Watching The International is very entertaining, regardless of how well you know the game. There are always big highlights, fight-turning ultimates and split-second outplays to be amazed by. But what if we said that the fight begins much earlier? That a big part of the battle happens during the drafting stage and it is when some matches are decided?
Most contemporary meta-games revolve around several heroes that stand out and dictate the overall approach to the draft. The balance is never ideal, so there are heroes who are slightly stronger than the others and in a highly competitive environment even a small edge can make a big difference.
These are usually the trend-setting heroes for the whole tournament, also known as “heroes to beat”. They are heavily prioritised during the drafting stage and are frequently what give first pick its competitive edge.
In many cases, teams that are well-prepared for the tournament will know how to play such heroes, while also having several strategies prepared against them. Also different teams might have a different read on what heroes should be prioritised, but usually only some of them are correct in their assessment.
Dota heroes are still very unique, with a set of highly pronounced strengths and weaknesses, so even the strongest of heroes will have a counter-play and even the weakest heroes have some good matchups. Identifying and even forcing a bad game for a strong hero and a good game for a weak one is one of the most important skills when it comes to drafting.
Given how “strong” heroes can still be countered, they can be a bit of a trap if picked in the first phase. Hence the current meta of semi-flex picks being high priority.
We feel like we are still in the realm of meta-meta-gaming, just one level lower: we are still discussing how teams have been approaching various meta games in general. And we feel in this meta-meta-game the most important concept is flex-picking in the first phase.
Strong heroes who can fill multiple roles are the biggest priority in a well-balanced patch. Special attention should be paid to heroes that can flex between support and core roles, as their playstyles are drastically different.
There are also heroes who can flex within core roles, most commonly between mid and safelane, but there are historically offlane heroes who can be played as a tempo position one.
Heroes that are both flexible and strong in vacuum are usually the ones that end up being most prioritised during the beginning of the drafting stage.
From two sections above, there is one unfortunate conclusion we kind of agree with. If teams are well prepared with perfect understanding of the current hero power levels, the drafting stages between games will look very similar and the game will be mostly about execution, not strategy.
The counterpoint to it is that no team has a perfect understanding of the meta, period. Professional players sometimes tend to overvalue comfortable heroes to play, as was the case with Mars for the longest of times.
Sometimes, they undervalue good fits as well. Most recently it was 
Ancient Apparition who jumped in popularity in-between same-patch tournaments, because teams got more time to prepare and analyse the game.
Most recent small balance patch will undoubtedly shake things up a bit and the upcoming Road to TI will hopefully be about different ideas clashing against each other and stronger ideas winning.
There are also sizable gaps between different parts of the tournament for teams to experiment and adapt. Perhaps because of it we will see three different metas in over three weekends, which sounds very exciting.
And finally we are ready to discuss the current state of the game and what kind of meta we should expect, at least at the start of the tournament. Based on recent tournaments, it feels like the current patch mostly favours very tanky supports, similarly tanky utility cores and flash-farming carries.
This is going to change slightly, though. 7.34d introduced several key changes to the game, most notably nerfing heroes like 
Phantom Assassin, 
Sven and 
Gyrocopter. We don’t think it is enough to rotate them out of the meta completely, but it did open up room for other carries.
Most notably, Phantom Assassin with her cheap Break was a bit too good against 
Spectre and we are already seeing the latter rise in popularity, while Sven nerfs are signalling the mini-return of illusion-based heroes.
In the support position, there is a bit of a shift from tanky save supports back towards lane dominators. Instead of 
Warlock and 
Vengeful Spirit we are seeing a lot more 
Bane, 
Undying and 
Treant Protector. All in all, though, the support meta lately felt very open with only a handful of outright over- or underpowered picks.
Finally, in the second and third position we still see a lot of tanky heroes dominating the meta. Heroes like 
Kunkka, 
Primal Beast and 
Earth Spirit who can start the fight and stay alive for a long time are prioritised in both roles. There is an occasional, more niche 
Puck and 
Outworld Destroyer in mid, but we don’t expect them to be as meta, as heroes who can survive without playing perfectly.
This is our quick overview and expectations for the start of the International, but our read can be wrong — after all a new patch dropped recently and professional players might have a different idea of what works and what doesn’t.
In our opinion, it doesn’t look like it is going to be that much different from what we saw in DreamLeague. Only the carry pool opened up a bit, but for the most part, at least for the Road to TI part, we are expecting teams to play standard meta.
What happens next is anyone’s guess, though. The beauty of the current format is that teams will have two five-day periods to adapt. It is unlikely we will see one strategy dominating the whole tournament, at least we hope so.
Do you like the long TI period, or did you prefer the old approach? Share your thoughts on that matter and your meta read in the comment section below.