better known as Day[9], is famous for streaming games like StarCraft II online. He argued, in his GDC Next talk, that developers can and should make their games better for streaming, and explained how.

Twitch has 45 million uniques a month, and viewers typically stick around for 90 minutes.

"With the rise of online video there's now an incredibly easy free way to make marketing tools for your game that people are hungry to watch," says Plott. "Most importantly, it's native to what a game is."

"If you can make it easy, convenient and fun for the community creators to make their own content then great, you get free marketing." And you build a community around your games if you do it right: "Online video is about community... with video, we have comments, linking and Tweeting, people are encouraged to send each other videos. It's about your community doing it for you."

Almost 50 percent of video content for a game on YouTube is community-generated, says Plott. And a huge proportion of this, of course, is pure gameplay video. "In thinking of the online video audience you are thinking about how to make the gameplay itself viewable.

Making a Game Pleasurable to Watch

One thing Plott emphasized is that a game has to be easy and pleasurable for not just "influencers" and pros to stream, but for all players. That's because "you want to make sure there's a constant rise up" of novice players to community contributors.

And if your game is appealing when streamed, you have a chance to capture both viewers "who have this game as a lifestyle hobby" as well as new players: "They want to stop in and watch some gaming content and you want the easiest, smoothest path for them to become hardcore players," says Plott.

And you have to think about what your game will look like to a new player when viewed by a newcomer. They didn't sit through all of your tutorials and play for tens of hours -- but they should be able to understand what they see.

"There needs to be a way for a new guy to hop in and get what I call 'game summary information,'" says Plott -- information like a player's objective, how powerful they are, and more. If a new player can, at a glance, get a bead on the game, they'll be more likely to stick around.

Mind Mechanics and Chunkiness

Games with strategic decisions are particularly well-suited to streams, says Plott: "The core, fundamental question in mechanics that makes a game fun to watch is 'What would I do differently?'"

This applies to skill-based games like Street Fighter, League of Legends, or Super Meat Boy and also sandbox games like Minecraft or SimCity... but not so much single-player linear games. But ones with choices (The Walking Dead) or possibiltiies of different play styles (The Last of Us) do well, Plott says.

But most important, perhaps, is what he calls "mind mechanics" -- where players can anticipate what the player will do and discuss it. Plott referred to Blizzard's collectible card game Hearthstone in this regard.

"There's a single action," he says, after careful deliberation. The player makes one move. "The viewers get that first 80 percent of gameplay, they can watch and see the cards and make these same deliberations on their own; the only difference" between the player's experience, and the audience's experience "is when the last step gets chosen. The last moment is what everybody wants to talk about."

Not all games work this way, and that's okay. Think Super Meat Boy: "Is there a way for someone to record a two or three minute feat in your game that is amazing?" Games that are broken into discreet chunks are good, too: "Is there a way to record just a chapter of the game?"

Related Jobs

Nexon America, Inc.
Nexon America, Inc. — El Segundo, California, United States

Localization Coordinator
Petroglyph Games
Petroglyph Games — Las Vegas, Nevada, United States

DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
Xsolla — Sherman Oaks, California, United States

Senior Business Development Manager


Chris Zukowski
profile image
Do you think the viewers empathize with the streamer or are there for the game? I feel like they watch Spelunky because they want to see the streamer struggle and triumph.

Robert Crouch
profile image
I think the viewers come for the game and then empathize with the streamer.

That experience though, gives a stronger connection to the game. Spelunky as an example was a game that I first saw streamed. Watching the streamer play and fail made me hang around to see if he could get it. But more importantly for mossmouth, it made me want to try it myself, because I started to see where he was screwing up and was certain I could do better.

It was advertising, but from a viewer's perspective it was the sincerest kind of advertising. The game was laid bare for me to see.

Andrew Wallace's comment about Hearthstone is an example of how a game has variable streaming effectiveness. Spelunky is very easy for a new viewer to understand. Starcraft is pretty easy to appreciate. Hearthstone, or M:TG are a bit too complex to learn in real time. Even something like League of Legends requires a pretty strong understanding of each hero and the metagame to get anything from.

Starcraft 2 is a pretty complex game. I would say that it's probably strategically more variable than something like League of Legends. But just because it might be more complex, doesn't make it harder to watch. Watching SC2 is easy, you know the goals, one side wants to wipe out the other. You quickly pick up the big building blocks while watching; keeping bases is good, destroying them is a small victory.

League of Legends is harder to watch for a new player. You have heroes with non-obvious skills if you haven't seen them before. The goal is more nebulous. You know that killing enemy heroes is good, but a lot of things that are important to winning kind of get lost. Even combat is frenetic and fast with a bunch of skill combos that if you can't identify them by the visuals you aren't sure exactly what's happened.

That said, LoL is a more popular stream than SC2 recently. I think that has a lot to do with the fact that it's LoL players who watch the stream, and they do so instead of playing. I think that has to do with the fact that LoL can be a very stressful game, and it's a game that's pretty dependent on external factors like your time. Once you're familiar with the game, you can get some of the pleasure of playing the game by watching, without having to suffer yourself through an uncooperative team, or being in a situation where you're severely outclassed.

In summary, I think that the design of the game dictates how easy it is to start watching a stream. The ease of watching a stream of an unknown game is not entirely coupled with the complexity of the game, instead how it presents itself real-time.

I think that the viewers can empathize with a good streamer, and enjoy some of the achievements of the streamer without achieving anything themselves. I think that these feelings can push the viewer to want to emulate those achievements.

I think that the design of the game can create a situation where the game is more fun to watch than to play in some circumstances. I think this is generally the case when the viewer feels they could not possibly emulate the streamer's achievements, or when the player would become stressed playing the game. I'm not sure if this is a good thing for a game or not. I think it might be, because it allows viewerships to form communities, for streamers to stay motivated to keep streaming, and for this streaming advertising to continue. Another example of this is speedruns, another popular category on twitch. People watch that instead of trying to compete because of how much more commitment it would take to even get close to the streamer's level.

I think that a high skill-cap and a noticeable way of measuring that is a good way to make stream content that's interesting to watch instead of playing.
I think that a game that can be frustrating (Something like LoL is a breeding ground for frustration with the combination of external factors, cumulative gains and losses and social dynamics.) can utilize streaming to give the players a way to experience the game when they are less willing to tolerate that potential frustration.

If designers can make their game interesting to watch while streamed, I think it can not only promote their game like advertisement, but it can also make it more enjoyable by making it more intense, but giving players a way to cool down while still experiencing it.

Andrew Wallace
profile image
I think it's interesting he uses Hearthstone as an example. I tried to watch a Hearthstone stream from a SC2 caster I liked and was terribly bored, because I didn't have nearly enough time to read the cards to figure out what was actually going on. Really interesting points to consider overall, though.

Scott Lavigne
profile image
I think there's initially going to be some overheard for gaining knowledge before this can apply in most games, but Hearthstone is overall pretty good about it. Even if you don't read a card, the game is pretty good about visually conveying what happens. There are different effects for different mechanics, and everything has a symbol/animation with differently colored numbers to tell you how something's adjusting.

I think in the example of Hearthstone, the idea is that the viewers look at the streamer's hand and as the streamer deliberates on what to do (usually for at least a minute, probably with very little dialogue), the viewer can decide how they would spend their resources in that situation and begin talking about it in chat. It gives viewers a feeling of vicarious accomplishment to figure out how to play a turn like a pro, I think, even if they arrive at the right answer for the wrong reasons or resolve to disagree with the streamer altogether. Viewers get the urge to play the game themselves so they can have these decisions play out. Every turn they watch the streamer is a tiny hit of the game, but all isolated instances since they can't see the results.