6.80 has been out long enough that we now have a pretty good estimate of its effects on hero balance. There are a lot of places one could go for this information, such as DotaBuff, but when they show you patch win rates what you’re seeing is the win rate over every single game of public Dota, which means the majority of those games will be relatively low skill contests. I take a different approach that allows me to look at hero win rates exclusively in games that the Dota 2 client categorizes as Very High skill. This often paints a drastically different picture than the aggregate win rates. Two recent examples of this are Earth Spirit and Io. According to overall win rates they were amongst the worst, but when you look exclusively at Very High games their win rates improved by over ten percentage points, which painted a much more accurate picture of their potential.
While I’m quite confident in the value of this approach, the downside is that I cannot create a complete list of games and must depend on sampling. This means that the numbers I provide are estimates of the actual win rates, so don’t get married to the positioning of some of the rankings. However, the sample sizes should be large enough that we can be confident that the larger trends represent actual existing trends, even if we have some doubt as to their precise magnitude. With that being said, let’s start off with the heroes that show the largest improvements in 6.80.
For all the supports that were buffed in 6.80, only three show up on this list. Earthshaker’s buffs were pretty sizable, and it’s pretty much guaranteed that you’ll see him more often at virtually all levels of play. Sand King and Lina have also exhibited significant improvement. Ancient Apparition narrowly missed the cutoff for this list, but he’s received buffs for several patches in a row and could be on the cusp of becoming a semi-regular pickup competitively.
From a position perspective, the bigger shakeup appears to be taking place in the carry position. We have buffs going to carries who haven’t seen regular competitive play in a long time (Sven and Faceless Void), carries who were nerfed out of competitive contention (Troll Warlord, Morphling, and Huskar despite never seeing the light of day in Captain’s mode), and carries who are commonly seen exclusively as pubstompers (Sniper, Bloodseeker, and Riki). In addition to a couple of long-term climbers who will show up later, there’s a lot of potential here for some undervalued pickups, especially given the decline of the dominance of the defensive tri-lane in 6.79. With multi-core lineups becoming more prevalent and 2-1-2 laning being a viable option, carries who were previously seen as not being good enough to warrant a 4-protect-1 strategy might have a shot at achieving previously unheard of metagame relevance. But the conservative read is that Faceless Void and Sven are the most likely winners because they’re closest to what people are used to.
Brewmaster is a hero whose gains in the patch are unlikely to go realized in most of his games. Relatively speaking, his ultimate is fairly difficult to use effectively, and the cooldown reduction it received in the patch is only valuable if you’re diligent about finding opportunities to use said ult. That being said, among better players it’s looking like 6.80 could be a sizable improvement to his greatest strength.
Centaur Warrunner is in a similar if less extreme boat. Traditionally a hero that does significantly better in Normal than Very High, Stampede is the one element of his kit that appears to become more valuable in higher skill games. 6.80’s cooldown buff to Stampede is a more targeted buff towards players that can regularly coordinate Stampede initiations with their entire team.
In addition to the Centaur, Axe and Sniper are heroes that have traditionally had a low skill skew in their win rates, but somewhat surprisingly their 6.80 changes appear to have gone some length towards addressing this. Unfortunately for Meepo, the bulk of his 6.80 improvement has so far been in lower skilled games.
Moving on to nerfs, we have a much shorter list with much starker changes. The two huge ones were the twin 6.79 pub terrors of Earth Spirit and Broodmother. Earth Spirit’s nerf was warranted, but perhaps a tad excessive. I wouldn’t be surprised if it got softened in a future patch, such as dialing back the nerf to remnant cooldown. Broodmother is a much more difficult case. She deserved a buff, but the direction of the 6.79 attempt, that is letting her walk off the bloody map, was poorly received. It will probably take some more experimentation to try to find the right flavor of annoying for her.
From a strictly competitive standpoint, Alchemist was the chief casualty of the patch. His 6.80 hit was a sizable one, and it remains to be seen whether he’ll be dropped or if teams will attempt to salvage him.
The nerf to Legion Commander looks a lot worse than it is. Moment of Courage got hit hard, but the (relatively few) people who have adapted to this have continued to do reasonably well. You can read more about her post-6.80 fortunes here.
Finally we have Venomancer. Of the three dominant supports of 6.79, Veno definitely got hit the hardest. He’s still absolutely viable in pubs, but he might become a more situational pick competitively. Visage and Crystal Maiden both look comparatively unscathed and will likely continue to be viable.
With the buffs and nerfs covered, here are the heroes that improve the most in high skill games in 6.80.
I don’t have much to add here. Most of the heroes on this list are the usual suspects, but there are two heroes of note.
Ember Spirit’s first month was relatively quiet compared to Earth Spirit, but the level of pub play on Ember has improved significantly since his release. This improvement is almost entirely confined to Very High play, but it’s there. I attribute it to better mechanics combined with a more realistic evaluation of his carry potential, or lack thereof. He appears to be a reasonably competent momentum ganker, but it remains to be seen if he’ll have some special niche that keeps him from getting overshadowed by other similar heroes, like, say, Storm Spirit.
In 6.79, Nyx Assassin suffered a huge drop in win rate solely off of a change to the targeting on Impale, but in the intervening months Very High players have acclimated to this change. He’s still harder to play than he was in 6.78, but the rumors of his demise have been greatly exaggerated.
Finally we have the top 15 list for Very High games. It doesn’t necessarily have a ton of predictive power for competitive play, but it’s still something people like to see.
The more things change, the more things stay the same it seems, as the top 3 could have been pulled out of 6.74 two years ago. In fairness to Lycan, he’s been much less conspicuous this time round during his quiet climb back up the charts. His Normal win rate has lagged behind recently, which suggests that his success might be a tad less braindead than it was in the past.
People moaning about Treant Protector nerfs are being ridiculous. His competitive game may have been toned down, but he’s still very strong in a pub environment that is rarely coordinated enough to punish his weaker early levels. He’s also still capable of surprising teams competitively but only as an occasional surprise pick. Start relying on him and opponents will develop antibodies pretty quickly.
In terms of popularity, Mirana has been pretty ridiculous. She has the most games played in my week long sample, and that includes beating out both Terrorblade and Phoenix.
And finally, RIP Skeleton King. See you in the Baller Tier in the sky.