GosuGamers WarCraft

Membership

Forgot password?

Navigation

News

Replays

Forum

GosuBet

Rankings

Blogs

Latest WarCraft 3 VODs

BlizzCon Europe 2010 - Winners Ceremony

7 months ago

BlizzCon Europe 2010 - ThorZaiN interview

7 months ago

BlizzCon Europe 2010 - DeMusliM interview

7 months ago

» View all VODs (3)

Search

Blogs \ The Metagame of 6.67c

Kjervar @ 12th August 2010 08:10 (Read 14585 times).

I've never done a blog before so this will be quite bad, but I am going to describe how successful lineups in the current stable version all follow a certain 'skeleton'.

I like to think of this kind of a lineup as a staircase strategy where you work your way through the stages of the game trying to survive each stage till you get to the point where you have the biggest edge over the other team and then trying to finish the game. To describe this way of thinking I'm going to attempt to categorize (oh boy...) the relevant heroes into different stages of dominance (that is when is that hero most effective). Obviously this is very subjective but I'll do my best also, since some heroes can be given various levels of farm and therefore made important in different stages of the game I will list the same hero several times since it really depends on farm in what stage it dominates.

Before I list the heroes I'm going to define each tier:

Tier 5 Early game dominance can survive whole game with out ever getting time to farm. Never ever stops to last hit, constantly roams the map creating ganks, counter ganking or babysitting. All of these heroes can fit into other tiers as well. A few versions ago one of these was considered vital, however, I think the metagame has evolved past this and now every hero should get some time to farm mainly because the money needed to make a hero ten times more difficult to kill can be farmed in ~5 minutes.

Tier 4 This tier is the lowest tier used in current meta it's basically tier 5 but only goes to gank/counter gank if there is a real need, if not, then the real focus is farming. This is a really ill defined role but I think it is necessary to separate it from tier 5. If your team wins all of its lanes then this tier is essentially tier 3

Tier 3 Most common and biggest tier, most teams pick from the same few supporters and same few carries but then its the middle part that defines the lineup. This tier gets to free farm until it has its core and maybe a little beyond that.

Tier 2 This would be your 'Semi-carry' however that term is used over any hero with expensive items so I'm going to keep calling it tier 2. These heroes get all the farm they want so long as there is a free lane for the tier 1, they also get the less save lanes to farm since these heroes normally have some sort of a escape trick that hard carries just don't have.

Tier 1 These are the ZSMJ heroes, these heroes get all the love and are always the ones that battle it out at the end.

Now this isn't anything new, everyone thinks of heroes in some sort of early-mid-late kind of fashion but the good thing about numbering it is calculating what kind of lineup you are running and trying to make it as versatile as you can while still keeping it full of synergy and randomness :)

This is the way I categorized the more relevant heroes you might do it differently it doesn't really matter the point is that after you have done that you will have a lot easier time drafting in CM.

5 Tier
VS, CM, THD, WD, Lion, Veno, Dazzle, Enigma?,

4 Tier
VS, CM, THD, WD, Lion, Veno, Dazzle, Enigma, Warlock, Windrunner?, Clockwerk, Puck?, Chen, Rhasta, Visage, NA, Seer,

3 Tier
WD?, Lion, Veno?, Enigma, Warlock, Windrunner, Clockwerk, Puck, Chen, Rhasta, Visage, NA, Seer, Batrider, Weaver, Jugger?, Beastmaster, Kunkka?, Brewmaster, ES, Alchemist, SK, Tide, Slardar, Sven, TC?, Storm, Krobelus, Pugna, Mirana, SF, Necrolyte,

2 Tier
Kunkka, Beastmaster, Sven, Alchemist, Slardar, Jugger, Doom, Morphling, Mirana, Weaver, SF, Storm, Windrunner, Necrolyte, Harbinger?, Razor, TA,

1 Tier
Kunkka?, Sven, Alchemist, Doom?, Slardar, Morphling, PL, Naga Siren, Jugger?, Medusa, Drow, Razor?, TA?, Windrunner?, Storm?, Harbinger, Necrolyte?,

The '?' means that putting it that high/low is pushing it a little and I didn't only listed important/unique heroes the other ones are all comparable to at least one of the heroes I listed.

Again if you would do this stage differently then that's fine since the only reason to do this at all is to use for drafting

I came up with this thing (no idea if it's been done before) by going over the ESWC ehome replays. They always drafted the same pattern even though the used different heroes. They made a staircase like 4-3-3-2-1 or 5-4-3-2-1 with drow or medusa at the top normally and what they did was the used their key heroes at each stage to defend the higher tiers till they found the stage where their advantage was big enough for them to press it and win the game, and if the game ran late then they had their big carry which got all the time in the world to farm while all of the other heroes where looking for fights.

So this form of thinking works, the question is what causes it not to work?
The answer to that is: heroes that belong to multiple tiers. Those pesky heroes like Lion make your lineup a lot more versatile, the reason asians pick all ranged heroes is they want to be able to decide the laning based on if the want their lion (or some other hero) to be a tier 3 or a tier 4. That's why great teams are able to pick a ganking sniper because they see that if they play their sniper into the game earlier instead of farming with it they can gain a bigger advantage than the farmed sniper would've given them in the late game.

On to calculating your lineups strength... you want the sum of your heroes' tiers to be between 13 and 17 where lower numbers = more turtlin'. It's easier to win in late game so something higher than 17 is to risky, while under 13 will cause you to have more late game than is necessary and will most likely result in a loss due to outpick. (if your opponent picks more than a 17 you will just go for a lineup that is rather mid ranged and win him earlier...)

When evaluating your opponents strength consider his laning options and try to lane your lanes so they counter his (for example SK > Sven in lane...) and also try to do it so that the farm your heroes are getting gets them their items at right times (for example ehomes lion got solo mid to get its dagger, that strategy would fail if they wanted to roam it AND rush a dagger...).

To recap:
1. You want to win your lanes by counter you opponents lanes.
2. You want to keep your teams sum equal to or higher than his so that you don't lose the late game (unless you plan on a push setup or they have way to much late game power)
3. You want to try to keep as much information about your plans as possible from your enemy. Select heroes that belong to many tiers and have many uses and synergies as early picks (Windrunner is a good example).
4. You want to have at least one hero for each stage of the game so that you have no weaknesses
I should add that if you have before hand decided that you want to win the late game and pick PL then ban 4 of the most powerful and popular late gamers and then pick a lot of tier 3's and 4's + your PL.

That is the way I think you should be thinking while drafting and it can be seen in recent replays that the Asian teams are thinking in some system similar to this one they sometimes go with very unorthodox laning options to make sure that the farm in the laning stage goes to correct heroes. It really goes to show that picking versatile heroes and controlling the game by distributing farm correctly is more important than picking powerful heroes. (By that I mean every pick is viable given the right circumstances)

In order to understand what I'm on about you have to go through the tournament replays from the past month and note what heroes are most consistently picked/banned and the context they are picked in (first or last pick? are they always laned the same way? are there any common combos that are often picked?).

I was going to go into more uses for this way of analyzing the picking phase of the game and more in depth about the current meta (analyzing picks and laning choices in replays and then reason why the matches turned out the way the did). Sadly I don't have time and because this wall of text is rather hard to understand already and because the people that get what I'm trying to say had probably already figured everything out I've said here (especially since this is rather obvious stuff) then putting more effort into this enormous wall would be wasted.

So thank you for reading if you had the patience and please share your views and opinions, also if you have the troll gene feel free to put a billion spoilers into one another that end in a tl;dr...

Comments \ Paging

1, 2

comments

#51

leRage 1 year ago

Nice work but it's not enough to describe the action, every picker has an unique strategy (well, except copy-cats), every picker can get outpicked and it's not about numbers, since most heroes can undertake many roles nowadays.
As an overall comment, nice job.

#52

lazyd00d 1 year ago

@kjervar: Lion is a male because he has a beard.

#53

cookiemonster 1 year ago

gj writing this blog. very well written. i think warlock is strong in this version.

#54

YourThe_rapist 1 year ago

rofl female lion
support heros are for girls (maiden, venge)
therefore lion = girl?

SEXIST

anyways the keyword here is VERSATILITY

#55

udm 1 year ago

it's a good read though as you said it's pretty obvious for ppl who watch chinese replays and have some idea about dota in general

also i disagree on some of your heroes in the tiers (especially in tier 1)

oh and yes there are male witches too :D

#56

littech 1 year ago

#18 +1

#57

Jules_Winnfield 1 year ago

Hjörvar "kjervar" Logi got rewarded 250 points for this blog entry!

#58

ztoa03 1 year ago

Thanks for making me understand the line-ups for 6.67c better... This is really a great read.

#59

orly 1 year ago

you should write more articles. It's analysis like this that truly represents what dota should be focused on

#60

brasi 1 year ago

Good post.
Although i dont see anything special in here.
People like numbers (easier rating of statements), Lion(as if he wasnt the same for ages) and Ehome (asian dota) in general.
Had a good read, but numbers for Tiers in Dota has been done thousand times and no rating has ever be the chance to decipher teams drafting and gameplay in a sufficient way exactly.

#61

mohib 1 year ago

wow nice read!

#62

JonnyL 1 year ago

so picking strong supports first is new to doters? interesting

#63

The_4Lb1N0S 1 year ago

*Facepalm*
Awesome!!!
But you forget "visage?" in a lower tier imo, and sven shouldn't go in tier 5 as he isn't born to farm all the time (unlike medusa)
At least you finally told why asians rape europeans: they understood this metagame, they understood dota, and european's strategies are not going in the good way.

#64

levkan 1 year ago

U JUST BROKE THE DOTA SYSTEM !! ! ! !

#65

Sets 1 year ago

Great blog man. It's a nice foundation which people can analyze more details into this game. Thumbs up to you.

#66

kut3b0y96 1 year ago

epic read, gj man

#67

sabuk13 1 year ago

hey man it ws gr8 luvd iy

#68

kjervar 1 year ago

^ I put a great deal of work into deciphering what you said, honestly I've spent 17 days literally trying to think of every possible meaning that sentence could have and I finally worked it out.

You said you liked my enormous wall of text to which I will reply: Thank you!

Other than that I'm pretty sure it's time for this thing to disappear somewhere deep into the vast randomness that is the internet, never to be seen (and consequently read) again.