This graph shows the average skill progression (μ) over the number of matches played for players in different skill percentiles: Top 1%, Top 10%, Top 25%, Top 50%, and Below Top 50%. The x-axis represents the match number, and the y-axis shows the average skill level (μ) for each group. Top-ranked players generally have more matches, so their curves extend further to the right.
Players in higher skill brackets (Top 1%, Top 5%, Top 10%) start at a higher skill level and maintain a steeper improvement curve compared to those in lower brackets. Top players not only begin at a higher level but also tend to improve faster.
The length of the light-blue line indicates that 50% of Top 1% players have played 2630 games, while the orange line shows that 50% of Top 5% (excluding Top 1%) players have played 761 games.
You can zoom in on the graphs by selecting an area with your mouse. Double-click to reset the zoom.
For the Bottom 50% of players, there is a noticeable drop in skill level early on.
This graph focuses on the first 300 matches, highlighting early skill development. It shows initial learning curves and the early formation of skill gaps between player segments. This view helps us understand how quickly players in different segments improve in their early matches and how the skill gaps start to emerge.
This graph displays the skill progression for players who started with a skill level of 16 or below after 50 games and have played at least 300 matches. The deciles are calculated based on their first 1000 matches, showcasing the development of skills for those with lower initial skill levels.
After an initial decline, skill levels stabilize and begin to improve around 100 matches played. This trend continues, indicating that players, even those starting with lower skills, will see improvement if they continue playing.