The mid-lane outer turret is often seen as the most important structure in the early game: it makes deep invades more difficult, limits pathing to some extent, and provides a retreat when teams lose objectives. A few days ago, we introduced a stat called First Mid Tower Rate (FMTR).
What can the First Mid Tower Rate tell us?
FMTR shows the percentage of games a team has gotten the mid-lane outer turret first. This means that the other team either didn’t get the mid-lane outer turret or got it later on in the game. Graph 1 compares a team’s win rate with their first mid tower rate. The trend line depicts the win rate for a specific first mid tower rate across all games played in the EU & NA LCS, LMS, and LCK. As you can see, most of the top teams either have a FMTR above 50% or seem to be outliers like the ROX Tigers (a weird team in terms of early game stats and win rate). The top teams in all regions have significantly higher win rates than their FMTR, already hinting at the stats limitations. On average, our FMTR seems to offer decent information about how teams win games. However, it is susceptible to outliers because of the limited amount of data we have from this split.
I used data from all LMS, LCK, EU LCS, and NA LCS games that were played before Monday (July 4th). Given the limited sample size, I will not go into details on the LMS, but wanted to include it for those who are interested.
If you are interested, you can access the graphic below as Tableau visualization and filter it according to regions. Unfortunately Tableau does not format well for smartphones. Please keep in mind that the trend line adjusts according to all regions currently selected in the viz.
When do teams take the outer mid turret?
The boxplot below gives you an idea about the distribution of the average time at which teams take the enemy mid tower. Notice that this is not necessarily the first outer mid tower anymore. Most teams drop the enemy outer mid after 18 minutes, already well on their way into the mid game. Teams in the North American LCS take down mid towers the quickest. Immortals (17:48) lead the way, but are an extreme outlier.1 The EU LCS seems to be the most balanced region in terms of time distribution. CJ ENTUS actually is an extreme outlier in the LCK. It comes as no surprise that teams with times above the upper quartile of the boxplot are also the worst performing teams in every region. Teams with average mid tower takedown times below the median seem to be the ones winning games (unless you are Unicorns of Love, then you obviously do not care about stats at all).
We now have a general idea of FMTR and the average time teams take mid outer turrets. In the last table, I have broken down the win rates for each team when they get the first mid tower and when they fail to do so.
Most teams know how to go from taking mid outer first to closing out the game. It actually gets more interesting when we look at last two columns. Even most top teams struggle winning games when they get the outer mid after their opponents. Notable exceptions are Team SoloMid, G2 and again the ROX Tigers. TSM win 75% of games even if they took the mid outer turret last, G2 win 56%, and ROX win 50%. However, the Tigers did so in 14 games. TSM and G2 only had to do so in a 8 or 9 games each.
In conclusion, the first mid tower rating seems a decent stat in terms of a team’s ability to close out games. If we want to say more about the stat in itself, we would have to see whether top teams sometimes trade it for specific drakes. Another important factor might be the time distance between both outer mids falling.
If you have any questions, comments or critique, you can voice them in the comments, our contact form or write me at flo[at]league-analytics.com. If you like what we do here and want to get updates on new articles, stats, and more, follow us on twitter.
1 Correction: An earlier version of this article stated that Team Liquid took mid towers at 17:36 minutes on average. They actually take it around 20:28 minutes. The boxplot graph has been updated to incorporate the correction.