In this post I’m going to talk about how variance is a major factor of many multiplayer games’ success. For those following the Zems project, you’ll notice that we plan on making some changes after getting feedback on our LD submission that are quite contrary to the positive feedback we received. To the uninformed, our LD entry was an board/card game hybrid that had no variance because it allowed the player to pick any card from his/her deck instead of the typical ‘draw a card’ function most card games have. The number 1 praise we received from the LD was something along the lines of this: “I love the fact that I can’t lose because of drawing something unlucky.” So why are we removing this feature from the game when everyone loves it? That’s what I’ll be covering in this post.

Games Without Variance

I think it’s important to begin by covering a couple games that do not have variance. Let’s start with the oldest one – Chess.

Chess was at its greatest peak in popularity during the Cold War, and a large part of this goes to Bobby Fischer, a grandmaster who became famous for winning the World Championship in 1972 after 24 years of Soviet dominance. Chess has an advantage over many games today in that it is a centuries-old game that almost everyone knows. However, despite the fact that Chess isn’t too complex to learn and has been around for a long, long time, Chess has been on a decline for the past two decades. There are a number of theories as to why this is, but I will present some of my own here:

  • Chess is slowly getting closer and closer to being “solved.” Computer chess has pushed the game to the point where computer programs are able to draw the majority of the time in computer-vs-computer tournaments. Even in non-computer chess, the best grandmasters of today are able to push most games to a draw, as evidenced by the recent World Championship where the first four games ended without a victor.
  • Chess is purely skill-based. While many see this as a positive aspect, what purely skill-based games generally (but not always) do is create one-sided positive experiences. This means that the better Chess player will, most often than not, beat the weaker Chess player. This discourages both players from playing each other because the weaker player feels like he or she will get stomped whereas the stronger player may not find sufficient challenge to want to play the weaker opponent.
  • Chess doesn’t have virally exciting moments. Sure, you may have thought deeper than your opponent and will find excitement when you gain the upper hand because of it, but you’re not going to share this on your Facebook or tweet about it – chess positions and moves just aren’t viral. They aren’t exciting to share, and a growing number of popular games today design not only for the exciting moments, but they aim to make those moments viral as well.

Now for a modern example: StarCraft 2. It’s the successor to the game that practically launched the eSports phenomenon a decade ago, but has since failed to maintain popularity (even within the eSports scene) since its release. Sure, there are a good number of people who care about SC2 tournaments, but the vast majority of gamers are more interested in Call of Duty or League of Legends. As with Chess, there are a number of theories as to why this might be, and I’ll present some of my own theories below.

Note that I played SC2 during season 1 and reached Diamond ranking just as Masters rank was introduced, so I am no stranger to the game and can most likely still play at gold level if I was forced to play a game today.

  • StarCraft is purely skill-based. The same points from Chess apply here, except the difference is that this is real-time skill instead of turn-based (slow) skill. In fast-paced, intense games, every small thing counts. This means that forgetting to chrono boost (or misclicking) your first Zealot so he arrives 1 second earlier than usual could lead to you failing to wall-off against the zergling rush. If those lings get into your base and hit your resource line, you’ve effectively lost the game within the first 5 minutes. This is skill demand on a very subtle level and many players are dismayed by the fact that they have to account for this all the way down in bronze league (the lowest of the low on the skill system) because bronze opponents typically like to employ risky rush strategies. Does this sound fun at all to you? No? Why would you want to keep playing this game after finding out that you need to either perfect this small micro-aspect of your early game or just auto-lose to any well-timed zergling rush?
  • Repetitive games. Back when I played StarCraft, even in season 1, much of the game was ‘solved’ in terms of builds. I scout their base and if I see X, I know they’re on strategy Y and I need to employ strategy Z to stay on par. This leads to repetitive games that only create unique moments after the 10 minute mark or if my opponent was on some unorthodox strategy. StarCraft doesn’t really lend itself to an environment of constantly changing situations until later on in the game. Balance updates have nerfed certain early builds and promoted new ones, but by the time the Heart of the Swarm expansion was released, it seemed like nobody cared about StarCraft 2 because the amount of innovation was so low. Everyone had seen practically everything, and the most optimal builds for each faction for each matchup and each adaptive strategy had been figured out by the pros.

It is true that non-variance games have a certain crowd, and there is nothing wrong in developing a purely skill-based game for that crowd. However, I believe that doing so means you are intentionally targeting a minority audience, as the majority of gamers prefer an element of luck to exist. In the following section, I will analyze a popular game that features variance as part of the gaming experience.

Games With Variance

Multiplayer games with variance tend to all enjoy similar benefits due to their inclusion of randomness, and because of this I won’t be analyzing several games since they tend to all share common themes. The game I will focus on is League of Legends, the game that is currently the most popular eSport in the world and arguably the most popular video game of the past year.

  • The team aspect. Why is it that players stuck in the bottom of bronze league (the lowest rank in the skill system) who can’t seem to ever get better continue to play League of Legends daily? The answer is simple – league isn’t all about skill. The fact that you are on a team means whether you win or lose the game isn’t solely in your hands, but also heavily dependent on 4 other people. This means that even though you are probably at your appropriate skill level, you might win a game you didn’t “deserve” to because you had one or two stronger players on your team. This effect is even stronger in shooter games such as Call of Duty where team sizes are much larger. Overall, the team aspect creates situations where the game becomes exciting because the outcome can’t be foretold or even well predicted.
  • Fog of War + Information Denial. For games without a minimap, this can be classified as hidden information that is very hard to access. While StarCraft also has this, StarCraft suffers from the problem of high predictability, especially in the early parts of the game. When you scout the opponent’s base in StarCraft, you can often make a very accurate prediction of what the opponent’s strategy is and what he/she will be doing for the next few minutes and won’t need to bother getting information again for some time. But what happens when someone leaves their lane in League of Legends? You can’t actively scout the enemy team. That missing enemy character could be on his way to middle lane to 2v1 your mid player. This creates for exciting moments for both teams because everyone knows something is up, and more often than not, you can’t predict what is actually going to happen. This also applies to the jungler role, which is a player who practically starts the game hidden from the opposing team and is a looming threat that can pop out of the jungle at any time to surprise someone in a lane. This dynamic creates excitement and variance that even skill-based players tend to enjoy.
  • Champion Select. I’m not just referring to the fact that the game changes player experiences constantly by having a large character pool to pick from – League of Legends also releases new champions all the time, which prevents the game from becoming dull because innovation is always on the horizon. Besides new champions constantly being made, the champion pool encourages players to pick different characters to play, allowing the player to solve his own problem of the game becoming repetitive (which is inevitable once a player becomes proficient at core game mechanics).

In short, variance makes games exciting. It means that players are willing to keep playing a game even when they know they are terrible at it, but more importantly it means they enjoy playing the game regardless of their skill level. I want to particularly highlight that this sort of game design can trump even the negative aspects of a game. The MOBA genre is renowned for having some of the worst communities, and yet MOBAs are all the craze right now despite douchebag levels being high. While I’m not saying that employing well-designed variance in your game is an excuse to ignore what kind of community your game fosters, I do believe that solid design can greatly help prevent negative parts of your game from bringing your product down.

Reintroducing Variance to Zems Online CCG

To wrap up this post, I want to return to my project. I think it’s important to consider all feedback, whether positive or negative, but to also take both sides with a grain of salt. While most of our positive feedback has praised the no-variance nature of the game, I think the majority of people who offered feedback would actually prefer the game to have variance. People may be overjoyed at the fact that they can’t lose anymore because of bad luck, but over time they will probably realize their skill level in the game has peaked at some point and will begin complaining about how purely skill-based the game is and how it is no longer fun for them. I don’t want this to happen. I hope that my analysis of games above has shown why variance is exciting and can keep ‘low skilled’ players engaged (and there will always be more weaker players than stronger ones). I want Zems to be a game everyone can enjoy, and for these reasons I will be taking steps to bring variance to the game without making it a luck-fest. Specific changes will be covered in detail on the Zems official blog, but the analysis and reasoning stated in this post serve as the main reasoning as to why I am doing this in the first place.