Game Balancing: Data-driven Approach
Game design: why the details are important
Igor Klyukin, COO Pixonic
I am entirely convinced that a lot of people in the industry are trying to find a “silver bullet” in the form of advanced analytics, marketing secrets, or “ingenious gameplay”, forgetting that success depends on a product’s quality. But quality is in the details, something that falls under the responsibility of front line developers.
By building a bridge between the success and performance of a particular specialist, I want to continue this discussion on the intricacies and details of game design. Good ideas are often ruined as they are carried out, or the opposite happens and poor ideas show their true fascinating colors when developed by a professional.
Let’s consider an example:
The chief designer (who is often a producer as well) gives a junior designer a task (who is often the only designer of the project), “We’re going to hold an event in our MMORPG, where crystal slivers have to be dropped from mobs, and 125 slivers can be traded in for a helmet. We need to make it in such a way that a player spends about 5 hours farming slivers in order to get the helmet.”
What will the game designer do?
According to the statistics and our own measurements, we know that a player kills 30 mobs every hour. In 5 hours a player can kill 150 mobs, so we must then divide 125 by 150, which gives us 0.83 sliver per mob (125/150 = 0.83). The game designer then sets the probability of 1 sliver dropping at 0.83.
So the task has been completed, hasn’t it?
Mathematically speaking, this is true. But a good designer would have gone about it in a different way. To begin with, he would have analyzed the resulting situation:
- A drop probability of 0.83 means that in most cases (83%) a sliver falls from the mob, and sometimes (17%) it does not.
- From the player’s point of view, the most common outcome of the situation is the “norm”, while the rarer case seems to be almost a failure when compared to the “norm”. So, this means that the gamer farms, but is sometimes unlucky.
- Every time this unluckiness strikes, he experiences negative emotions that are a completely unnecessary punishment.
What can be done about that?
1. To begin with, a good designer would say, “We should change the task conditions in such a way that during the event a player always receives an award, so we’ll drop 250 slivers instead of 125 and we’ll charge 250 slivers for the helmet in order to maintain the balance”.
The result is that now players always get a reward, as we have eliminated the risk of “failure”.
2. Now it turns out that the number of slivers to be dropped from one monster has to be 250/150 = 1.67. That is, on average, more than one. This is no longer a problem, but rather an opportunity. There is an opportunity to turn crystal farming into a more interesting pastime than it would have been if this task were handled by a typical game designer. A good designer would say, “let’s drop different numbers of crystals: 1, 3, 15 or even 250 at once! I’ll calculate the probabilities, so that on average there would be 1.67 slivers from every mob”. Note that the drops are distributed in such a way that the situation has shifted from “losing is the norm” to “winning is the norm”. The player will “win” an average of 3 slivers every 12 minutes, remaining “fit and 3 or 4 times during the farming session they will have windfalls of 15 slivers at a time.
3. In addition to enriching the task with positive emotions, the designer also makes it more difficult for the players to count the time they need to farm for the helmet because now they are collecting 1, 3, or perhaps 15 slivers at a time. Someone will pick up the required number faster, while others will take longer, but the gaming community will be less likely to blame developers for unnecessarily lengthy farming.
4. Moreover, a good designer introduces the possibility of dropping 250 slivers at once, but very few gamers on the server will experience a drop of this magnitude. These lucky players will generate buzz and people will start talking amongst themselves like, “I know one guy, and his brother’s friend knows someone who had 250 slivers dropped at once!” Every gamer who hears about it is going to start hoping they will win the jackpot too. Everyone loves having a chance to win the lottery, and they will get their hopes up that they could be the lucky ones. And even if farming slivers still seems like a long time to them, they will hear the voice of hope whispering, “Kill another monster, maybe there are 250 crystals in it…”.
The overall point of the story is that this is not rocket science: it is an example of the crucial role that the details of actual game play should have during game design. This is something that deserves more attention or at least a hunt for professionals who know how to do it in the first place.
Sign in to comment
War Robots: 3 year anniversaryApril 14
All of last year's highlights brought together in our infographic
Three years on: Pixonic on the evolution of War RobotsApril 6
On the eve of War Robots' 3-year anniversary, PocketGamer sat down with Pixonic COO to talk about the project's current progress