Pub Tier List 6.81
Disclaimer: We've assembled this list largely from win rates in public games. It may be different in many ways to professional play, but it offers insight on why certain heroes have a distinct impact in pubs.
Ten days is a long enough period for the changes introduced in a patch to take a statistically significant effect on the public matchmaking metagame. As previously, we will be creating a tier list based on the Win Rate of heroes in pubs, with explanations and analysis on the most interesting cases.
The order of the heroes in each Tier is based on their Win Rate, with highest first. Additionally, we will be adding a +, - or = signs to indicate the changes from the last Tier List.
Tier 1: Pub Stompers (56%+)
Necrophos (+) 
Abaddon (=) 
Omniknight (+) 
Wraith King (=)
Abaddon and Wraith King again establish their Tier 1 positions, with their pub friendly skill set of not dying.
Abaddon's spike in win rate was due to a bug that gave his ultimate the new Aghanim's upgrade, without needing Aghanim's Scepter. It is not surprising his pick rate rocketed in the same period.
In the past, our Tier 1 6.79 heroes shared a strength in team fight potential. Today, they have a common trait in resiliency. We have two new entrances into our top tier.
Necrophos: (55.10% -> 58.89%) Our #1 hero laughs at your respawn timer and has vaulted to the top win rate spot. Despite many PSAs on his carry potential, he is often relegated in low bracket pubs to a support or even, shiver, a jungle role. Aside from his strength in team fights, his durability is the trait he shares with our Tier 1 heroes. The Reaper's Scythe change (30% increase in respawn timer) seems to have a significant impact on pubs, and we are excited to see his potential in professional play.
Omniknight: (54.05% -> 56.35%) Rarely seen in competitive play due to his weakness in the lane, he is incredible in pubs because no one knows how to lane. Pubs tend to shift towards greedy lineups. If your team can have five carries or five mid heroes, whats the difference in having just one greedy support? Omniknight received a multitude of buffs. An extra second to his ultimate and the changes to Repel appear to have the greatest impact.
Tier 2: Pub Winners (52%-56%)
Lich (-) 
Zeus (=) 
Vengeful Spirit (=) 
Witch Doctor (=) 
Centaur Warrunner (=) 
Warlock (=) 
Brewmaster (+) 
Silencer (=) 
Death Prophet (=) 
Sven (=) 
Treant Protector (=) 
Viper (=) 
Mirana (=) 
Bloodseeker (=) 
Ogre Magi (+) 
Razor (+) 
Ursa (=) 
Crystal Maiden (=) 
Venomancer (=) 
Tidehunter (=) 
Disruptor (=)
Brewmaster: (50.77% -> 54.09%) Heroes that require micro are not very friendly to pub players (with the exception of Invoker, our 2nd most picked hero). Brewmaster is one of the least picked heroes in the pub meta, but unlike his fellow unpopular heroes such as Io, Anti-mage, and Broodmother, his win rate has been consistently greater than 50%. The last buff to Blink Dagger also proved to help a hero in constant strain for mana. And the latest buffs to his passive, granting a guaranteed critical, adds a good bit of damage for his initiation.
Razor: (49.40% -> 52.71%) The strength of a strength buff (from 1.7 to 2.3 per level). Razor remains as one of the best anti carries, and this increase in his durability is proving to be more significant than we thought. We expect he is able to extract more potential from his Static Link, and his Aghanim's upgrade makes him an adequate pusher as well.
Vengeful Spirit: (54.26% -> 54.77%) We anticipated that the VS change to her aura would have a more significant impact due to her feeding tendencies in pubs. You actually don't mind dying when you swap. However, we failed to realize that most pub players prioritize her other abilities before her aura. Among pub players, her most popular ability build prioritizes Vengeance Aura last. She still remains one of the most friendly pub friendly supports and we would consider her a top draft pick in CD and CM games.
Tier 3: Average Pubbers (49%-52%)
Spectre (-) 
Dazzle (-) 
Pudge (+) 
Drow Ranger (-) 
Undying (+) 
Riki (-) 
Slardar (-) 
Faceless Void (+) 
Jakiro (=) 
Lycan (-) 
Earthshaker (=) 
Shadow Shaman (-) 
Luna (-) 
Ancient Apparition (=) 
Slark (=) 
Spirit Breaker (=) 
Sniper (=) 
Bristleback (=) 
Phantom Lancer (=) 
Juggernaut (+) 
Keeper of the Light (=) 
Weaver (=) 
Kunkka (+) 
Clockwerk (=) 
Sand King (=) 
Doom (+) 
Medusa (+)
Dazzle: (53.33% -> 51.96%) One of our favorite heroes received a significant nerf in the ministun of Poison Touch. Though it will still stun at the end, the initial stun was often situationally crucial for quickly interrupting channelled spells, including the most powerful escape item in pubs: the Teleport Scroll. Dazzle still remains an excellent team fight hero, with Shallow Grave saving your YOLO teammates.
Faceless Void: (47.94% -> 51.50%) Void is now no longer a losing hero. His increase of 2 in his base agility was negligible, so who knew how huge a turn rate buff could be? Time Walk became a little more effective as an escape tool. We are also seeing an increase in popularity in the competitive scene as well, with Aghanim Scepter a common first big item among carry players like Burning and EternalEnvy. This item choice is unpopular in pubs, but perhaps should be reconsidered, with its 70% win rate.
Lycan: (55.38% -> 51.35%) Lycan has one of the most significant drops in win rate this patch. It was a well-deserved nerf, as he had the highest win rate in competitive play and was a mainstay in the first ban phase. However, it was this nerf in conjunction with the buff to Roshan's armor and the nerf to Necronomicon that hamstrung Lycan in this patch. Still, he remains an effective hero, especially in pubs where weak lanes can give your team room for a jungler.
Tier 4: Pub Losers (46%-49%)
Enigma (=) 
Night Stalker (-) 
Phoenix (=) 
Elder Titan (=) 
Axe (- -) 
Troll Warlord (=) 
Lion (=) 
Shadow Fiend (=) 
Huskar (=) 
Dark Seer (=) 
Pugna (=) 
Invoker (=) 
Beastmaster (=) 
Meepo (=) 
Skywrath Mage (+) 
Leshrac (=) 
Lifestealer (=) 
Nyx Assassin (=) 
Dragon Knight (=) 
Phantom Assassin (+) 
Nature's Prophet (=) 
Chaos Knight (+) 
Tiny (=)
Night Stalker: (49.48% -> 48.91%) By all means a very small and insignificant change. Interestingly, the 6.81 patch has introduced a buff to his upgraded ultimate - [missing skill: night-stalker-darkness-5278]. It is possibly one of the reasons for the lower Win Rate, since a lot of people were trying to get that extra effectiveness, rushing for an 
Aghanim's Scepter, instead of getting more useful items, such as 
Black King Bar.
It is further substantiated by the fact that Night Stalker did not receive a huge popularity boost after the 6.81 patch and the fall in the Win Rate can not be attributed to an extended player base on the hero, who were not playing him efficiently. Hence, it is probably the case of existing players trying out new things that are extremely interesting, albeit gimmicky.
Axe: (52.49% -> 48.62%) We initially assumed that the increase in consistency (and predictability) of 
Counter Helix would only increase the Win Rate on this manly hero. We were wrong.
The dunk of almost 4% derived from a change on an ability that in the long run should be exactly the same is peculiar. On the other hand in a real-life situation, where you manage to 
Berserker's Call several enemies, you only have 3.2 seconds of being unconditionally hit by the enemy. In this time frame, the amount of successful Counter Helixes that use PRD is probably lower than with a true random. Hence the decrease in damage output and Win Rate.
If, for example, there was a way to ensure that Axe is consistently hit by enemies in a small range around him for 10+ seconds, the change would not be as drastic.
Skywrath Mage (44.75% -> 46.94%) An increase in movement speed that made this hero the second fastest in the game did pay off bit time. It has effectively increased the Win Rate for the hero by 2.19%, which is very substantial. Considering the fact that levelling 
Concussive Shot past the first point at early levels is less popular and yields lower Win Rate (see: ability builds), the buff to it could not have such a drastic impact.
It further proves the point about secondary stats, such as Turn Rate, Speed, Cast Point etc., have a huge impact on the gameplay. These buffs, as well as an increased popularity of the hero among pub players could also put him on the radar for the professional teams. It will be very entertaining to watch him being played with heroes like [missing hero: outworld-devourer] on the same team.
Phantom Assassin (43.74% -> 46.38%) In our 6.81 changelog analysis I have stated that while the buffs to the hero might be really strong, they are still rather meaningless, considering the hero is somewhat weak by design and extremely inconsistent. While I still believe that this holds true for the competitive scene, it is impossible to deny the impact it had on the public matchmaking.
An increase of Win Rate by 2.64% is very impressive. Looking at our ability build page for PA, we can see that lower priority on 
Blur is still more effective. It does make the window of opportunity before the enemy cores get their 
Monkey King Bars a lot more distinct, but it rarely makes this window longer. Hence, having a potentially deadly nuke in the form of 
Stifling Dagger is still preferable.
Chaos Knight (45.13% -> 46.05%) It is interesting how an increase in inconsistency makes some heroes stronger, while an increase in consistency makes others weaker.
To be fair, there was no drawback for the buff to the Chaos Knight introduced in 6.81. And skilling 
Phantasm at level 6 is still sub-optimal, as seen on our ability builds page. However, there is no denying that this inconsistent buff has not made the hero broken, while significantly increasing his Win Rate and popularity. Hence it can be considered a good change and it will be interesting to see the hero in the competitive action once more.
Tier 5: Pub Throwers (41%-46%)
Terrorblade (- -) 
Legion Commander (=) 
Lina (-) 
Timbersaw (=) 
Gyrocopter (=) 
Batrider (=) 
Templar Assassin (=) 
Bane (=) 
Bounty Hunter (=) 
Lone Druid (=) 
Magnus (=) 
Clinkz (=) 
Windranger (=) 
Rubick (=) 
Visage (=) 
Chen (=) 
Puck (=) [missing hero: outworld-devourer] (=) 
Storm Spirit (=) 
Naga Siren (=) 
Broodmother (+) 
Tinker (=) 
Enchantress (=) 
Morphling (=) 
Queen of Pain (=) 
Alchemist (=) 
Ember Spirit (=)
Terrorblade: (52.21% -> 45.98%) While the decrease in Win Rate could be expected, it is the decrease in popularity of this annoying hero is what I am personally really happy about:
Did the buffs hit the hero too hard? Quite possibly. Can he still be a threat in good hands with access to really crazy steroids and high survivability later in the game? Yes.
Much like Naga Siren (who also took a rather strong nerf) this hero can become a "comeback" hero with ease. If he is left unattended for a prolonged period of time he will farm his items and he will destroy pretty much everything on his path. Now, however, it require a lot more skill, patience and time.
Lina: (46.30% -> 45.55%) The hero has received some very interesting buffs making her the second only to Sniper in terms of range. Looking at her popularity, we can see that there was a distinct spike in it right after the patch, probably contributing to a lower Win Rate on a hero that has been consistently receiving buffs from patch to patch:
Cross-referencing the Win Rate and Popularity graphs on our trends page yields a disappointing truth - after the initial spike in popularity resulting in a drop of Win Rate, the Win Rate did manage to return to its original value, however it did not increase despite the buffs.
Lina is an extremely interesting hero to play with one of the biggest bursts in the game that transition her into a decent semi-carry with the help of 
Fiery Soul. The fact that she has decent DPS in all portions of the game make her a very decent pick for any position, except for offlane. Her Aghanim's Scepter upgrade also has no substitutes, as it is the only spell in game that can deal 1250 magic damage that will go through magic immunity.
We will keep our eye on the hero, since her Win Rate still has a decent potential to grow. For now, however, she will reside in the Pub Throwers tier...
Broodmother: (39.20% -> 42.70%) She is a niche pick. She only works well if played by certain players in certain situations and that remained true. And a 3.5% increase in Win Rate with absolutely no effect on popularity only proves that.
It is frequently the case that heroes that require a lot of microing and generally have a low Win Rate in public matchmaking are beastly in the competitive scene. Broodmother is possibly on of these cases, since a lot of her Win Rate comes from higher skill brackets.
Since the hero was previously available in the Captain's Mode, it is possible that we will see her appear in TI4. And, honestly, no matter how much I dislike Rat Dota, Broodmother is always entertaining to watch.
Tier 6: Repick Heroes (<41%)
Shadow Demon (-) [missing hero: antimage] (=) 
Tusk (- -) 
Io (=) 
Earth Spirit (...)
Shadow Demon: (41.11% -> 40.87%) The changes to this hero were not substantial, except for an Aghanim's Scepter upgrade that is really hard to come by when this hero is played as a support, since he also requires a decent amount of mobility to be effective at later stages of the game.
The fact that this buff is extremely situational and cannot be achieved in every game is what made this hero drop a little bit of his game. It is the possibility of an extremely potent upgrade, however, is what probably makes the situation even worse - disregarding some minor things like [missing item: observer-wards] and even 
Force Staff to save money to get this desired double ultimate is a bad idea. And, as with many other bad ideas, pub players will always love and cherish it.
Tusk: (46.60% -> 40.18%) Interestingly, there was no crazy increase in popularity we have assumed when we saw a drop in Win Rate - after all it is usually the main reason behind sudden drops - something situationally effective (albeit gimmicky) is added to a hero and a horde of new players decide to try it out concentrating all their effort around this change, rather than playing a balanced game.
There was no popularity increase for Tusk. Moreover, the change to 
Snowball is widely considered a buff, and there is not explanation as to why there would be a drop in Win Rate.
Quite possibly, the case is similar to the one of Night Stalker where a multitude of players decide to try and play around the changed ability and miserably fail at it. It is still not enough to explain a 6.42% Win Rate decrease. We will keep an eye on the changes to the heroes Win Rate as well as his competitive scene performance and will try to come up with an answer to this mystery.
Closing Comments
The 6.81 patch was not that game-changing, but it has refined an excellent 6.80 dealing with the more popular heroes without breaking them, while introducing a variety of new options for the players at all levels of gameplay.
We will be releasing "Weekly Trends" Blog Posts each Sunday from now on to see how the meta will be developing and will cover the most interesting changes to come, so stay tuned!
As always:
Thank you for Reading!
No comment on Ember spirit, oh well :( Still a nice review though!
imo the winrate drop for tusk is because people dont launch the snowball so the enemies have 4 seconds to escape/react
not sure if that many people are noob enough to do that tho
Some heroes, like Razor, I suspect had an increase in win rate due to the suggested item lists being updated (less Vanguards, more Meks).
@Jebus Slept Great point! Vanguard had a drop in item use in 6.81. It was his #2 built item in 6.80.
@Nøstalgia Ember Spirit did stay in the same tier, but his win rate dropped by nearly 7%, the largest decrease among all the heroes. The damage nerf hit him harder than we expected. He still remains a top pick/ban in competitive play, so it's disappointing to see his lackluster performance in pubs.
Good read!
I think some of the reasons why some hero inexplicably dropped in winrates is because of how they match up to other heroes who were buffed.
The only thing which amazes me is that how invoker is one of the most picked heroes.
People still attempt to play Terrorblade as he was back in 6.80: a high-impact, early-engaging carry. Unfortunately with a nerf to his Q and his overall survivability he's literally the squishiest motherfucker out there. Peeps need to know that TB needs more items now; drums are now core (no more Yasha/SnY/Manta rushes), stats-first is a viable skillbuild and he REALLY needs to play safe early. No more first blood attempts. Once people realise to play him more like a Lycan, he'll be as popular as ever.
Initially, I was rather surprised that the 6.81 changes lowered Llch's win rate. You'd think that unlimited chain bounces and tower ice armor would pretty much give him an extra half percent to one percent on his win rate.
I think with enough given time, Lich actually might pull himself back into tier 1. While he had an initial dip in his win rate due to the people greedily rushing a scepter, his win rate has actually been approaching +56% again. Give it a month (or 2) and maybe he'll be back in tier 1? I think people are are beginning to understand while the ult is cool and all it's not something that you should try out instantly. And now since everyone knows about it people aren't as inclined to rush it anymore.
Yes, I know I look like a scumbug since I only play Lich and wraith king most of the time, but the buffs to Lich have actually been rather beneficial. Ice Armor to towers is very helpful for lane control since a combination of sacrifice (artificial pulls and denies), ice armor (anti-push), and nukes (force heroes out of lane) allow Llch to almost command the lane in the laning stages in regards to pubs. He can pretty much decide when he can pull, push, and harass. Auto-attacking is far from penalized when you can instantly pull and safety hide behind an Ice-Armored tower.
I have to say that this article was not as well written compared to the previous articles. First of all, there was quite a conspicuous error. "Did the buffs hit the hero too hard?" was said for Terrorblade, but… he wasn't, and for sure you meant nerf there.
That aside, some changes seem very trivial. Yes, Shadow Demon's 'tier' changed with respect to your (quite arbitrarily) predefined tiers but… seriously? A change of 0.24% warranted a comment, and not Phoenix, Naga Siren, or Ember Spirit? Plus, some comments are pretty pointless – like the Chaos Knight one, which you realised, but this isn't a conversation – if you find something to be not really true, you should go back and edit it out instead of justify it.
Also, more of a minor complaint, why are the tiers defined this way? 56/52/49/46/41 have arbitrary gaps between the chosen boundaries. Instead of trying to fit an equal number of heroes into the middle tiers, more realistic numbers should be used. For instance, why is 49%-52% average and not 48%-52% or 49-51%? It doesn't make sense.
That aside, it was still obviously put together with effort, and an enjoyable read overall.
As another note, the WIN RATE AND MATCH VOLUME BY LEVEL in the trends tab (which you linked to in the article) still isn't fixed. I would really like it to be as it would reveal extremely interesting patterns. Since it's error is smaller at higher levels, you can already see the rise for Anti-Mage above average or the fall for Crystal Maiden below average but it's useless at lower levels because they count every match that has been to that level.
BUFF EARTH SPIRIT PLS ICEFRAUD :(
I'd enjoy if the bias was toned down a bit. Maybe the author had a bad time against Terrorblades but really, constantly bashing him without providing any relevant criticism or analysis is rather annoying. A 6.23% drop is HUGE. To say that he was "maybe" overnerfed is putting it lightly. And prepatch 52.21% isn't even that bad.
lina can be played offlane solo... she can get last hits with skill 1 , & she's little of everything, support, carry, nuker.
Look, i know this is not the place to ask, but i've ran out of options searching for myself and have to ask you guys at dotabuff out.
Some time ago, checking the 'Dazzle' and 'Flow' posts I found, soon after, a webpage that was named more or less 'increase your solo mmr in Dota 2'. There were around 3-4 tabs of content to it, it had a white layout and most importantly: it held GREAT advice to keep a good state of mind that would help one step the right way into a DOTA 2 match. I still remember the basics but would like to read it again, though google and my memory have proven unsuficcient, as I only find threads in Reddit and dota.dev about not having a high enough mmr.
Does someone know the name of this site and would they be kind enough to share it so I can save it (as I should have done) this time?
Thank you in advance.
Anytime they add an aghs buff you can expect a win rate drop because almost all the time for those heroes aghs is a sub optimal item and should be a luxury item. Lich for example, I'll bet many are rushing aghs now for the uber ult but shiva's, veil, drums, scythe are still better item choices in my view. Same with the explanation for night stalker.
@That one guy Are you referring to our State of Mind post? That was a great piece by KawaiiSocks. You can check it out here! http://dotabuff.com/blog/2014-01-16-state-of-mind
That post was the one I referred to as 'Flow', as I have seen it called that way once or twice here. I mentioned two Dotabuff posts because those are the closest memories I hold relative to that other page, which, I insist, was a separate webpage. Thank you though, I read the article back on the day and it was pretty neat
The amount of un-professionalism shown in the article is ludicrous. Keep your Terrorblade bashing to yourself.
I predicted that in 6.81, supporters are going to have a higher win rate since some of them have been added a Aghanim's effect. That increases our joy as a supporter in a longer game. However, I was surprised by the result that Lina has got a decrease in win rate.
This kind of blog posts should be added on dota2 news better than http://www.pcgamer.com/tag/three-lane-highway/
Some comment on changes that had been expected but didn't happen could be nice. Husk, for example. Had Burning Spear become no longer a UAM in 6.81, a lot of people were expecting Husk to become strong pub stomping hero. Didn't happen for some reason, and he still stays in 'losers' bracket.
@Lord Bison Saw what you said, thought I'd help you out. Is it this one?
http://slashstrikesguidetosoloranked.weebly.com