6.81 Hero Shifts: Day 1

(Better tables available at

http://www.teamliquid.net/forum/dota-2/449572-hero-shifts-681-after-one-day

or

https://docs.google.com/spreadsheets/d/1QAoemqiZcW-MhlNS6ThGmZ_5oTlqhQ3bSC91jUPw2Og/edit?usp=sharing)

Going to do a quick and dirty comparison for the hero changes in patch 6.81 using the shift on Dotabuff between their4-30 win rate and their previous monthly win rate.  Someone will inevitably complain that it’s “too early” to look at win rates, and I will patiently ignore the epistemological nightmare involved with unpacking that statement.  To try to shortcut that, there have been enough games that large results are guaranteed to be indicative of something.  It is true that sometimes it takes the general public some time to digest a new change.  Typically this effect is strongest among heroes with a low win rate or a large win rate drop and tends to be positive as the community learns to play them.  Small results might be indicative of a small but noteworthy change, or they might just be statistical noise.  Non-results are not proof of no change.  Some hero traits are only noteworthy in an organized setting and do not make a statistically significant impact on pub play.

Also, I don’t claim to have an exhaustive knowledge of the bugs fixed in 6.81 nor the bugs introduced.  I also don’t know offhand which heroes saw the most noteworthy changes to their recommended item lists.  6.81 is kinda unique for hero balance in this regard, so keep in mind that the listed hero changes do not tell the entire story.

Now let’s start with the Competitive Hero Nerfs:

Big

  • Ember Spirit -6.57%
  • Lycan -4.50%
  • Luna -3.22%

Noteworthy

  • Naga Siren -2.25%
  • Visage -1.83%
  • Dazzle -1.65%
  • Centaur Warrrunner -1.34%
  • Invoker -1.07%

Negligible

  • Ancient Apparition -0.56%
  • Mirana -0.32%
  • Batrider +0.13%

Ember Spirit got crushed, with his first day win rate now hovering around 40%.  This is a somewhat larger shift than I expected, but a small damage nerf to a short cooldown ability adds up.  People might still need time to adapt to this one, and it would be interesting to see some returns by skill build.  A relatively early 4 points in Searing Chains might be a must now with treating it as a one point wonder being even worse than it already was pre-patch.

HP nerfs are a big deal, but I wasn’t certain that only losing HP on his ult would hurt Lycan as much as it did.  In any case good.  Maybe DK (and DK alone) had figured out how to stop the hero, but Lycan had the highest competitive win rate of 6.80 of any hero with over 100 games.  It’s also positive that Lycan is the hero most hurt by Roshan receiving an extra point of armor in the patch.

Of course, the 2nd highest win rate was Mirana at 59.9% and third highest draft rate, and she barely got touched in 6.81.  Expect to continue to see a lot of her just about everywhere.

Luna proves once again that base primary stats are pretty important.  I wouldn’t be surprised if we see another Naix situation where Luna is still viable but in a much narrower window leading to a pretty substantial win rate drop until teams stop treating her as a default option for their carry.

I felt that Naga Siren‘s carry potential was overrated, but I’m not going to cry if Naga the Living Roomba falls off the face of competitive Dota.  Ember Spirit’s competitive evolution is the much bigger loss.

I’m disappointed at the Dazzle nerf.  He was popular in competitive 6.80 (7th most games played) but he wasn’t unreasonably successful at a 51.8% win rate.  Maybe there’s a more long-term intention here that I’m not seeing, but this feels harsher than was warranted.

Finally we have Visage whose drop is, I suspect, more of a result of the negative attack speed bug fix to Grave Chill than the modest cooldown nerf to the ability.  But we’ll get more into that attack speed bug fix in a moment.

Moving on to the Pub Nerfs:

  • Terrorblade -8.00%
  • Phoenix -2.87%

Not going to say much about Terrorblade.  The hit to level 1 Reflection’s duration makes him a lot less threatening in lane, and Strength nerfs of any kind are a big deal.  He’s currently hovering around a 46% win rate, so his status as premiere pubstomp carry should be considered obsolete.

Phoenix is actually a bit more interesting.  She took a couple of separate nerfs, but she was also considered to be one of the biggest beneficiaries of the negative attack speed bug.  Given this, you’d expect her win rate to have taken a bigger hit, but so far it’s been a relatively modest drop.  We’ve already talked about Visage as being another hero with negative attack speed, but here are a few others.

  • Enchantress -1.13%
  • Lich -1.22%
  • Viper +0.17%

Enchantress received a minor buff to Untouchable’s duration presumably to offset the loss of effectiveness, but the net result appears to be a nerf.

Despite buffs to both Frozen Armor and his Aghanim’s effect, Lich is down as well.  I’m assuming this is because every ability other than Sacrifice comes with an attack slow, but it could also be a result of the Aghanim’s effect, which we’ll get to shortly.

Viper didn’t see any direct changes in the patch, but every ability besides his passive provides an negative attack speed debuff.  Confusingly Viper is slightly positive in the patch.  Could this be due to Mek replacing Vanguard in his recommended items?  It’s impossible to say yet, but it’s worth keeping an eye on.

With Lich already mentioned, time to move on to Aghanim’s Upgrade Buddies:

  • Omniknight +2.83%
  • Abaddon +2.29%
  • Keeper of the Light +0.03%
  • Shadow Demon -0.29%
  • Windrunner -0.51%
  • Lina -0.65%

Aghanim’s Effects are one of the few buffs that can hurt a hero’s public win rate because they encourage players to change their playstyles in a way that might actually hurt their team.  I still feel that Keeper of the Light‘s new Aghanim’s effect is amazing, but it’s not surprising that it hasn’t had a positive impact.  It’s a great item for a support with excess cash, but it doesn’t turn him into a legitimate, farm-deserving semi-carry.  KotL mid is dumb (generally, not going to rule out edge cases but you shouldn’t pub around edge cases), and KotL stealing farm in lane from actual carries is even dumber.

So the general takeaway from this is that it’s hard to judge the value of an Aghanim’s upgrade from pub returns because how you get the upgrade can be the most important factor of whether it was worth it for a support.  With that in mind, Abaddon‘s Aghanim’s Upgrade is looking surprisingly potent because I can’t figure out what else would be driving his win rate increase.  He even has a (relatively minor) attack speed debuff.  Maybe Mek as a core item is new for him.  In any case, his day 1 win rate is 58.88%, and if it holds up he’ll likely have the highest pub win rate for the foreseeable future.

Omniknight sees an even higher win rate increase than Abaddon, but he received a set of buffs in addition to his Aghanim’s effect also being buffed.  I suspect that the base duration increase to his ultimate is the big driver of his win rate, but it should be mentioned that his Aghanim’s effect has been substantially improved this patch.

Don’t conclude anything about Windrunner yet.  She’s the only one of the Agh Effect heroes that haven’t seen a surge in usage, so it’s impossible to say whether people are actually experimenting yet.

In other item driven shifts, let’s look at the No Longer Unique Attack Modifier Duo:

  •  Huskar +1.31%
  • Ursa +0.23%

Ursa is a funny case in that what seems like it ought to be a huge change is barely registering, but the bottom line is that this change does nothing if you don’t take advantage of it and just rush a Vlad’s as always.  Once people start experimenting with newly available itemization pathways and the better ones start spreading through the collective consciousness of pub play, then we’ll likely see a bit more movement.

Huskar is a simpler case.  Since his rework Burning Spears has been hugely important to his success, but came at the expense of having lifesteal.  People kept on buying lifesteal, either at the expense of using Burning Spears as much as they should or at reduced rates of return on item progression.  Now there’s no trade-off, so have fun disregarding viability and just rushing a Satanic because it’s hilarious.

In the final specialty carry, we have the Less of a Pub Disasters:

  • Broodmother +4.26%
  • Earth Spirit +1.70%

Substantial improvements to two of the three least successful pub heroes in 6.80, but neither hero is exactly tearing things up.  Broodmother‘s 6.81 day one win rate is just 43.29% and Earth Spirit‘s is 34.45%.  Both heroes are a bit problematic from a balance perspective given how much stronger they are in experienced hands.  Expect to see more carefully applied buffs in the future in both cases.  Nevertheless, Broodmother has so far received the largest net win rate boost of the patch.

With all of the side stories taken care of, I’m going to just group the rest of the patch changes by hero role, starting with the Carries:

Substantial Buffs:

  • Faceless Void +3.43%
  • Phantom Assassin +2.65%
  • Troll Warlord +1.89%
  • Juggernaut +1.79%
  • Sven +1.67%
  • Chaos Knight +1.34%
  • Legion Commander +1.25%

Faceless Void is an interesting case, as I can’t remember the last time we saw a turn rate buff.  Combined with a small base agility buff (half the magnitude of Luna’s base agility nerf), these two buffs make Void the most obvious winner of the patch out of the carries.  He might still be too specialized to become an top Pick/Ban competitively, but expect to see more of him in teamfight strats like DK ran vs Empire in the StarSeries finals or like C9 ran against RoX.KIS earlier today.

Juggernaut, Sven, and Troll Warlord all get noticeable boosts off of their buffs.  It’s uncertain whether they’ll see any competitive play off of this, but of the three Juggernaut was the least irrelevent in 6.80.  Troll Warlord is an interesting case in that professional teams did get a lot out of his ultimate when he entered CM, but the nerf to Whirling Axes damage appeared to kill off his viability mid.  Without mid as a laning option, it might be difficult for teams to fidn a satisfactory way to lane him.

Legion Commander‘s buff is far more substantial than it appears.  The vast majority of people playing her in pubs are still running jungle strats that ignore Overwhelming Odds.  People who lane her and max Overwhelming Odds first were already more successful and most of her buffs were targeted exclusively at this skill.  She may have also benefited from the Duel bug-fixes in the patch, but it’s difficult to say how much of a role those played.

Phantom Assassin and Chaos Knight both receive much-needed buffs, and PA’s net benefit was actually quite substantial.  They still don’t look like especially strong picks in general with day one win rates of 46.43% and 46.22% respectively.

Substantial Nerfs:

  • Riki -1.15%
  • Wraith King -0.94%

I don’t know why Riki is down.  As hilarious as the idea of him becoming too effective at denying his teammates is, that’s probably not driving this.  Smokescreen does come with an attack speed slow though…

Wraith King saw both a buff to his Crit passive (+25% at all levels) and a -1 nerf to his armor, and I guess if +1 armor makes you take literally no damage then -1 armor must make you take literally infinite damage.  In all seriousness, this tradeoff appears to be a net negative over all pub play, but I wouldn’t be surprised if the results play out differently if you look exclusively at higher level play.

Small or Statistically Insignificant:

  • Gyrocopter +0.68%
  • Lone Druid +0.63%
  • Medusa +0.62%
  • Bloodseeker +0.44%
  • Anti-Mage +0.24%
  • Sniper +0.18%
  • Drow Ranger +0.04%
  • Tiny -0.08%
  • Doom -0.28%
  • Shadow Fiend -0.43%

For all the talk that Bloodseeker‘s 6.81 buffs have gotten, I don’t feel that they’re as big of a deal as the ones he saw in previous patches.  This is more of a “Maybe Sempai Will Notice Me” buff from Icefrog, because Bloodseeker is ripe for the noticing.

Moving on to Semi-Carries (and no, I do not want to have a long and tedious discussion on the precise definition of “semi-carry” right now):

Substantial Buffs:

  • Necrophos +3.42%
  • Razor +3.30%
  • Brewmaster +2.93%
  • Kunkka +1.74%
  • Silencer +1.60%
  • Beastmaster +1.29%

Necro is seeing a pretty huge boost off of a fairly unique set of changes.  I’m going to assume Reaper’s Scythe adding extra death timer is the big driver here, and I’m honestly a little shocked at the magnitude of the effect.    His day 1 win rate if 58.48%, which is less than a half a point behind Abaddon’s.

Strength buffs are a big deal, and Razor is largely riding off a .6 increase to his strength per level.  Like Necro, he’s not a character you can just shove into any lineup situation, but his ability to jam a right-clicker in lane is unparalleled and with Scepter he’s a pretty strong pushing hero, so I’ll be surprised if we don’t see any teams trying to work him into their 6.81 playbooks.

Not only is Brewmaster apparently getting a lot out of his patch buffs, he’s also doing it while having an Attack Speed slow on Thunder Clap.  Maybe 6.81 will be the patch that returns Brewmaster to prominence.

Similarly, Kunkka‘s buffs might end up being pretty significant.  Especially because I am 100% certain that the average pub Kunkka is not getting as much out of X Marks the Spot lasting twice as long on allies as they could be.

Does Silencer belong here instead of with the supports?  I don’t care, but you have to admit that the sizable buff to his agility growth is a semi-carryish kind of buff.  In any case, I’m assuming it’s doing more of the work here than the cooldown decrease to Curse of the Silent, and maybe we’ll see a shift in how Silencer is played as a result.

Beastmaster‘s +4 base damage is statistically relevant, but not especially exciting.

LOL:

  • Tusk -7.69%
  • Axe -3.87%
  • Zeus -0.94%

I really wouldn’t have guessed that Tusk would have the second highest win rate decrease of the patch, ending the day at 38.55%.  It’s too early to tell whether the new Snowball is an net nerf or if virtually nobody has a clue how it works yet.

Axe losing almost 4% with no direct changes other than Counter Helix being moved to Pseudo Random is almost as hilarious.  I don’t claim to know what’s going on here, but I am curious as to why Axe sees such a huge negative change when Legion Commander’s also received the Pseudo Random treatment but has a positive shift.  It could come down to the actual value of proc rate between the two abilities being vastly different so that Helix benefits far more from proc clustering, or it could be that the Overwhelming Odds change is so strong that it’s canceling out the effect.

I also have no idea why Zeus is down off his changes.  Yes, he saw some vision range hits, but you would expect the cast time on arc lightning and the new utility of Lightning Bolt to make up for it.  Whatever the case, this likely has no bearing on his potential for competitive viability, as an organized team will make better use of the scouting potential of Lightning Bolt.

Small or Statistically Insignificant:

  • Templar Assassin +0.49%
  • Meepo +0.42%
  • Elder Titan +0.39%
  • Bristleback +0.18%
  • Pugna +0.18%
  • Magnus +0.17%
  • Spiritbreaker +0.10%
  • Bounty Hunter +0.02%
  • Nightstalker -0.17%
  • Queen of Pain -0.26%
  • Tinker -0.46%

The only surprising entry here is Meepo.  I would have expected +10 movement speed to have a more noticeable effect.

And finally closing with supports, first we have the more Support Oriented Junglers:

  • Chen -1.58%
  • Enigma +1.80%

Chen is another curious case as his only direct 6.81 change is a relatively insignificant cooldown buff to his least often skilled ability in Penitence.  I have no idea what’s hurting Chen in 6.81.

Enigma on the other hand is doing pretty well for himself, presumably off the area increase to Midnight Pulse.

Now for Traditional Supports:

Substantial Buffs:

  • Undying +4.12%
  • Skywrath Mage +1.99%
  • Witch Doctor +1.81%
  • Vengeful Spirit +0.79%

Undying is far and away the biggest 6.81 winner of the supports, and so far he has the second highest net win rate increase of 6.81.  This boost is likely driven by the Zombie Deathlust buff for Tombstone, so watch out for that.

Unlike Meepo, Skywrath Mage actually sees a decent boost off a +10 movespeed buff.  Well, that and a slight damage increase to Concussive Shot.  The more important question is how many people have I managed to piss off by listing Skywrath as a support and Silencer as a Semi-carry?

Witch Doctor sees a decent increase that’s primarily driven by extra damage on his ult.

Vengeful Spirit sees a small increase, but that’s not surprising given how rare it is for players to max Vengeance Aura.  6.81 may have substantially changed her skill priority, but for now it’s too early to call.

Nerfed:

Nyx Assassin: -1.69%

I’m skeptical that a 11 to 13 second cooldown on impale is driving that much of a decline for Nyx Assassin.  Maybe it’s more important than I give it credit for, but there’s a decent chance something else is at play here (or just statistical noise).

Small or Statistically Insignificant:

  • Leshrac +0.59%
  • Warlock +0.17%
  • Tidehunter +0.07%
  • Lion -0.62%
  • Disruptor -0.83%

Tidehunter might be another case of the average pub player not utilizing the 6.81 changes effectively.  Organized teams with actual ancient stacking strategies might get more out of it, but I am surprised that the damage reduction buff alone isn’t making much of a splash.

Disruptor and Lion are both down off of buffs.  They weren’t major buffs, but I would still assume this to be noise until a more substantial trend is proven.

While writing all this, Dotabuff released their actual 6.81 table, happily invalidating all of my effort.  I’m going to include the 6.80 vs 6.81 table at the bottom, but here are a few heroes with large win rate shifts despite no direct patch changes:

  • Pudge +3.95%
  • Storm Spirit -1.40%
  • Death Prophet -1.56%
  • Sand King -1.65%

I have no idea what’s driving these shifts, particularly that absurdly high Pudge performance.  Maybe something is going on with the 6.81 mid meta, but for now it’s anyone’s guess.

 

Advertisements

3 Responses to 6.81 Hero Shifts: Day 1

  1. pzkw says:

    The WK change is most easily explained by the change shifting perception of him away from what he’s a good at (supporting) towards a category he’s still sub-par at (carrying).

    There’s no refuting that he has a godly stun at early levels, enough to be the envy of most roamers, what’s much less clear is how useful his aura and crit actually are for carrying since he’s generally even easier than Naix to kite and doesn’t have an inbuilt BKB to help compensate and allow a racecar build and the aura work on WK’s carry whether it’s him or not. The ultimate as well is not well understood in terms of it’s effect on his carrying. To get the most out of it, you want to position in a way that it’s going to trigger that ungodly slow on a lot of people at a time when your team can maximise counterinitiation, but that means he’s uniquely unable to actually add his farm into that portion of the battle. Using it as a straight aegis is vulnerable to lots of easy and popular counters like EMP and Necrobooks.

    My gut feeling is that his winrate is driven by games where he isn’t played as a traditional position 1, either playing as a very aggressive early support who bounces back off easy farm or as a safelane farmer who goes into items that foster early aggression like blink. In this context, something that encourages a lot of players to play him as a straight carry as well as encourage a lot of players who don’t really understand the hero to pick him up would undermine his winrate.

  2. Phoenix is a he, not a she. Please research.

  3. ClothesOptional says:

    All you need to do is check out the matchups page on Dotabuff to see what’s driving his huge increase. Some of his biggest rivals (Lycan,Naga,TB) got huge nerfs. I’m actually surprised you didn’t see this.

    The rest of the chose shifts are probably related to match ups as well, Sand King is a counter to illusion based heroes like above.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: