The International 5 – Team Performance Ratings

In the past 24 hours, all regions around the world have completed their qualifiers for The International 6 that will begin, in about one month’s time, on the 8th of August. As we rapidly approach the main event, it is an ideal time to review past tournament performances from the teams and players that participated in The International 5 and this year’s Majors.

Last year’s The International 5 had several novel narratives occurring throughout the entirety of the larger than life tournament. During Summer 2015, when the TI5 prize pool surpassed 10 million US dollars (USD) I remember thinking, “Wow, I didn’t think it would reach this point this year.” Then the prize pool continued to rapidly grow, nearly doubling, and finally ended at US$18,429,613. 18 MILLION USD! This is the largest prize pool for an event in the history of esports.

Another headline from TI5 was CDEC’s grind from the wild card event all the way to the grand finals. It is the most successful wild card team, ever. Their success was such a shock, that none of the analysts or casters had predicted CDEC would even reach the semifinals of the event. For them to reach the Grand Finals, and appearing to be the favorites in the matchup versus Evil Geniuses (CDEC defeated EG in the upper bracket final 2-0 in a best of 3) made it that much sweeter. The demise of Secret was another unexpected storyline. Secret appeared to be one of the more dominant, top-tier, teams coming out of the group stage but ended the tournament with a disappointing 8th place finish.

There were many surprising narratives, but when I started to crunch all of the data surrounding TI5, I was intrigued to see that some of the teams were underwhelming or overwhelming. Some of the analysts and fans had previously made the assumption that it was otherwise. Some continue to make these assumptions.

Before I continue, let’s take a look at the data for the teams that participated in TI5’s main event.

The overall performance ratings for the teams that participated in the main event of The International 5 may appear misleading but that is primarily because of the disparity in skill within the two groups during the group stage. To give you an idea, of the teams that placed in the top 6 in TI5, 5 of the teams came from Group B. Meaning that a team (Virtus.Pro) that went directly into the lower bracket outperformed every team from Group A, except LGD, and finished in 6th place. They even defeated Secret during the main event, and then were annihilated by LGD, which ended their tournament run. Because of this disparity in performance during the main event, choking is real, it allowed for some lower finishers to have buffered stats. This also helps explain the LGD stat line. This information will be explored and explained further when we breakdown individual team performance over time through TI5, in a different article.

When looking at the team ratings from TI5, there are two intriguing narratives that are occurring here. The first is that there is a large variation in the performance produce by teams. There is a rating variation of 27.29 from the performance of MVP.Hot6ix to the performance produced by LGD. A level of variation of this magnitude appearing among the top 16 teams of Dota 2 was not something that I wanted to discover. This shows that there is still development that needs to occur in this game for it to reach a level that allows for a stronger comparison to professional sports. The variation in any traditional sporting league might have a similar circumstance, but there are double or even triple the numbers of teams in the league. TI5 is the equivalent of a playoff where the better teams are the only qualifiers. This performance variation will be further delved into when we look at player performance variation based on a player’s role. The other narrative that is promising and positive is the clustering that appears in the ratings. There are a couple of clusters where teams put out similar stat lines and then there is a jump in ratings to the next cluster. Also there is a very large cluster in the middle of the ratings. This is a great sign because it shows that there are multiple levels of competitiveness within the tournament. If we look at sports like basketball, when the NBA enters the playoffs there is a larger difference in performance from the lower ranked playoff teams and the top ranked teams. This variation in team skill is what causes the quick early series and things slow as they get further into the playoffs. It is similar, for the most part, for Dota 2. This is something that is great to see in an esport because it allows confirmation that things are going in the right direction. There is progression.

The team that took home the Aegis of the Immortal, Evil Geniuses, had a very impressive stat line. If you look at their Last Hits Per Minute (LHPM) and Experience Per Minute (XPM) they ranked 1st. In Gold Per Minute (GPM), EG was ranked 3rd. In regards to the Kill to Death Ratio (KDR) and Assists to Death Ratio (ADR), EG places 4th. EG placed 12th in Hero Damage Per Minute (HDPM). The fact that the winning team wasn’t handily the top of each stat line is also promising and shows that there is competition.

The International 5 is a very positive sign of more great things on the horizon for the future of esports. The rate in which major tournaments and infrastructure have improved in esports is an impressive feat that is rarely lauded for the great achievement that it is. It’s an exciting time as we draw closer to The International 6 and all the drama that comes with it.