6.80 Win Rate Shifts: A First Look

Dotabuff updated their win rate page to support 6.80 filtering today, and you know what that means: painfully premature win rate speculation!  Even if it’s early, we can still see some clear winners and losers emerging.  Just keep in mind that some of these trends are likely illusory products of the short time frame and unusual matchmaking conditions, and on the flip side of things, there may be real trends in the patch that are getting lost in the aggregate.  I hope to have a N/H/VH breakdown in no more than two weeks that might shed further light on this, along with some Terrorblade and Phoenix stuff early next week.  Moving on…

The Winners

DotaBuff680Winners

The biggest boost of the patch goes to Meepo unsurprisingly.  6.80 expands on the buffs he received in 6.79 by boosting his ult schedule up one more level at all ranks to 3/10/17, as well as changing it to provide a significant reduction to his revive timer.  The benefits he gets from this are straightforward and obvious, but it remains to be seen to what extent this boost will  into success in higher skilled games where he has traditionally struggled.

And while we’re here, I’d like to point out that Dotabuff’s method for differentiating 6.79 and 6.80 games may be systematically underestimating the degree of change between the two patches.  On Meepo’s personal page, his win rates for 1/30 and 1/31 are 48.93% and 47.90%, which must average out significantly higher than his listed 46.90% in 6.80.  Additionally, on the 6.79c listing both Terrorblade and Phoenix have games, which suggests to me that Dotabuff’s boundaries between the patches may be fuzzy.

Coming in next is Axe who saw both a sizable HP Regen buff as well as boosts to his Berserker’s Call ability.  Well come back to this again later, but Base HP Regen is pretty big boost, especially when it comes to public win rates.  I don’t feel that the cooldown improvement on Berserker’s Call is terribly significant, but adding half a second of disable to the first rank of the skill will make his early CC much more potent.  Like Meepo though, Axe is a hero whose win rates have traditionally trended downward in the stronger skill brackets, so we’ll want to look at individual bracket returns before we draw any conclusions on how significant these changes are.

Faceless Void is an interesting case.  He received two fairly mundane buffs.  The first of these is a cooldown reduction on Chronosphere, which is of course nice, but it is late-loaded and provides zero benefit at level 6.  Generally buffs and nerfs that make a large change in a hero’s overall win rate come into play earlier in the game, so my suspicion is that this change isn’t driving the numbers that we’re seeing.  Time Walk also saw a significant reduction in its mana cost, going from 120 to 90.  This should allow Void to use the skill more liberally in the early game and is probably a bigger win rate driver than the cooldown change.  But the real X-factor in these changes is the buff that grants Faceless Void 1000 movement speed inside of Chronospheres.  How much this impacts the average success of a Chronosphere is just straight up hard to estimate.

Whatever the case, I’ve been less pessimistic than average towards the hero in the past, and these changes have a shot at turning him into a sleeper pick in the upcoming months.  I’d compare him to Naga Siren in that his value as a pick depends greatly on how practiced your team is at setting up and capitalizing on big ults.  The one thing holding him back is an outdated adherence towards treating him exclusively as a late-game, 4-protect-1 pick, when Chronosphere allows Void to contribute as early as 6.

By contrast, Earthshaker is an incredibly simple case.  In 6.80 he moves faster and has higher strength gain.  And while 6.80 Faceless Void might have potential to be a dark horse selection in competitive play, more Earthshaker is almost a guarantee.  Oh, and manaless Blink Dagger should probably be mentioned too.

Sniper is another straightforward buff with Take Aim providing x*20 extra range per rank of the ability.  I still strongly suspect that my Sniper Skill Build Analysis will hold up, but I’ll redo the test once we’re a little further into the patch.

If you were wondering where all the +1 Armor buffs were in this patch, it turns out that Sven is hogging them all.  +3 Base Armor is a rather non-trivial amount of effective HP at all phases on the game, especially on a hero tied for the 14th highest starting Strength, the 22nd highest Strength growth, and one of the biggest motivations to stack strength in his ultimate God’s Strength.  It wasn’t that long ago that Sven was (briefly) the terror of competitive play, and maybe this is big enough to get him back in the spotlight.  Oh and Warcry has a slightly shorter cooldown at all ranks.

Continuing the theme of simple changes, Riki’s only buff was .75 more HP Regen.  I mentioned during Axe that we’d come back to the importance of HP Regen, but this wasn’t actually what I had in mind.  In any case, the change adds a bit more stability to Riki’s somewhat fragile laning phase.  Will it be enough to turn him into something more than a situational pick/Wisp tether receiver?  Probably not, but it’s still a significant boost.

Lina being this high is a bit of a surprise to me.  Her two major changes were increased duration and movespeed on her passive Fiery Soul.  My thoughts were that a buff solely to Fiery Soul would be more negligible due to how much priority her other abilities have, but perhaps I have misjudged.  In any case, it’s an intriguing direction for Lina buffs given how unique of a mechanic Fiery Soul is for her, and perhaps a Lina SBA is in order for the new patch.

Ruining the trend of simple patch changes, Troll Warlord’s changelog is a bit of a doozy.  Both changes were to his passive Fervor which now has a flat number of stacks per level and a scaling amount of attack speed per stack.  To try to quantify things a bit, in previous patches at rank 4 of Fervor he could achieve 40 increased attack speed after 2 attacks on the same target.  Now he can achieve 64 attack speed, but it requires 4 attacks so the ramp up time is doubled.  At lower ranks its a mixed blessing, but ranks 3 and 4 were undeniably buffed.  It could be that the late game benefit outweighs anything he might have lost due to the increased ramp up time of early ranks.  An alternate, and perhaps not competing position is that the early ranks of Fervor mattered at least as much for killing neutrals as they did for killing players, and for killing neutrals the added ramp up time is negligible.

And closing this top 10 list,  we arrive at Sand King.  Sand King’s lone change is a sizable radius boost on Sand Storm at lower ranks of the skill.  To tell the truth, I’m not especially well-versed on the implications of this.  My understanding is it allows him to jungle more effectively at early levels.  If my understanding is correct, then we could easily see more Sand King play as the jungling support in a traditional defensive trilane, like the old standard of Sand King/Leshrac or perhaps combining with the new and improved Lina.  Also free Blink Dagger.

The Losers

DotaBuff680Losers

Earth Spirit being a distinct #1 on this list should surprise no one.  He received a sizable list of changes that I will not go into here, but suffice to say most of them were nerfs.  As I mentioned in my blurb in the TeamLiquid Patch Impressions, Earth Spirit’s 6.79 win rate increase from Normal to Very High was the largest I’ve ever seen.  From what little I’ve seen, that success disparity still exists in 6.80, so the hero may be significantly stronger than that sub-40% win rate suggests.

On the competitive side of things, Alchemist took a beating this patch, likely as a result of his top 10 finish on the list of most commonly picked heroes in competitive games and his top finish in win rate among those 10 heroes.  The loss of night vision distance is more important than you would think (for a fairly liberal definition of the word ‘you,’ anyway), but the big driver here was likely the hits to Unstable Concoction.  Max damage Concoctions should be much more difficult to land reliably, but that’s a fair balance direction given Alch’s dominance during this last patch period.

Moving back to pub, Broodmother’s 6.79 rework is getting re-reworked before she ever sees the light of day in Captain’s Mode.  She was doing well in Very High games, possibly not absurdly well, but this nerf was more driven by mass dislike of the implementation of her 6.79 buffs than balance concerns.  I see these changes as a back to the drawing board moment for the hero, and in the meanwhile a version’s available that’s way less likely to make her opponents hate life.

When I was talking about the importance of HP Regen, Venomancer was who I had in mind who lost .5 in this patch.  Sure, he also lost 5 base damage, but I’m inclined to believe that Poison Sting makes up for that in a lot of scenarios.  He’s still very much a usable hero, but 6.80 is noticeable downgrade in all scenarios.

I’m not even going to try to summarize the changes Legion Commander saw to Moment of Courage this patch.  I don’t see this as a balance change so much as it’s trying to improve the implementation of MoC in the new engine.  If the new version proves to be a nerf, numbers can be tweaked in a future patch once the behavior of MoC has stabilized.

Visage is no stranger to comp-driven nerfs, and 6.80 brings him two new ones in the loss of 5 movement speed and increased mana cost on his slow.  In exchange, his familiars get true magic immunity, but it’s still a net nerf.  Regardless, I suspect he’ll weather this well enough, as he ate nerfs in both 6.79 and 6.79c and still was the third most picked hero in the patch while putting up a 52.5% win rate.

Timbersaw loses 3 base strength.  It shouldn’t cripple him in general, but I’m inclined to believe that he may have already been a bit overrepresented in comp play, so this might be harsher than what Visage is dealing with.

In another straightforward nerf, Outworld Devourer lost 50 range on Astral Imprisonment.  It may not seem like much, but AI is the cornerstone of Outworld’s mid domination, so it’s not surprising that even a relatively small range nerf can noticeably shift the outcome of his laning matchups.

As the most picked hero in 6.79 by a sizable 180 game margin, a Crystal Maiden nerf was almost inevitable, and to be honest, she made out pretty well all things considered.  A loss of 3 base intelligence hurts, but from a mana perspective her Brilliance Aura softens a lot of the blow.  I would not significantly downgrade her viability over this.

And finally we have Death Prophet.  In 6.79 she became the new Warlock of pub games, and her 6.80 nerfs were targeted at reducing the impact of Exorcism by increasing its cooldown by 15 seconds.  I believe though that her recent changes are an attempt to put her into consideration as a competitively viable hero, so in exchange for her nerf she got an increased area of effect on her Silence.  When it comes to pub play, Exorcism is the big driver for Death Prophet’s win rate so it’s no surprise that 6.80 is a net loss on that front, but in more highly skilled and organized environments the outlook for DP in 6.80 is a lot less clear.  It still seems doubtful that she’ll be picked up with any regularity, but it’s certainly not for lack of trying.

And that about wraps things up.  For anyone wanting to look through the results themselves, I’ve sloppily converted it to a google spreadsheet.  Altogether it’s not much, but keep on the lookout for further patch delving next week.

Advertisements

4 Responses to 6.80 Win Rate Shifts: A First Look

  1. Pantalaimon says:

    Thanks for the winrate shift analysis, always interesting to see how things progress in the early stages :)

    Can we expect any analysis of AD mode in the future? I was thinking that it might be interesting to see how people prioritise picking abilities, and/or how sucessful abilities are (as opposed to heroes). This mode would seem to give an interesting chance to focus on heroes or abilities as separate entities rather than innately tied together.

    • phantasmal says:

      Yes, there will be AD analysis.

      Also, don’t undersell how interesting hero success is in the mode either. They may be random, but they’re not all equal.

      • BLUEPOWERVAN says:

        I am actually super psyched to look at the hero success — As a late picking weaver I found out just how much the hero depends on sukuchi for success. Pretty much all the escape skills picked up before I could pick — ended up picking a lot of spells and realizing not only didn’t I have the hp, movement or attack speed to lane well, I also didn’t have the mana to stun people to get kills or escape either.

        I think this can shed some light on otherwise hard to discern balance phenomena such as literally unkillable +1 armour buffs.

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: