Guide
Hey everyone! It’s Highnoon, the one responsible for the “magic metrics” and other mystical technologies that often remain under a veil of mystery and mostly hidden from players view. In this Q&A, I’ve tried my best to respond to the most interesting questions you asked and demonstrated a number of metrics that have never been seen anywhere but our company backstage. Unfortunately, there are a number of questions left without an answer since some metrics and figures must be secured from the players’ gaze due to objective reasons…
Hope you will find it intriguing!
When it comes to assessing gameplay quality, we don’t really have any key parameter to evaluate player behaviour or pattern. That is because each gameplay aspect is important in its own way.
On one hand, we closely spectate how players pass the main storyline. We follow what dungeons and bosses are too difficult or too long to walkthrough, and it typically goes against the fundamental ideas of game designers regarding them. If we detect an excessive number of attempts to kill a boss (or too much time to kill the boss), it goes straight into the works of the Game Designers and is fixed with (or sometimes without) the nearest maintenance. Apart from the main quest, we also check up on how deep are players involved with completing side quests.
On the other hand, we constantly monitor the high-end players and their preferences on armour, weapons, the velocity of grinding out a specific recourse, players faction balance during outposts fights, and estimate the engagement in each activity on the map relative to the total volume of the selected activity type. This is just a small part of what we keep an eye on to “maintain the quality of the game” to the already existing content. This is also in turn monitored by the Game Design team. Another aspect is a health audit of the project. We monitor the speed of movement of players from one location to another, identify the most difficult and frustrating quests and dungeons, searching for activities that lead to ragequits. This kind of audit, for example, has led to the current development of the Swamps and Roadside rebuild.
Some metrics are regularly monitored without the involvement of analysts by the Economists, Balancers, Level Designers, etc., but some metrics are monitored primarily by us (even though there are few of them left). We try to transfer our expertise as much as possible to the whole Team so that they can independently and regularly monitor the correctness of their mechanics and features.
In general, we monitor everything! Our basic principles are:
We see something abnormal, we transfer it to the Game Designers for further review and fixes.If we see something different from the target figures (if there are any), we also transfer it to the Game Designers for further fixes.
And of course there is genuine interest, we regularly try to find out new things and evaluate the user experience in a new way to make the game more attractive and interesting for more groups of players.
Easy victories are frequently easy to forget, but for you, as a Covenant member, victories can often be a bit harder to achieve. Here is a picture of the outcome of the outpost battles on NA at Inflow:
It can be seen that the majority of Outposts do get claimed by Frontier members, with a ratio of about 40% for the Covenant and 60% for the Frontier. This problem arises due to a slight number advantage to the Covenant, which allows this faction to take the Outposts a bit more typically due to their population. Don’t fret throwing tomatoes at us! Within a couple of days, this issue will start to disappear.
Here are the same graphics on EU:
45% for the Covenant, 55% for Frontier.
Let’s also have a look at Mines, EU region:
45% for the Mercenaries, 55% for Rise
And the NA:
Mercs get more comfortable than Risers.
Does this data prove that one faction has a higher chance of capturing an outpost on average than another? Definitely yes. Does it increase the chance of the dominant faction capturing an outpost? Also yes, but we have to point out how much. You'll only feel the fundamental difference over a very long distance if you play outposts regularly. In a random single skirmish, the outcome can still be anything. However, measures have been taken, the situation should stabilize soon!
Despite the Analytics Team not taking part in balancing out weapons, we know the process well. In practice, weapons are not “balanced based on statistics from Session Battles” they are balanced by performance on different locations, against different armour, at different ranges and with different upgrade level. Thus, weapons are balanced both on “global data”, i.e. on the whole zone, and on specific contexts of their usage. In this case, the more narrow the context becomes, the less reliable data we obtain, because the parameter selection becomes smaller, which is well aware of the balance department.
Balancing items separately for each region makes no sense, because:
-
Player launch the same game and play in the same conditions. Only the players differ from one region to another. There is no reason to create different game conditions on different regions, that will only confuse both players and developers.
-
If balancing is done separately for each region, it will be difficult to reliably find the context of weapon usage, because on less densely populated servers the selection is already small.
That’s a good suggestion! Although there are reasons to it as well.
Firstly, the game’s visual language is essential for player's better perception and immersion. The player should clearly understand that AKS-74U is weaker than FAMAS at least on gameplay level, and at the most it should “resemble reality” at least a bit by TTK of real weapons. You may assume the example I will provide you now is exaggerated, but still: if a player with a Shaman's Boomstick jumps out at you and headshots you when you were in Scythian or Rigel, it will feel at least strange, and at most unrealistic. The player will stop understanding what behaviour he should expect from each weapon. Just like in Skyrim it would be very strange if an Iron Mace hit as hard as a Daedric Mace.
According to the latest data, about 12% of players successfully reach the North. However, we are actively working to increase this number by improving the game experience on Southern locations. We aim to find and solve those problems that cause boredom and frustration for the player, while not seeking to take away the actual challenge from them.
In fact, we often think about making an open service with basic metrics for players. That applies not only to the number of active players in factions. However, so far we have not come to a consensus on this issue, although I personally think that some figures would be curious and a useful for the player to see. If the majority would really be interested in this, and you show it with your activity, maybe the consensus will start to be found faster! 🙂
I would also point out that a lot of information services have already been implemented thanks to enthusiastic players, StalcraftHQ API is a good example of it.
I won't give specific numbers, but the situation is similar in all regions: Hounds is the most popular armour, next come the super heavy armour. Then there are all the others. Hounds is popular due to its unique speed characteristic, super heavy is an ultimatum solution for session battles and CW, and the rest of the armour seems to be too situational, making it bartered as 3–4 armours per account. However, there are examples of appreciative players who only play Vanguard, Punisher or Peresvet.
The number of players in factions doesn't sound that intriguing, especially considering that the player count doesn't reflect the actual number of players playing, as well as there are boundaries to the number of hours played and matchmaking.
But I can share some interesting metrics with you (wherever you see the letter q or quant in the name, it's quantile 🙂):Number of attempts to beat the Guard boss:
Number of attempts to beat the Vagant boss:
Number of kills by one Murmur player of other players in Black Raid:
The most popular containers and backpacks at master rank (hello to grinders with a Soviet backpack 🙂):
Effectiveness of beginners' farming of the Swamp Stone resource (units per hour):
That's only 1% of what we have! There's a lot more interesting but often more complicated stuff out there. Ask questions — I'll show you more interesting stuff in the future 🙂
Sure!
Red Forest:
Yanov:
As for displaying chances, I'm afraid that's a bit out of my area of responsibility. However, as for pity-system and similar, we are actively thinking about it. We even have an overview of the mechanism of how it will work and what cases to integrate it into. Do not consider this an announcement! It's more like thinking out loud…
As for basic event analytics, we know literally everything about the event. From how many seconds the game is found in, how many zombies the player killed, what they bought at the vending machine, and what crates they opened and what dropped out of them. However, detecting illegitimate co-operation is a little more difficult, as it's not a specific event, and we try to find it through mediated complex calculations. I won't go into details, but just to say — many illegitimate co-operators will be surprised in the near future, and the Baron's Guard may change a lot…
I'm afraid that is not at all true. Moreover, even though I am an analyst, I often favour “asking the players” more than anything else.
Player opinion is an extremely valuable resource for tracing out problems that cannot be measured in numbers. Many player complaints about difficulty in passing bosses, issues in certain locations or activities highlight ways for us to find issues, which we later confirm by analysing the data. Let me tell you more. A number of key decisions during the release of the major update in the summer were made based on feedback from ordinary players who participated in early testing.
I'm also actively promoting the idea of player surveys to identify pains, issues, and bottlenecks in our favourite game. It is still only in the concept stage, but even now, player opinion is not the enemy of metrics or the developer. This is our roadmap in improving the game. But don't take it too literally, any opinion is validated with data from the game whenever possible, ha-ha! 🙂
That's all for now! And remember — this is not the last time I prepare exciting graphs and facts for you. Feel free to ask questions, and maybe next time you will see exactly the graph you asked for. Especially now when you realise approximately what kind of graphs I can show you 🙂
And remember — metrics don't lie. Only the person who counts and interprets them can lie!