Player Engagement Data

Hi All,

I have recently seen a lot of videos talking about the Path of Exile 2 analytics and the 1 week snapshot. One of the primary topics covered in the screenshot post by Lilylicious (https://www.gamepressure.com/newsroom/poe-2-devs-revealed-most-popular-early-access-classes-version-020/z27a28). I am not critiquing or criticizing the aggregate metrics or the data that was published. I am certain that they know what they are doing. However, some of the content creators who see this post have been talking about the Average player level and the median player level that people are getting to in the one week snapshot. Average player level is 75 and median player level is 89. Both average and median are metrics that seek to tell you what most players are able to achieve. In theory, both of these metrics are one way of measuring the 50% or halfway mark of a population. These measures are useful to tell you an estimate or summary of a larger data set. Both of these measures are more useful when you also know the variance, standard deviation, low, high, or interquartile range because those metrics tell you how spread out the data or population trends are. In this case our low is probably above 1 and our high is less than or equal to 100 (player levels start at 1 and cap at 100). We have already seen that there is a difference in the mean (average) and the median. Both of these metrics seek to show what the 50% mark of the player base is. in an ideal world, these metrics would be the same. When these measures are different than one another, then we actually are given more information about the total data set. A lot of the people see these metrics and think "oh, this is a great demonstration that players like the game because 50% of all players are 1) getting through the campaign and 2) are playing a decent chuink of the end game". I don't actually think this is entirely corect. I would agree with it if we were only seeing the median. However, levels are weighted, which means that is is easier to get to a lower level than it is to get to a higher level. Except, leveling actually gets easier after the campagin because you have dropped currency, figured out your build, and geared up enough to kind of know what you are doing. So even though levels are weighted, we would actually expect players levels to be greater than level 70 because that is the point where your character actually comes online (I think I finished the campagin at level 68 - 71). I don't agree that these metrics are saying that most players are finishing the campaign and playing the end game. @ least 50% of all players in the snapshot period are doing that, which is great for the upper 50% of players. However, for the lower 50% of players, which are probably the noobs, I do not think they are finishing the campaign or they are finishing and then quitting the game. I am not saying this is the entirety of the lower 50%. I guess I also do not know what percentage of the player base truly needs to continue playing for the game to be a success. My point is that when the average and median are not equal, we learn more about the dataset. When the average is lower than the median, we actually learn that the data set is left skewed, which means that there are a hefty number of players at very low levels that are dramatically pulling down the average from the benchmarked 50% point of the data (the median) (this is recursion towards the null, null be 1 in this case). I do not think as many players are completing the campaign as you might expect when seeing that the median player level is 89 and average level is 75. To verify, you would need the full data set to know more, but I would be skeptical of how positive a result this is. On the flipside, at least 50% of all players are pressing very far into the end game. A majority of players (50% or more) are playing a lot and seem to be enjoying the game. Also the 1 -week snapshot is far enough into early access that those who are still playing are probably going to continue playing based on the hype around a new release having warn off. I still would not be surprised if player numbers are about to drop if they have not dropped already because the 14 level difference in average and median just doesn't sit right me. It just seems weird that one metric is saying that the upper 50% is getting far and the lower 50% are maybe getting a taste of maps and than stopping their level progression. It is just weird to me.

Honorable mention, please release the templar and please add in the imbalanced guard ascendancy notable on the warrior side of the tree.