![]() I agree that they have not communicated well about a majority of things, however you need to stop and think about the audience they would be speaking to and how that audience has reacted to them being communicative in the past. If Riot has compiled research across the board about the changes and found that they generally liked the effects that it has had on player experience, then there shouldn't necessarily be a huge burden for them to make sweeping changes to correct for queue times. ![]() Secondly, queue times are a byproduct of the generally enormous changes to champion select for this season. Rewarding players who play unpopular positions has always been a central goal of their's, so worse queue times for 75% of players and significantly shorter ones for the other 25% isn't innately a problem that requires fixing this requires more research and needs quantitative analysis. ![]() I'm not disagreeing with your points, because the factual ones are indisputably true, however I feel like you're making some significant assumptions and generalizations here.įirstly, even if Riot is being somewhat illusive about queue times, the conclusion that you have reached that queue times are worse for a majority of players doesn't necessarily mean that Riot has failed to accomplish their goal. I think Riot have done a poor job in user research, a poor job in communication, and have been extremely evasive. I suspect what's going on here is that the median queue is shorter but the 75th or 90th percentile queue is much longer – and you remember the outliers, so the player experience is much worse and much more frustrating. Evidence suggests those are much, much longer than they used to be. The meaningful numbers they should be looking at are the median and the extrema – 95th and 99th percentile queue time. So Riot haven't quite done this, I believe they've implied it and talked around it, which is shady, but they've not actually just fed people deliberately misleading data. So we've established that that's either incompetence (you failed high-school statistics) or deliberate deceit, both of which are firing offenses. It doesn't tell you anything useful, and the only people who don't know this are the people who you have a professional duty not to mislead. So using a mean as a measure of central tendency is actively misleading to the point of professional negligence. We can take it as read that whoever gets to work on this kind of problem has that basic level of competence.Ī time interval can never be less than zero therefore the distribution of time-to-game is going to be very, very skewed (there'll be a lot of values around zero and a long tail going off to infinity). ![]() Part of basic statistical literacy is having very basic intuition about distributions of data. ![]() So, how is it lying to players? The person looking at this will be, in title, probably a data scientist. If you deliberately mislead customers, you shouldn't have your job any more. Let me explain quoting mean queue time is lying to customers. That's a big statement, but I'll back it up. (I think Riot have avoided the worst possible misleading statements, but have been shading the truth really, really hard the way they've behaved has been really distasteful and professionally offensive, honestly.) If Riot are looking at means here – and to be fair, they've been careful avoiding that precise language, as far as I've noticed – then the employee responsible should be fired, either for incompetence or for deceit. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |