The International 2018: An Incredible Finale

The International 8 concluded this weekend in thrilling fashion. Not since TI 3's Alliance vs Navi's 3-2 Grand Final has there been such excitement. OG, a team whose two initials have no meaning, OG, a team reformed with just three members right before the TI8 open qualifiers, a team with no events played, a team containing a player who had played in no prior event, vanquished every foe, and claimed the aegis in an astonishing, outrageous, absurd best of 5. How did it happen? It still seems so incredible.

Before the final, and OG's performance, the meta. The group stage contained many 1-1 splits of 2 game series, yet the Main Event was surprisingly one sided. Of the 17 best of 3s, 13 were decided in 2 games, with only 4 needing the full 3. And with 47 games played on the Main Stage, and a total of 470 picks, of 116 heroes, 81 were picked. A few in particular distinguished themselves.

Drow, with her global aura to grant bonus damage to allied ranged heroes, went 7-2 on the Main Stage, and 20 – 9 over the entire event. She enabled other heroes such as Weaver, Vengeful Spirit, Lina, Mirana, Invoker, and Clinkz to push earlier and fight better with the bonus damage. Invoker performed solo as well, with a 5-2 and 16-8 record.

Some heroes who performed only moderately (or even poorly) in the group stage such as Clinkz (17-23), Spectre (12-15), and Vengeful Spirit (35-34), excelled in the Main Stage respectively, 8-3, 7-1, and 13-7. Vengeful Spirit, was the most picked hero with 89 games overall, and teams selected Tiny the second most, even though he had only achieved a 32-46 win rate. A supremely strong laner, he didn't deliver as most games developed.

After Vengeful Spirit, Phoenix and Earthshaker were the most successful supports, securing a winning lane for their cores early, and controlling the battlefield late.
Finally, there was Broodmother. The hero has often been called a cheese pick. In Dota, a cheese pick is when a team selects one of a few specific heroes (or two hero combinations) late in the drafting stage, which are nearly invincible. Some examples of picks considered cheese are Huskar and Dazzle combo, the old Techies and Tiny, Drow and Visage, or a late picked Broodmother. It always seemed to me that players and fans attempts to label certain strategies cheese, were merely complaining, and many picks formerly considered cheesing, are now common strategies, such as the ability to kill Roshan at level one, or rat effectively. But even if the term remains, Broodmother can no longer be consdiered a cheese draft. With a 4-0 on the main stage, 12-6 overall, and victories after a first pick in the draft, she is a dominating, but fair force for those who know how to use her, like Matumbaman and RAMZES666.

But there's a meta beyond the heroes. Some commentators said, teams preferred the Radiant side in the group stage, but on the Main Stage it didn't seem to matter, and of the 10 games played in the Grand Final, Upper Bracket Final, and Lower Bracket Final, Dire won 6. Others claimed that the team with the last pick were more likely to win, but last picking team won only 6 of 10 final games, nothing conclusive. The biggest conclusion was, win the lanes, win the game. This opinion has been in vogue for quite a while, since the change to denying creeps in 7.07. Yet, OG proved this conclusion incorrect, as I'll demonstrate shortly.

A brief overview of the Grand Final. In game 1, OG fell behind 8k gold at 22 minutes, and didn't gain the lead until the final 5 minutes of the match, from 29 – 34. OG victory!

Both games 2 and 3 were stomps by LGD, where OG was never ahead. LGD won both.

Game 4, OG managed to secure the lanes, and lead for the 35 minutes, peaking at an 8k gold advantage, before falling behind by 15k gold and losing two lanes of barracks, before finally recovering at the 54 minute mark, and concluding the game at 1 hour and 5 minutes.

And in game 5, OG slipped behind by 7k gold until the 25th minute, and then won at 36 minutes.

OG only won the laning stage in one of the five games, had to make a reasonable comeback in two of their victories, and an incredible recovery in the 4th game, even though they won the laning stage in that game. Each game OG won was a crazy experience, and if you want to watch, here are the highlights of games 1,2, and 3 and of 4 and 5.
Before concluding, let's look at the complete outcome. Commentators kept claiming that TI8's results was entirely unpredictable, and that everyone's brackets were ruined after the first day. Mine certainly was. There were many difficult decisions, like VP vs LGD, EG v Secret, Secret v Vici, Storm v Secret, and EG v VP.

But in fact, there weren't many upsetting results, only 3; Serenity beating Fnatic, Winstrike beating Newbee, and every single OG series. They were unpredictable, they were clever, and they were often seconds away from defeat. Of the four best of threes (of eighteen total series) which required the complete set of games, OG was in two of them. OG was only average except when they were brilliant, and fortunately they were brilliant and tenacious enough to repeatedly claw themselves to victory from serious deficits.

Even ODPixel can barely believe the result as he declares OG champions. Yet they deserved it, and few would say there are players more worthy (and likable) than Notail and his crew.

After the group stage, and OG's average success, my system thought they were a terrible choice for TI champions. But, finally, with OG's victory, my system believes thinks they would have a reasonable chance at a repeat. Based on the chart below, my predictions after the Group Stage were reasonably precient, as my top 4 were all in the top 5. Only OG surprised, because there wasn't enough data, and they're performance in the group stage was only average.
A few final thoughts. My MVP of OG, both in the upper bracket final and the grand final was Jerax with amazing plays on Earthshaker, Io, and Rubik. Game saving in the 3rd game of the Winner's Final.

A nice surprise of Valve inviting Merlini to TI8 to tricast the Grand Finals.

The yearly rotation is broken. In the seven prior Internationals, a Chinese team had won every one in an even year. It's over. Yet, though so many came so close, no player or organization has managed to win The International twice, yet.

Two new heroes are scheduled for release, one immediately, and the other in the winter.

I was a disappointed the third Open AI vs Pro Gamers was canceled, even though Open AI had already lost twice. There was even a post on Reddit, which I can no longer find, how all the members of the original TI 1 winning Navi Squad were in Vancouver. It would devastating to learn they were scheduled to play only for it to be canceled.

And that concludes the 2017/2018 Defense of the Ancients 2 season. Don't worry, the qualifiers for the first Minor and Major begin September 17th and 23rd respectively, with the main events scheduled for late October and early November. I'll be keeping track of them, but there won't be another Dota article until the end of January 2019.

Recent:

Relevant:

Comments