you can view individual games by hovering over the point on the line. As for your idea of averages over more specific time frames, I agree with this. Total average doesn't mean much.
what if you were complete garbage 3 years ago, and were even complete garbage 3 months ago, an then you have a Rocky Balboa moment and become awesome. Well now your gpm is 4x higher but your average is still basically the same. Obviously exaggerated but that's the problem with this current super simple averaging.
Lütfen yorum yapmak için giriş yapınız
When looking at any of your trend graphs for Win rate, KDA Ratio, GPM, LH, DN, it looks good to begin with, until you realize the graph spans a change within <0.5% of total value.
If you change the time to "All Time" you realize whats wrong. The graph is right now showing you your average value, instead of the actual value. That means some change to start out with, then just a leveling of the value to a constant line, if you have many games played. Sure it's nice to see this heading slightly in the right direction, but it doesn't really show a good graphical representation of your actual values for different periods of time.
Suggestion:
Add an option to show the actual value from each game in these graphs, preferably an option to choose "individual games or daily/weekly/monthly averages" too to give a more even curve. This would give a much better impression of what's been happening, like "oh that month that I focused on support I can notice an increase in win rate" or "I've been randoming a lot the past week and that really shows"
Also, why no subforums on dotabuff? It's quite an active forum and would be cool to split off some general/strategy/suggestions/offtopic atleast