The International 2019

Enough time has passed ever since the main tournament of the year has concluded and new competitive season is about to begin. OG has broken yet another curse and proved their championship to be more than “just a fluke”, while all the new tournament trends completely destroyed all the expectations and predictions. It’s time to look back once again and look at the tournament’s numbers.

Meta

While most of TI9 trends came directly from the ranked meta (and overall predictions based on teams playstyle proved to be on point), I will be wrong if I say that the meta was normal and predictable.

• Total matches played: 193 (144 – group stage)
• Radiant winrate: 46.11%
• Median match duration: 39:13
• Uncontested heroes: 0
• Unpicked heroes: 3 (Keeper of the Light, Ursa, Clockwerk)

Ever since the 7.22 patch one of the most expected trends was playing heroes in an unconventional way, changing positions and increasing pick variativity. This expectation turned out to be a reality. Firstly, gap in popularity between the most popular heroes and median ones turned out to be way smaller, just like the pool of the most popular heroes. Secondly, there were a lot of surprises in terms of new playstyles for old heroes, like Carry Io or Support Invoker.

But there were two major factors that decided the success of TI9: “solved” meta (globally the game remained just the same as a year before) and meta that favored timings that is comfortable for a lot of heroes. Three hero types that don’t feel so good in this meta are the ones, who need to farm a lot, tempo heroes and, lastly, heroes unable to kill Roshan efficiently. While there are a lot of heroes who hit one of those categories, it’s hard to design a good lineup that will be weak in at least on of these ways since most of the heroes are either on the “fine” and “solid” territory or even able to close weak spots of their team by themselves.

This also led to rising of another interesting trend: heroes, who are hard to mess with early on, turned out to be very successful. Examples of this kind of heroes may be Alchemist, Arc Warden, Wraith King and Medusa. While the latter seems kind of weak because of being slow and unable to contribute something while killing Roshan, the whole idea still remains the same. The most common success story was based around one simple strategy: one hero, who is strong at farming and hard to shut down early, keeps farming for his whole team while another core hero (who is usually pressured a lot) doesn’t need a lot to start playing active, and the latter three heroes are capable to force a fight by themselves.

There are two heroes who deserved a special mention: Alchemist and Enchantress. They are not only ones of the most contested heroes, but also great pictures of the meta trends. Alchemist isn’t necessarily strong hero on lane, but he needs just a couple of levels to start efficiently farm early on. This is also the case when you can’t just ignore and let a hero go at any stage of the game or any advantage you had may be lost in an instant, while it will be hard to deal with the hero going into later stages of the game. In addition, the hero is great at killing Roshan and is able to kill team oriented expensive items (like Assault Cuirass) relatively early, effectively enabling his whole team and paying off his farm. Wraith King is pretty much identical to Alchemist in that regard and Arc Warden is a bit slower in the beginning.

Enchantress depicts another trend: the relevancy of strong laners who are capable of pressuring opponents and delaying his power spikes as much as possible. Another hero who’s great at accomplishing the same thing is Ogre Magi.

Heroes

Best heroes (based on rank)

First draft stage

Picks | Bans

Bans

Hero combos

Positions

Players and teams

• Most matches played: Team Liquid (32)
• Team with the highest winrate: OG (82.14%)
• Team with the widest hero pool: Team Liquid (62)
• Team with the smallest hero pool: Mineski (37)
• Players with the widest hero pools: Secret.MidOne and Liquid.Miracle- (19)
• Highest hero diversity: Secret.MidOne with 19 unique heroes in 28 matches

It’s hard to note someting in particular by just looking at stats here. One interesting trend, however, was lowering average stats of players of the most successful teams. The reason behind this is most likely meta that forces players to fight a lot and give up on farming as much.

It’s also worth to note the performance of OG, who are sometimes called “the new Wings”. The main difference of the team’s playstyle and, most likely, what helped them to beat Alchemist meta and win the tournament is flexible laning stage and mimicking lategame moves during early to mid game. Later into the tournament there was another team who started using this playstyle – Team Secret.

The highest average stats of players

Teams summary

Match records

• Match with the most kills combined (95) and the most Roshan kills (6) – Na`Vi vs EG, game 2
• Match with the most networth stomp (50,430) – Fnatic vs VG, game 2
• Match with the most networth comeback (23,107) – KEEN vs TNC, game 2
• The longest match (1:26:28) – TNC vs Chaos, game 2

• The shortest match (15:07) – RNG vs Infamous, game 1

• Highest GPM – 1,176 by NiP.Ace as Alchemist in a match Fnatic vs NiP, game 1
• Highest XPM – 992 by Infamous.Chris Luck as Broodmother in a match Infamous vs NiP, game 1
• Most Kills – 27 by EG.SumaiL as Storm Spirit in a match Na`Vi vs EG, game 2
• Most Deaths – 17 by Newbee.pieliedie as Crystal Maiden in a match PSG.LGD vs Newbee, game 2
• Most Assists – 37 by EG.s4 as Tidehunter in a match Na`Vi vs EG, game 2
• KDA with 0 deaths – 34 by OG.ana as Spectre in a match OG vs RNG, game 1
• KDA with 1+ deaths – 35 by EG.SumaiL as Leshrac in a match Secret vs EG, game 1
• Most Gold Earned – 77,056 by CHAOS.MATUMBAMAN as Leshrac in a match TNC vs Chaos, game 2
• Most Last Hits – 1,130 by Na`Vi.MagicaL as Templar Assassin in a match Na`Vi vs EG, game 2
• Most Hero Damage – 128,438 by TNC.Gabbi as Arc Warden in a match TNC vs Chaos, game 2
• Most Tower Damage – 23,138 by Alliance.mICKe as Morphling in a match Newbee vs Alliance, game 1
• Most Hero Healing – 48,170 by TNC.eyyou as Chen in a match Liquid vs TNC, game 1
• Most Hero Damage Taken – 136,986 by TNC.Gabbi as Arc Warden in a match TNC vs Chaos, game 2
• Most Observer Wards Placed – 34 by TNC.Tims as Grimstroke in a match TNC vs Chaos, game 2
• Most Sentry Wards Placed – 68 by VG.Dy as Witch Doctor in a match VG vs TNC, game 1
• Most Observer Wards Destroyed – 13 by Mineski.ninjaboogie as Ogre Magi in a match Na`Vi vs Mineski
• Most Map Pings – 611 by Fnatic.iceiceice as Sand King in a match Fnatic vs NiP, game 1
• Most Stuns – 361.07 seconds by NiP.Saksa as Nyx Assassin in a match Fnatic vs NiP, game 1

What’s next?

The story of TI9 was in many ways a continuation of TI8. The same bracket, participants, champions – it was almost like it was set to be a revanche. A lot of it was decided by the meta: there wasn’t a big gameplay update (not counting 7.22) ever since TI8 so basic gameplay principles and ideas remained untouched. However, it may not be the case anymore with the upcoming Outlanders update.

It doesn’t seem like the update is coming anytime soon since we already got a minor gameplay update and there are new tournaments, qualifiers, major and minor tournaments to be held soon. Most likely, the update is coming sometime in November, and we can even expect to get another minor patch before that.

However, there is another factor that is not going to be a thing anymore – teams rosters. We may expect some huge reshuffles and roster changes in the biggest teams which may make the competitive season completely wild and unpredictable.