One of the biggest parts of the battle pass compendium is the predictions section. Players have a chance to earn over 20 levels by picking out the most successful heroes, teams, and players of the event.
The trouble is that while players can win rewards for their predictions, picking out the right answers can be difficult. The International is an event where wild things can happen, from teams making surprise grand finals runs to big meta surprises. That makes each year’s picks a tricky endeavor as players need to weigh past tournament results, recent patches, and their own instincts about what to pick. TI9 will be no different.
Fortunately for you, we’ve got advice for every major pick in this year’s predictions.
Though Sand King was the slam dunk pick here for several months, Grimstroke has taken over as the most popular support hero in Dota 2. He offers excellent damage output among his peers and has some strong disables as well. He was the most-picked hero at the Epicenter Major and there is a good chance he retains that title at TI9.
The most banned hero category has been shifting around in a big way over recent months. Heroes like Drow Ranger and Nyx Assassin have been hit hard by nerfs, while others including Dark Seer and Chen have been all over the place in terms of popularity.
Things shifted again in the TI9 qualifiers, with Ember Spirit taking over as the most-banned.
Now we’re getting into the categories that are more luck than science. While we can do extensive research to pick out picks and bans, hero with the highest win rate could be a surprise pick that gels with the TI meta, or some other hero that gets picked five times but happens to win four of those games for an otherwise difficult 80% win rate.
Any off-meta hero that has a very strong speciality could work here, but we’re going with Sniper.
Just like hero with the highest win rate, hero with the highest kill average tends to be won by a surprise entrant. Queen of Pain is obviously capable of scoring loads of kills and should be able to hit that five-pick threshold.
Bounty Hunter and Zeus are both strong candidates for highest assist average, but they probably aren’t going to be picked enough to pull the trigger on here. Nyx Assassin is a great hero for setting up kills and has been an enduring favorite in the meta.
Lone Druid is punished harder for dying than any other core. That makes him a great pick for hero with the lowest death average. In the hands of a top-level player, Druid will only die a few times in a game regardless of the outcome.
He isn’t especially popular right now, but Arc Warden is an incredibly fast farmer that has the tools to drag games out for extended periods of time. Add to that how he is a favorite among some of the top carry players entering TI9 and he is very well positioned to win this category.
Basically everything that was said about Arc Warden in the last category can be said right here as well. The only other important note is that his ability to wield two Hands of Midas can make for games where he jumps out ahead in levels very early.
The next two categories are complete guesswork since they boil down to just one game. For both, you’re going to want to look for a survivable core that is capable of fragging out. Morphling is a great snowballing carry that can find kills at any point in the game, making him as good as any other possible pick.
All you can do for highest last hits in a game is pick a fast-farming core and pray. Alchemist, Arc Warden, and Anti-Mage are all good picks, but we’ll go with Tinker based on some of the recent buffs he has received.
OG has been looking great and that’s largely because of Anathan “Ana” Pham’s ability to rack up kills. OG mid player Topias “Topson” Taavitsainen gives him plenty of room in this regard and if OG can make any kind of a run at TI9, there’s a strong chance Ana will be near the top of this category.
The jury is still out on the Aliwi “w33” Omar era of Team Liquid, but the early results are very promising. Amer “Miracle-” Al-Barkawi also seems to be the top fragger of the team and with Liquid positioned to make a strong run, he’s a solid candidate for this category.
Highest assist average is one of the tougher categories to pin down. Both tournament winners and early eliminations can hypothetically end up winning this category, making this largely guesswork. Vici Gaming captain Pan “Fade” Yi typically ends up on the higher end of the assist totals and has a decent chance of being up at the top by the tournament’s end.
Most assists in a game is the least scientific category there is in the compendium. This one could be won by any player on any team. Xu “fy” Linsen from PSG.LGD won this category last year and could put in a repeat performance.
Michał “Nisha” Jankowski from Team Secret is the best pick for both last hit categories. The former Team Kinguin player is one of the most efficient farmers in Dota 2 today and Team Secret can sometimes find itself involved in longer games. Both of those factors make Nisha a great pick for all your last hit needs.
Those same points can be made again here, though there is much less certainty involved as the one-game category means it could hypothetically be any carry player that happens to be dragged into a particularly lengthy game.
This category basically boils down to which player is most likely to choose to have a massive game as Alchemist. There are plenty of solid options here including Syed “SumaiL” Hassan from Evil Geniuses and Vladimir “No[o]ne” Minenko from Virtus.pro, but Armel Paul “Armel” Tabios from TNC Predator is probably the best choice.
Nisha is a strong candidate for this category as well, but we’re looking back towards Team Liquid’s Miracle-. We’re feeling really bullish on where Liquid is headed, and the addition of W33 actually puts him into a stronger position to rake in money.
Rounding out the players category with most different heroes played is Team Secret’s Yeik “MidOne” Nai Zheng. He’s as versatile as they come in Dota 2, and with Team Secret positioned to make a long run in the tournament, he should have plenty of chances to add to his total.
With an incredibly strong season that included first-place finishes at two majors and three ESL One tournaments, Team Secret are the on-paper favorites to win TI9. They had a surprisingly short run at the Epicenter Major, but the team’s season to this point has been just too good to pass up.
Virtus.pro hasn’t looked as dominant in recent months as they did at the start of the season, but they’re still an aggressive team that should be a strong pick for the team with the most kills in a game. Even if they don’t win TI9, you can count on them to try and run over everybody they face.
Highest kill average is usually tied very closely with making the top four, so taking Team Secret here ais a smart move. But the fact that they won this category last year certainly doesn’t hurt.
As we discussed in the heroes category, the single-game picks are more luck than logic. Any team could potentially win fewest deaths in a game, but the best choice is just to pick someone that likes to win quickly. Vici Gaming is about as good of a pick as it gets.
Vici Gaming doesn’t play slow-paced Dota 2, but they’re still a great team to pick for team with most assists in a game for two reasons. First, they have a strong base defense game that allows them to drag things out. Second, Vici captain Fade is possibly the best Zeus player in the world and hasn’t been shy about playing him this season.
The team that wins the longest game will either be a team that plays a slow-paced game of Dota 2, or a team that has excellent high ground defense and a knack for making comebacks. Team Liquid is more the latter than the former, though they’re also not above drawing things out before looking to go for the win.
Longtime players might find this strange, but Chinese teams are playing some incredibly fast-paced Dota 2 in 2019. Vici Gaming in particular is great at taking advantage of an early lead and they have a good chance of winning the shortest game of the tournament.
Highest game length average traditionally ends up being won by a team that is good but washes out of the tournament early. Ninjas in Pyjamas fits that bill and the team actually has one of the highest average game lengths in all of professional Dota 2.
Team Secret won this category last year despite finishing in sixth place. The fact that they are the favorite to win this year makes them an even stronger pick.
This category is all about picking out a team that struggles early, winds up leaning on their favorite heroes, and washes out in the group stage. The most likely team to have this story at TI9 is Infamous.
Dota 2 is more top heavy now than ever before, with a handful of teams standing far above the rest. That should make for a lot of 2-0 series during the main event, which makes the 45 to 49 pick a strong one.
Wings Gaming showed that Dota 2 is an incredibly well-balanced game when they completely ignored the meta but still managed to win TI6. Other teams took notice of that approach and we saw almost every hero picked in both TI7 and TI8, and we’re not expecting that to change this year.
The total number of heroes banned doesn’t usually hit the 100 mark, but it can get pretty close. We’re going with 91-100. but 81-90 is also a very strong option.
We’re going to get some bloody games of Dota 2 at The International 2019 and that has us wanting to pick the maximum here.
For a long time, there was no doubt we’d get at least one game that approached the two-hour mark. But 2018 saw a lot of changes get introduced to shorten the average length of games. These days most games end in under 40 minutes, and it’s somewhat rare to see games even reach 60. We’re hoping that one game will drag out to 70 minutes, but we don’t expect it to go much higher than that.
The shortest game of The International is usually in the 15 to 20 minute range. There’s an off chance we could see a game end in under 15 minutes under the right circumstances, but you definitely shouldn’t go any higher than this.
Last year, SumaiL scored a whopping 31 kills for Evil Geniuses in a losing effort against OG. It might not be SumaiL this year, but I’m expecting somebody to frag out in similar fashion and rack up more than 25 kills.
This year’s meta will likely be an aggressive one with lots of fighting. That’s bad news for five-position supports and it has us looking to the high end of our options here, particularly towards the 18-20 pick.
You only need one long game with a hero like Zeus or Necrophos to hit more than 35 assists. It might not happen very often, but I’m betting it will happen at least once.
And finally, we come to the last and probably easiest pick of the entire compendium. At some point during TI9, we’re going to get a big game from an Alchemist or Antimage where they hit over 1000 GPM.