Starcraft Statistics: Tournament Multipliers and Understanding a Twitch Audience Part 1

Published on 03/04/2017 09:00 PST by ROOT Gaming

Introduction

Often times when discussing games and how they differ from a viewership standpoint we dumb it down to just totals. This is fun and easy but it doesn't look at things deeper. Now I want to note right away, this isn't a discussion about Starcraft's life or death or anything like that, rather this is a discussion about how looking deeper into Twitch numbers can reveal the nature of the game's audience.

I first thought about writing this after reading Newzoo's monthly review of the top Twitch games, all credit for this section goes to them. Now Newzoo does something interesting, they look at total hours but they also look at eSports hours. By doing this they can give us a look at a game's audience. For example if we have two games:

- Game A: 10 million hours watched, 9 million eSports hours

- Game B: 10 million hours watched, 5 million eSports hours

We can see both games are similarly popular but what makes them popular is very different. Game A's popularity is drawn from it's eSports events while Game B draws much more popularity from it's regular streamers. We are going to call this the Tournament Multiplier because it allows us to see how much a boost a game gains from a tournament on average.

Now this does vary month to month for most games, Dota 2 has few tournaments but in the months those tournaments take place viewership goes out the roof so some months will have high eSports hours but others will have lower eSports hours. So using a full year (2016) we will exclude games which had high viewership the month they came out but then dropped off in following months. A game like Call of Duty: Black Ops III is a good example, it had huge viewership for three or four months but disappeared after that. To assist with this a game will also need to be in the top 10 games in viewership for more than two months to qualify for this study so it won't be skewed.

A quick note before diving in, a game like Overwatch which has maintained it's popularity on Twitch, unlike a game like Call of Duty, didn't have a full 2016 and about much of its total viewership hours came in beta or early after it's release and as such didn't have much tournament scene, which came a few months later, this is why some of it's data is a bit odd.

The Data

Let's start with Starcraft 2. Starcraft 2's breakdown had 50.85% of all hours watched coming from tournaments. This is a nice even split between tournaments and streamers. But without context we don't know it this is high or low, which makes it hard to analyze, so let's look at the table. 

Game

eSports Percentage

Starcraft 2
50.85%
Super Smash Bros.
48.13%
Dota 2
43.19%
CS:GO
37.81%
SMITE
29.65%
League of Legends
26.69%
Heroes of the Storm
24.54%
Rocket League
21.43%
Call of Duty: Black Ops III
14.50%


Hearthstone
14.38%
World of Tanks
14.18%
Overwatch
10.67%


Right off the bat we see Starcraft 2 is the highest game in eSports percentage for 2016.  Right below are the top fighting games in Super Smash and Street Fighter. At the bottom of the list we have Overwatch, which again I mentioned is a bit odd since much of it's viewership hours came in beta before any tournaments were organized so it's a bit of an outlier. After that we have World of Tanks and Hearthstone. With the numbers all here we should look at what they mean.

First we should look at the big three games on Twitch: League of Legends, Dota 2 and CS:GO. All three are very different despite all three being very large. League of Legends boasts some of the largest tournaments in eSports history but their streaming scene is also huge and even when no tournaments are going on League generally still tops Twitch with some of its streamers having larger audiences than the tournament totals of smaller games. While not as robust as League we find that CS:GO still has a strong streaming scene that occupies over 60% of its viewership hours. Dota 2 is closer to CS:GO but it's streamers don't occupy nearly as much viewers as the others and when there are no tournaments Dota 2 often drops out of the top 5 games on Twitch. Each of these big games is unique, League is so large that its massive tournament scene is only paled by it's even larger streaming crowd. CS:GO has the best balance of the big three while Dota relies the most on it's tournaments to carry it.

The next tier I want to look at is those games while are primarily streamer focused games. These are World of Tanks, Hearthstone, Call of Duty and Rocket League. These four games all had either high peaks of viewership (Call of Duty), maintain a stable but small crowd (World of Tanks) or are very large games (Hearthstone and Rocket League). All of these games are incredible popular almost exclusively in part due to their streamers. Nothing exemplifies this more than Hearthstone. Now Hearthstone is one of the most popular games on Twitch, ranking between 3rd and 4th most popular depending on the month but it's thanks to names like TrumpSC, Kolento, Kripp, Amaz and others. The strange thing when looking at this category of games is they all have tournament scenes that are strongly pushed by their developers and publishers but almost none are very successful eSports. Again to look at Hearthstone, despite being one of the most popular games out there we find that its tournaments are generally lower in viewership than you'd expect.

Next up is the group that is a bit confused about what they want to be. It is SMITE and Heroes of the Storm. These two games are the 3rd and 4th largest MOBAs but neither has been able to gain either the tournament numbers of a game like Dota 2 or the streaming success of League. Now both are between 25% and 30% eSports viewership percentage which means they generally get by on streamer hours. But unlike League which has such a huge audience that a low eSports viewership percentage doens't hurt it because both sides are so large. SMITE and Heroes don't have that luxury and their eSports side flounders and their viewership is middling.

The final group I want to examine is the one I want to call the Legacy section. I'll explain that in a bit but first the games in this category are Starcraft 2 and Smash. Both of these games sit at or around 50% viewership coming from tournaments. Of the two Starcraft 2 is the much larger game, it had more total hours, more eSports hours and more streamer hours by double that of Smash but their break down is similar, while both have streamers who are popular they thrive due to tournaments. This is where we get to the Legacy title for this group, both games (in the case of Smash it is both Melee and Smash 4) are ones considered the godfathers of the eSports scene. When a Smash tournament is on you get viewers from all fighting games and those who may not follow the scene but played Smash as kids or with their kids currently. The same with Starcraft 2, while many may not know the streamer when a tournament comes on you have League fans, Dota fans, CS:GO fans watching. This is because both games have strong name recognition, fans of all eSports know about Smash and Starcraft. This unique sense of Legacy is one reason why both games are still revered despite neither being the largest when it comes to viewership. Companies like ESL and DreamHack want these games on their promotional material because of the cross promotion they all.

Next Step

What we can take away from this already is helpful but next week I want to dive into this Tournament Multiplier more by examining exactly what kind of jump each of these games gets by having tournaments and whether it's smart for big games with low eSports percentage to keep pushing tournaments and what small games can focus on.

We'll also look at streamers within different games and compare them.

About the author:

Topher is an American football and eSports writer with a focus on statistical metadata research. You can follow Topher on Twitter

Wardi 343