Developer's Blog

Remember me ?             Register

Drafting in DotA -cm


- This guide does not cover the very basics! If you play on a very casual level or simply are new to the game, there are other aspects you should focus on, which are not in this guide.
- Aspects are left out. If people wan't me to write more, I will.
- Visuals are kept simple and maybe not as fancy as you want it.
- Grammar is far from perfect.
- Walls of text. No way around it considering the subject. You can put on some music.
- This guide does not discuss the aspect of acting as a captain ingame.
- This guide discusses the -cm mode and not the -cd mode.

Table of Content

Changelog (started 22 August):

22 August:
- Added minor sentences all around
- Added example for the chapter "Casual level drafting"
- Added example and explanation for the lanecombination 0-1-4 in the chapter "The current metagame"
- Added example and explanation for the section about choice of lanes in the chapter "The current metagame"

24 August:
- Minor edits to sentences all around
- Added two scenarios in the final chapter

27 August:
- Added link and description to the section about core heroes in the chapter "The current metagame"
- Scenario layout issue fixed in the last chapter

6 September:
- Scenario 3 written

13 October:
- New scenarios in the last section!
- Minor adds all around.


I’ve decided to write this guide mainly because I feel that I can contribute on this subject, but also to learn in the process. Back when I began to draft, I found it very hard to find useful information from the people, who actually knew, what they were talking about. No guides were to be found etc., and I hope this will help at least a few people out there. The title is drafting for intermediate, since it definitely isn't for beginners. Whether it is for intermediate or advanced, I frankly don't know, and I'd like to spare myself of the "this is far too simple for advanced"- comments. I could only find one published guide to drafting on PD, and I felt that it was outdated and flawed. My knowledge picking-wise is limited for sure, and I still have much to learn. Furthermore, this is only my take on the drafting phase. However, I hope that you will all feel that you’ve learned a thing or two regardless of your level of skill. I like that DotA is such a complex game strategywise, which has led me to analyze a lot while both playing, picking and watching, and I feel that I have many valid conclusions to draw.

"Always pick hard carry last,” I sometimes hear. I even sometimes still see people picking with the old-school mentality. However, DotA has reached a much more complex level since the 6.48b version. First of all, the range of valid heroes has grown. Especially due to the evolvement of the support role, which has helped widen the range of heroes used. Back in the day we had the 2-1-2 lanes, and 1 side-lane would typically be some stun/nuke combo, whereas the other side-lane would consist of a hard carry and a babysitter. The babysitter role is still valid for sure, but with the discovery of the tri-lane, the added support items and the buffing of jungling and pushing, there is now a ton of other ways to play the support heroes, not to mention DotA in general, but I’ll get back to that. The reason I will be using the timespan from 6.48b till now to explain certain things is simply that it is the time I started playing competetive.

Rock/paper/scissors aspect:

I came to the conclusion that it would be better to let you read the entire article instead of a summary. First of all, it is already explained for everyone to understand, and the article is fairly short. It is an interesting point of view indeed, but it should only be kept in the back of your head, considering the fact that DotA obviously isn't as simple, as it is put. To understand some of my theory later on, you will have to read this.
A Look at the Rock Paper Scissors of -CM

The conclusion is simply:

Gang > Push
Turtle > Gang
Push > Turtle

You will need to read the article to understand why though.

If you disagree or still do not fully understand, you should read the spoiler below. It is taken from my reply in the comments.

although i think the rock-paper-scissors-concept is flawed (i know it's not your idea but you quote it without any criticism).
i think it isn't very good, because obviously every team needs to be able to push. the time to push is obviously strategy- and lineup-dependent.
also it doesnt include counter-ganking. if a turtling-team recognizes the opponents ganking-/pushingpower and plans accordingly they can still pull of the turtling with good counterganks.
First of all, I believe I do express skeptism, when I say that DotA isn't as simple as it is put (like you say yourself), and the concept should be kept in the back of your head. The RPS-concept (let's call it that) is not flawed itself the way I see it, and I'll try to comment on your examples.

Every hero in the pool is able to push with the fact that they can all attack. I'm not sure I'm completely following what you're saying. Counter-ganking and the example you mention is outplaying the opponent the way I see it. When you pull off turtling without outplaying the opponent, it is because you are playing versus a team with lack of pushing power. They can not push unless they gain a huge advantage from ganging, and that is pretty easily avoided. You turtle because you are not stronger than the opponent lineup yet. This means that if the opponent team has pushing power, they can push (like I said, since they're stronger). When one finally understands this they'll be like "aaah".

and i think a pushing team needs to have a decent amount of early-gankpotential because otherwise their pushes will be stopped by counterganks hurting them badly.
This is true (except you can't countergang push by definition), and this is where the article may fail to convince some people, since they consider pushing five core pushers. Here is how I transmit the theory to actual gameplay:

The opponent is using a trilane with a hard carry such as sniper and therefore most likely wanting to go lategame (turtle). I would consider either of two things: The 0-1-4 or the 3-1-1 lane combination with the intention of pushing. Lets say I went for a trilane also, but a pushing one in the opposite sidelane (could be with an Enchantress or Chen), and I would be comfortable that the most they'd get out of their 3 heroes is 8 creeps a minute on Sniper. You can push tier one and two towers in 5 minutes easily, if they do not come to help the solo hero. From these two towers, I've (if lasthitted) gained a 2400 gold advantage for my team. I can afterwards pick up mid tower with ease and the rest in whichever order. The gold advantage from picking up towers is underrated by many. Leaving the Sniper to freefarm means absolutely nothing. Even your supports with no lasthits will gain more gold (atleast when you're succesful). Picking up all these towers will give you an advantage big enough to superiorly win teamfights and push rax very early in the game. I suggest you read the 0-1-4 lanecombination (to understand that pushing is not five core pushers) and initial section in "The current metagame" chapter again. Note that the sniper lineup will be weaker because:
1. Sniper is weak early-mid.
1. Their two supports will have zero gold and little xp. It will practically be 5v3.

If the team picks a gangoriented lineup with no hard carries, you can't push, if they play it right. When you are pushing, you are very vulnerable to gangs. When you want to gang, the optimal scenario is when the opponents are laning without caring. Some would say that push > gang, because they're either imagining:
1. All lanes being pushed, and the gang team having no time to gang all lanes.
If this happened, you simply were outpicked looking at each and every hero and failed in the laning phase. This would happen if you for example put an AA to solo against Broodmother and does not mean that push > gang.
1. You're against a clashpushing lineup and concluding you can't gang a clashoriented 5-man group with tanks and healers.
If this happened you again failed the very early stages of the game. If you have a ganglinup, and the opponent has a clashoriented pushing lineup, you better put atleast as many heroes at each lane as they have. Like for example if they choose to pick up one of your towers fast with 3 or 4 heroes, you better put 3 or 4 heroes there aswell to stop their push. After they've claimed the first 2-3 of your towers, the problemscenario as mentioned above will happen. Basically, you have to stop them from picking up the first towers by not ignoring their 3- or 4-lane.

Lastly the turtle > gang, which I explained, when I answered the first quote.

This simply means:
Turtle > Gang
Push > Turtle
Gang > Push

In words everyone can understand:
Two years ago the metagame was gangoriented.
Then the trilane came and it was turtleoriented.
Now it is pushoriented.
You see the pattern?
Gangoriented metagame would/could be the next step in the pattern, but not yet, because with Icefrog's changes (buffs to pushing), pushing is superior at the moment. You even see the Chinese pushing now. Ganging is still the counter to the pushing, and therefore the current metagame can be defined as gangoriented pushing or aggressive turtling like some call it, when for example a Syllabear or AM is in your lineup also. Either way, it is pushing. Don't be fooled when you go to, look at a replay and found that lineups consisted of supports and semi-carry heroes with not too much pushing power. They will still (with almost certainty) be focusing on getting those towers.

I'll say this again: When one finally understands this one'll be like "aaah". You may need to experience it ingame.

In terms of push vs. push, it has to be mentioned that clashoriented pushing with tanks and healers gets frustratedly raped by a pushing lineup with heroes such as Broodmother and Furion.

Still don't understand or agree? I'll try to put it in another way:

Focus points of the three strategies:

Ganging: Killing heroes
Turtling: Killing creeps
Pushing: Destroying towers

The ganger would have an easy time killing the pusher, since the pusher lanes so aggresively and vulnerably. He can gang all he want, and since the pusher is not better than the ganger lategame, and the pusher is not focusing on creeps, the ganger will get stronger and stronger (than the pusher). The turtler would have an easy time against the ganger because he just has to stand back and kill the creeps long enough while avoiding gangs. He can be sure that his towers will not go down, since he is against a ganger, and he basically has all the time in the world. The turtler would fail against the pusher though, since he can't stall the game, when his towers are going down without him being able to do anything, because he isn't strong until later. Remember that a characteristic mindset for the turtling strategy is to avoid clashes as long as possible.

Split pushes and "the ganglineup doesn't have time to defend" etc. That's called outsmarting, outplaying the opponent or maybe even outpicking the opponent in terms of each and every hero, and has nothing to do with this.
Some are going to say "bla bla", "this doesn't work in real play" and this and that. However, this is transmitable to actual gameplay

The current metagame:

I am going to start this chapter with a brief introduction to the history of the metagame in Europe meassuring to a timespan of approximately two years.

For a long period of time, the metagame was very gang-oriented. I can't define the exact timespan, but it was atleast from 6.48b and up until early 2010, where the Chinese changed the European metagame for good. This was with the strategy of running two pure supports and relying entirely on the carries, which also meant the comeback of the trilane - now with a new playstyle. Since then, top teams such as Na'Vi, MYM and m5 have been adapting from ganging into a more pushoriented gangstyle, and with a lot of help from Icefrog it has now proven to be superior at the moment. I can explain this further if you wan't, but it isn't that relevant for you. Therefore I'm going to take it a level down.

An overview of the current popular lane-combinations:

The trilane (1-1-3 or 3-1-1):
Theoretically, the trilane with a hard carry is the strategy you wan't to play if you are up against a gang-lineup with weaker lategame and lack of pushing power. Especially if they run the old-school 2-1-2 lanes. Practically speaking though, it requires that you know how to track dual roamers with wards and mapawareness, and also that you know how to turtle. Ofcourse, if you are playing on a very casual level, opting for a trilane alone will probably win you the game, and you won't have much to worry about. A trilane could also be used to shut down the opponent trilane with a farming hard carry or to push. You have to decide and remember why you have the trilane.

If you have it for your hard carry to freefarm, who preferably has to be able to outcarry the opponents carry at the opposite lane (if that's what they opt for). If you are not keeping the opponent's solo away from XP, and you can't push either, you need to consider whether there is a reason for the supports to stay. One or both of them could go roam or babysit other lanes. I will not get into details with this, since it is not a guide on how to play the trilane. I am sure, there are plenty of those. However, you should read the section further below regarding pressure and choices of laning. It relates to the trilane.

The dual woods/roaming (1-1-1):
Officially introduced by the the former LGD here in 2011, I believe. The best combinations of the two would definitely involve an Enchantress or a Chen - preferably partnered with a reliable stun such as Venge. Note that this strategy very easily can be played as pushoriented ganging.

When mastered, this proves to be very succesful against the other lanecombinations. First of all, since not many know of it in the casual scene, and especially since many simply don't make an effort to counter it. Against it you need to track their movement with wards and mapawareness - perhaps even blocking their creep spawns. Going trilane against these dual-roamers is usually a bad idea. Instead opt for one support protecting each lane for example. I can explain further if needed.

The one jungling and one supporting (1-1-2 or 2-1-1):
This lanecombination grew in popularity after the great buffing of jungling, and it can be played in different ways. Mostly it is used with jungler who will come to assist the dual lane ganging and pushing, but it can also be played with a farming jungler such as Lycan, who atm. can outlevel everyone in the woods if undisturbed.

Note: With the current metagame, I would not put Enigma nor DS in woods to farm.

Dual babysitting (2-1-2):
This combination is in my opinion underrated and situationally very strong. If the opponents opt for an aggresive trilane on their short lane, a strong ranged hero with an ES protecting can actually make their strategy completely useless if played correctly. At their short lane it is hard for them to pull and keep you away from xp. Also, like I said, if played correctly, they can't kill the opponent farming because of ES. The Chinese proved this vs. MYM a long time ago, where they had Medusa and ES vs. an aggressive trilane of MYM on their shortlane turning out to be completely useless.

Let's say the opposite lane is WR vs. Puck, a harrasing AA or Lich for either one can turn that lane very onesided. This way you are getting xp and farm on all 3 lanes and actually winning one of them drawing the other two farmwise.

Mid babysitting (1-2-2, 2-2-1 or 1-2-1):
A hardcarry farming mid is not evolutionary stuff (the PL/KOTL combo was the first to be used mid, I believe), but it has somewhat grown in popularity lately. Especially Lich partnered with a Spectre or an Anti Mage is often seen - and with good reason. This way you are putting a lot of pressure on the mid lane forcing their supports to assist leaving more space for your sidelanes, and you will hardly be killable considering that it is mid lane, and it is Spectre and Anti Mage. It doesn't have to be Lich, Spectre nor Anti Mage though. It has to be a strong support though, since harrasing will be uphill a lot.

Opting for this combination also often suprises and confuses inexperienced teams, and need I say that is a good thing?

Clashoriented straight push (0-1-4 or 4-1-0):
The all-in push is for many mistakenly a mid rush. There are ofcourse many ways to do this, but from my experience, this is what works. You pick a combination of four heroes, of which a few has pushingabilities. Others can be semipushers preferably.
Note: Don't pick four core pushers, since:
1. There's no need.
2. You're going to need heroes who can kill the opponents by diving towers. Pushing power is counterable.
3. The opponents will smell it and potentially prepare themselves. In the bottom of this spoiler I will explain counters to this.
4. You wan't to have some lategame aswell.
5. The point actually isn't to push the throne/tree fast. You pick up towers quickly gaining a huge gold advantage, which will let you superiorly win teamfights even when diving tier 3 tower.

Why not 5-0-0?
1. No more than four are needed.
2. You don't want to leave two lanes for freefarm.
3. The guy mid will be ready to push after the rest of the team has cleared the tier one and two towers.
4. The guy mid will be able to farm mekansm very fast.

I've found that the best way is to move to the short lane, creepskip and kill them if they try to defend, but it is pretty situational. There really is a ton of ways to do this. The other lanes will often not tp to defend, since they think freefarm is better, and who buys tp at lvl 1? It is important that the mid player pressures, so that he won't go to defend.

The idea is to pick up all six towers and ultimately have an advantage big enough to take on the rax very fast. Remember that there are no pure supports and no pure carries. You're going to need threads/mana boots and magic wand on everyone aswell as janggo, mekansm, basilius ring, urn etc. Your supports will be having a lot of survivability, while the opponent supports have got naked boots making it practically 5v3. Remember that if they try really hard to defend, and you therefore can't push - just farm up the core items and go push again.

To give you an idea, a lineup could be:

Necrolyte would be laning mid to farm up Mekansm. Some would say Furion needs the solo, but he has strong attack at lvl 1 and two treants will do just fine for the first tower. Remember that there really is so many heroes you can use for this.

Note: This is just to give you an idea of how little pushing power you need. You could even go with less. The picks would very much depend on the opponent's picks! So far, I've concluded that it is a good idea to ban heroes such as AA, Weaver, Broodmother, Lich, Clock, Lycan and even Venomancer. If possible, let the opponents ban heroes such as Invoker and Weaver.

Counters to this:
1. Counter-pushing. Heroes such as Broodmother and Furion will frustrate these clashoriented pushing lineups obviously because every time they gather up, their other lanes are being pushed.
2. Not ignoring early pushing. Say you picked a standard lineup with supports and semi-carries, put four heroes whereever they are four or three whereever they are three. Simply do not let them pick up those first towers. You have a great chance to get a lot of kills. I'll repeat this: Securing one in your team freefarm means nothing compared to the gold gain of all those towers.

I'm not going to explain further, but feel free to ask questions.

Note: It's all roughly speaking though. You need to be able to adapt in the early stages of the game, if the opponents are not doing, what you expected them to, or if for example a teammate is failing horribly. These are also just the popular ones. Sometimes you will see three mid at lvl 1 to grab a quick tower or even completely adapting in the last second.

Choices of laning:

As both Scourge and Sentinel you are faced with a mid lane, a short/hard lane and a long lane. To explain the choices of laning, I'm going to explain what the idea has been during the trilane prime, since it has pretty much transitioned into the other lanecombinations aswell.

You lane your trilane based on how strong it is compared to the opponent's trilane. This typically means that if you have a hard carry on your trilane, you probably want to be on the long lane or in other words the safe lane. Here you are closer to tower, you can easily pull, and you got the forest advantage. However the heroes you have for solo also influences the choice of lanes, but it is pretty situational. It will come with experience.

If you currently can't predict who will be on the trilane amongst the opponent's lineup, you need to watch replays and get some experience, cause it would take too long for me to list a million scenarios.

An example of the above could be these two trilanes:


It should be obvious, that Sentinel has a way stronger trilane. It would be ridiculous for them to go bottom lane, since Razor can't outcarry Spectre, if they're both left to freefarm. Also, Visage and ES can't roam as well as CM and WD should it be trilane vs solo. They basically need the action at lane. The two trilanes would most likely clash at top lane. What I've seen some people do is to predict this and put the trilane with Spectre on the hard lane, but this is not wise in my opinion.

A thing you could do though is send Spectre mid for an either 1-3-1 or 1-2-2 lane combination or simply outplay them. Remember that this example is to show you, how you know at which lane to put your heroes. Don't pay too much attention to the heroes used in the example.


As for pressure, I can't tell you what to do, but you should keep the aspect in mind. I'll give you a few examples. Scourge has trilane top and Sentinel bot. It's a farmfest, and the sidelane solos are not having a nice time. However, an important job they have is to put enough amount of pressure on the lane to make sure that the two opponent supports do not go to gang mid for example. Although, not too much pressure, since that would result in getting caught.

A trilane would sometimes want put so much pressure on the lane that the opponent supports have to come help, which leaves room for your solo at the hard lane. It's pretty situational, you just have to keep in mind that you can pressure to dictate the movement of the opponent support/roamers.

Core heroes of the current version:

Listing a bunch of heroes that are strong in this version would not be of much help. I would need to explain my thoughts on each and every hero, which would take too long. If you have any questions, feel free to leave a comment. Also, I will most likely be discussing some of the core heroes in the final chapter.

However, Lumi has already made a list of the current core heroes based on how often they are picked or banned, so I'll give you the link here. If you wan't to interprete it as to whether a hero is strong, be critical, since that is not what it's meant to show. The data from the August list was collected from June to August, and for the subsequent months (September and so forth) conclusions are and will be drawn based on the previous month. Thanks to Lumi for clearing that up.
Hero Tier List – Aug 2011
Hero Tier List – Sep 2011
Hero Tier List – Oct 2011
Thanks to Yagachi for providing link.

Now back to the overall metagame. An aspect I think many are missing out on is how much this version is designed for pushing and jungling. Back in the day your towers would be invulnerable, when you teleported to them, and you could even have like four chickens teleporting, which means that you could turtle your ass off. This has been removed obviously, but also a ton of other things. At the top of my head I can mention the added cooldown on buybacks, the added pushing items, the buffing of core push heroes etc.
I read a great article on some time ago, which summed it all up, and I will see if I can track it down. Basically what you need to understand is that the current version is the most push-appealing there has ever been, and the top teams have explored that. There's also the aspect of the fact that the metagame has been turtling with the trilane era, which makes the pushing even more attractive. I'll say this again though. Don’t get me wrong. Pushing is not a combination of five ultimate pushing heroes with no synergy, who gather mid at level 1 to storm the opponents, like some think, I believe. For now I think you should listen to these two videos:

The first is Dendi from the team Na’Vi known to be great at pushing, and after his Invoker show-off he explains to us what the deal is with this aggressive playstyle they got going on.

So, when asked about his team and their pushing, he gives us this gold:

Dendi: Okay, I’ll try to explain. People think that it’s pushing, but that is not really what it is. If you get tower, you get advantage, and I think the right way to play DotA is to pick up towers. If you can pick up towers, you get advantage. So if you take heroes that let you push six towers, you still have lategame, and you defend your six towers, you always win, because you have gold advantage. Lategame is very random. When you go lategame, you can lose any moment. Any mistake - small mistake. What is the point of playing lategame, if you can finish them in twenty minutes? You risk. Sixty minutes and boom, you lose the game. I think it is really not good at the moment.

Some people refer to this strategy as aggressive turtling, when you have carries such as Syllabear or AM in the lineup. Either way, it is pushing though.

The next video is Lumi, who gives us an analysis of the newest version. You should especially take note in the jungling he talks about.

Like Lumi also discusses, the disappearance of Shadow Fiend pretty much symbolizes the game play of the current version, especially with the adding of Smoke of Deceit, and invention of the popular dual jungle/-roaming. Feel free to ask questions, and I’ll elaborate.

Maelk: "A few versions back, defensive DotA and in particular the utterly boring '5-man train'-style was very dominant, but nowadays it's rewarded being aggressive and pushing."
But knowing this, do we just go with the flow or is there a way to beat the metagame? First of all, there are a lot of ways to push. The metagame is very versatile in my opinion. Typically people think to counter push with mass antipush waveclearing heroes and turtling like there's no tomorrow. But as Na'Vi points out in several interviews, one would have atleast some lategame to back it up, and the pushing of towers are just for the gold advantage. Also, for the clashpushing the usual idea is to have little pushing power, but strong heroes such as Omni, Balanar, Viper or Slardar, so that they can towerdive you, kill and then push. Knowing this, creepclearing heroes are still a bit of a pinpoint to countering pushing, but there are other aspects to consider in terms of which sort of pushing they will be doing, and how you smell it in the first phase.

To first establish a key element, a theory on one aspect goes something like this: The current metagame is very push- & gangoriented, which means that towers will fall, and the games will finish relatively early giving the carry less time to farm. This should lead to a conclusion that you shouldn't rely on the once very popular trilane with a hard carry. There is simply not time and space for such a hero to reach its' potential, which is where semi-carries such as Lycan or Weaver come into the picture.

So generelly, don't pick hard carries. Anti-Mage, Spectre and perhaps Syllabear can be exceptions for their superiority. Broodmother for it's counterpushing capabilities.

Taking the rock/paper/scissor aspect into consideration, people are maybe trying to counterpick with the wrong approach. Typically people are thinking how do we defend, but think about this for a second. When you want to gang, heroes are typically hiding or caring, but here your opponents has or chooses to stand on lane big parts of the game very vulnerable to gangs. Practically speaking though, it comes down to each and every pick.

Regarding Na'Vi and aggressive pushing:
Tobi: "It basically comes down to the skill of the players."
This quote probably wasn't needed, but a great thing about this metagame is that everyone plays an important part. Teamwork really is key.

In continuation of that, another aspect in regards to the metagame is the role of the support. I've been told that it was NS who said it and then afterwards that it was Fear, who said it. Either way, it does not change the relevance. It isn't an exact quote, but something alongside this was said:

NS or Fear: The role of the support has evolved. It is simply not enough being the ward bitch anymore. You have to make an impact on the game, and you can.
This was especially with the Smoke of Deceit being added.

Improving and keeping up with the scenes:

Experience is monstrously important, and a captain is always going to make mistakes, which is okay. You learn far more from mistakes, which cost you the game than from just running over your opponents. It is very important that you realize and admit to these mistakes. Typically captains love their own picks and blame the team or individual skill, but this is bad. If you truly were right, and the opponents were outpicked, maybe consider other aspects. Yes, maybe they were outpicked, if the level of play in the game was different (read the section: Casual level drafting). In other words, what would’ve been the best picks for these players.

If you want to be good at drafting, you have to keep up with the scenes. Maybe you want to invent your own picks and not “copy the pros”, which is totally fine. However, you need to know what you’re up against. What could be coming, how heroes typically are played, and of course what you need to counter. The pros probably spend more hours playing DotA than you, and it is therefore also totally fine to copy their picks. But to do so, you need to understand the mentality behind the picks. This is in my opinion pretty simple and obvious stuff, but I can explain further, if you want. I felt, I had to write this section, even though it’s not very new and advanced. Personally I use GosuGamers DotA | Replays a lot, but beware of the forum. It is filled with quite a lot of trolls and some inexperienced players who give slightly unintelligent advice. The same goes for the strategic forum on, I'm afraid. While some advice given here is valid, a lot of people are handing out advice based on way too quickly drawn conclusions. I personally can't read through a topic in the strategic forum without getting a bit annoyed. However, it is better than no advice and people wanting to help should never be put down. Just be critical.

Knowing your team:

Obvious, yet overlooked. Before showing off your mastermind or anything else - know your team. I still see people out there not taking into account what people on the team can actually play. Remember that the point of picking well is so that you can ultimately win the game. Picking something that looks good and makes you proud of yourself does not make you a good drafter, if the picks are very unsuited for the team, and the game is lost. This goes for both newly formed teams and drafting for unknown players.

Regarding whether or not to communicate while picking, there are definitely pros and cons, while it also varys from team to team obviously. If several people on the team are good drafters, great things can be thought out together, but if your team rarely takes an interest in drafting and are shouting out stupid suggestions - considering them for the sake of being nice is not being a good captain. You are simply wasting valuable time and letting them mess with your head. Note that this is only my opinion.

Casual level drafting:

People tend to copy their picking-mentality when drafting for their team and paste it in whatever inhouse league or public game they’re playing. This is wrong, and I’ll try to explain.

First of all, which may be obvious to many of you, some heroes have more potential than others, but also need more skill to be played. I feel, I need to make this clear, so I will give you a rough example. Queen of Pain, Chen, Lina etc. are heroes with very much potential, but needs skill to be succesfully useful. Other heroes such as Sven, Naix, Leoric, Barathrum and perhaps Lycan are to reach a level of "usefulness" with less effort, but are also more limited in terms of what they're capable of. How is this valuable information? When you pick for your team, who assumingly have better individual skill than a random league or pub game, a team of paper heroes with great potential is good, because the players can use its’ potential. When you pick for a team, who don’t know each other and probably aren’t very good skill-wise, you need to pick heroes with potential that they can fulfil.

Practically speaking:


The thesis above can very well be true, if the level of skill in the game is too low. Knowing this, you can pick your way to many wins in public leagues etc. This is just a rough example - don't pay too much attention to each individual pick.

In public games at low level, the team of strength heroes will almost always win. You will be shocked to see a list of public games and then notice that somwhat 70 % of the winning teams have more strength heroes than the opponents. With such heroes there is room for mistakes, and they are therefore not very hard to play. Also consider other aspects. You need lategame to fall back on, considering you can’t expect the pub players to push when they have the advantage, or gang like your lineup may suggest. Then ofcourse there are also the aspects of picking heroes who are easy to farm up, focusing on strong lanes etc.

Practical examples, observations and pointers:

It is time for some practical examples you can relate to. I am going to make these scenarios in which I will justify why I am doing, what I am doing as well as write random thoughts of the process. Please keep in mind that the observations are not explained in details, and you may think to yourself "this is not advanced thinking..", but I didn't want to write 2000 words to every scenario. I hope you're satisfied with what is written.

It seems that some people would just like a list of counters, so I'll consider making that.

Scenario 1:

Picks- . -

Before starting this game we decided that we would just play a standard game, which means that we wouldn’t try to counter turtle with push etc. We would be picking as we go along, and there is nothing special to note about the first bans and picks.

First phase: Unless we are pushing, I am almost always banning or fast picking Spectre, since he is really strong at the moment and put this pressure on the opponent team. Invoker, Weaver and perhaps SD are very common bans, since they’re strong, but mostly really annoying to be facing. Lich and Doom are situational bans, but there’s no reason to put too much thought into those, since most people just ban them for being strong.

I start off with an AA pick, which doesn’t need much justification. I don’t want to reveal too much of my lanes yet. They opt for CM and Batrider. Two heroes, which leave them open for anything as well. From what I’ve seen, Batrider is a very hot pick/ban in Asia, whereas it is situational in Europe. However, these people probably just picked it, because they like it. Spectre, Weaver and Lich are out of the pool, and I feel a good pick here is Anti-Mage. They’ve picked two intelligence heroes and especially Batrider, which Anti-Mage does just fine against despite the long disable from Batrider. I pick up ES as well, which leaves me open for AA solo, trilane with AA (could be both defensive for farm at long lane or aggressively at short lane depending on their picks). Even AM mid with support to counter Batrider. The pressure is on them now, and they finish the first phase with a Tidehunter pick for clash power. What’s funny about this pick is that it tells me with 99 % certainty that they will trilane.

Second phase: I banned Balanar and Syllabear, to ban some of the carries, which would give me a hard time. Besides, the two are very strong heroes at the moment regardless of the Syllabear nerf. I consider banning Puck – but then again, I have Anti-Mage. A few teams would’ve banned Visage in their spot, but these guys didn’t – most likely because of the great nerf he was hit with after 69c and the fact that the pros do not pick him anymore. He is still viable (in my opinion) though in the “not pro”-scene, since they’re filled with old school trilanes. They ban Venge and Windrunner mainly to limit me lanewise, I’d guess.

I would’ve liked to last pick in the second phase, since it is tough to say, what they want to play. However, I grab Necrolyte because they don’t have many nukes (also of course for the tankiness of Batrider and Tide), and I figure that I will just put him against whatever side solo they will get. They follow up with Void, which makes it fairly obvious that they will trilane at their long lane. Therefore I decide to grab my Visage, and they finish off with Mirana for their off lane making the lanes tri vs. tri top, AA vs. Batrider mid and Necrolyte vs. Mirana bot.

Scenario 2:

Picks- . -

First phase: As for the bans, I prefer to play aggressive lineups, and therefore I often ban heroes such as Weaver, AA and WR. The opponent bans are fairly standard.

However, they start off with a Spectre pick. On one hand, this sucks. Spectre is very strong, and now the pressure is on me, since they will pretty much auto win if the game is played evenly the first 30 minutes. On the other hand, he is welcoming me to push, so that’s what I’ll do. I grab Balanar and Furion, which is one way to work your way towards clash pushing without revealing too much. As for their ES pick, it doesn’t matter too much, unless he manages to get dagger. For that to happen we would have to fail though.

Second phase: I ban Broodmother, since counter pushing would be devastating to this clash pushing. Also, I have close to no AoE-spells. Clockwork is also a bit of a threat to the 4-lane, since he can “join” the fight with rockets and still farm and leech exp at another lane. They ban Enchantress and Venomancer, because they obviously smell a bit of a push, but these heroes were not in consideration anyway.

They pick Puck, which shouldn’t need much justification, and I grab Tidehunter for some clash power. Especially his ultimate will be a problem for all of their weak heroes. They finish with Obsidian, since I’ve only got one intelligence hero. This could be a problem, but if we get to a point in the game, where Obsidian will rape us, Spectre would’ve already beaten him to it. Omniknight is picked regardless of not being intelligence, but his Repel will do well against Obsidian. Also, heal was needed for the clash pushing. It should be pretty clear that they opt for Destroyer mid, Puck at their short lane and Spectre, ES and CM on their long lane for a trilane. I counter this with a 4-lane top with Omniknight mid leading into clash pushing.

Scenario 3:

Picks- . -

This game is from a league, and I am playing with two of my mates. Level of skill is decent.

First phase: I ban AA and Clock, because I'm thinking of pushing with a trilane (we were only three mates), and leave the two unkown players for solo. I do not want to be disturbed by rocket nor AA ulti incase I'm up against three heroes. Also, the two heroes are very strong and has counterpotential to the pushing I plan on doing. Spectre I wouldn't usually ban if I was pushing with my team, but since you can't rely on the other two, I have to ban Spectre. He is simply an almost autowin in these games.

They ban Weaver, Invoker and SD. Standard bans with nothing to add. Although, looking at their picks, they had pushing planned and did not ban heroes such as ES. Let me tell you first; ES is not THE antipush hero, like many think. Pushing does not involve a lot of summons, and therefore he can't really stop pushes until he gets dagger, which he most of the time wont. I'm not saying he isn't good, but he is far overrated. Also note that with the current pushstrategies, standard bans are probably the best way to go in order to hide your intentions.

They first pick Furion, which is something I personally wouldn't have done. However, it isn't terrible to first pick him. He can be gamebreaking to one's strategy. I still have plans on pushing, but I lead out with CM and WR mostly to see what the hell he is up to before I pick important heroes. Also, I want to give the two unknown players heroes I know they can use. WR has pushingpower, treantclearing abilities and escapes, which is perfect.

Next picks are AM and Venomancer which indicates the aggressive turtling strategy. So far their picks are going to be hard to push against, and I pick up Balanar. A hero which I already had in mind for pushing, however he will also work great against their pushing, since they are not going for clashoriented pushing, which will make it very easy for Balanar to gang.

Second phase: I ban Lich and ES. Most of all two standard bans, since they're needing atleast one more support with their strategy. Both ES and Lich are strong with counterpotential to what I'll be doing. And also, the ES and Lich has great syngergy with AM, so I'm also trying to limit his potential.

They ban Doom and Puck, which I had no intention of picking. Not weak heroes though. I suppose the Puck is banned slightly for the reason that they wanted Broodmother.

They pick Brood, which leaves me with the obvious pick, Sand King, who really counters Broodmother on solo lane. At this point I am pretty sure that they'll put Brood bot, but they could put both Veno, AM and Furion mid, so I'll have to wait for their last pick, which turns out be to Tidehunter.

This makes me somewhat convinced that they will do trilane top with AM and support of Tidehunter and Venomancer, which they do. They could've done AM mid also though. Furion woods is an option too, but most of the time, you will fail to make Furion effective this way.

I put WR mid against Furion, and Sand King bot against Broodmother. For the top lane, I want to push and end up picking Enigma. You probably think Enigma needs solo or woods, but situationally you should think outside the box. They cannot clear Enigma summons, they lack disables for the Black Hole, and we stomp them picking up both tier one and two towers in 7-8 minutes. When they tried to defend tower, we just dived and killed Venomancer and the following two afterwards.

Windrunner keeps pressure on Furion, so that he does not help top. He also farms fast Mekansm, Sand King neutralizes Broodmother push bot (fails to kill him and actually dies once, since he isn't very good), and we end up destroying raxes at minute 15, while they according to their chat expected to be picking our towers up very easily.

I am trying to decide whether I should add replays to these.

Scenario 4:

Picks- . -

First phase: Bans are pretty standard, and they lead out with an old school first pick Venge, which tells me these guys probably aren’t the best. I pick Furion and CM to kind of see what they’re up to, but also to secure myself Furion obviously. Next picks are Lycan and SF, which is even weirder. Obvious pick next for me Batrider to both counter the weak SF, but also to counter BKB Lycan.

Second phase: Since they need a support and even more so AoE disables, I ban ES and Tidehunter, who would save them from complete lack of clash power. They ban Enchantress and Windrunner because they smell the push supposedly, but as we see, also because they want Chen.

With the Chen pick their synergy is no better and I decide to opt for a Juggernaut in all seriousness. One might say it isn’t that good of a pick here, since they have summons vs. the ulti and lots of physical damage vs. the spin, but the healing ward will do good for the pushing, and they are almost all pretty weak heroes. Last pick from them is Lion, and I’d assume they’d put Lion mid and SF to farm bot with Lycan solo top, however they decide to lane SF solo top instead.. Anyhow, I last pick Undying and opt for a pushing trilane top with CM, Juggernaut and Undying. With all this heal, dive power with Batrider and Juggernaut and summons from Furion the towers fall fairly easy.

Scenario 5:

Picks- . -

First phase: I had some intel on these opponents, and that is why I ban Broodmother and Lycan. Strong heroes, who are definitely justified a ban though. As for their bans it doesn’t get much more standard than that.

They lead out with ES, and with that pick it is pretty safe to say that they will not be pushing. This is why I opt for the Spectre, partnered with a Lich for synergy and options kept open. They decide to grab the Windrunner and Lion, which I hope for their sake will be used solo. Lion really should not be played as trilane support. He needs items and exp advantage to be used successfully. Just like Rhasta and Sand King. I think a mistake here was not grabbing Pugna instead of CM, but I kind of wanted to see what else they had in the bag.

Second phase: I ban Anti-Mage and Venge to limit them lane and lategamewise. Also because of the synergy Anti-Mage has with ES. They ban Doom and Pugna, which shouldn’t require explanation. Mainly the Lion is very much countered by these two heroes.

Their first pick is Enigma, and it confused me a bit. They would need some more physical dps or they are for sure not winning, but I also didn't think that they’d Enigma woods. They did however. For Enigma Black Hole and the rest of the weak bunch, I counter with Panda. His ulti does not get affected by Black Hole and he is very resistant to spell damage. They finish off with Morph. An old school pick, which indicates that they in fact will put Enigma woods. What’s also to note about this ES/Enigma lineup they’ve picked is that they’ll be clash forcing. For this reason and the fact that I will be doing aggressive turtling I grab myself Furion. This would mean: Spectre and Lich mid vs. Lion, Panda and CM top vs. WR and lastly, Furion bot vs. Morph and ES. The thing about bot lane is that ES will move mid pretty early into the game because Lion is struggling. The other two lanes would be pressured enough for Furion to get some farm also.

Scenario 6:

Picks- . -

First phase: Due to some intel on the opponents, I ban Batrider, Anti-Mage and Invoker just to get them out of their element. They ban SD, AA and Furion which could signal that they either want to turtle or clash push.

I lead out with Chen for a couple of reasons. Firstly, I had a nice Chen player on the team, meaning it was sort of to secure him this hero. Secondly, a first pick such as Chen throws many captains off their usual stream of thoughts with the trend first picks and so on. You really do not have to first pick this or that. You see, he even reacts picking two key heroes to his lineup first. Both of which are easily countered. I grab Beastmaster to rape this Doom at the mid lane (he also does great against Weaver) and Lycan, since they have no disables. The only real counter to Lycan left in the pool is Enigma, which I think they should’ve picked instead of Venge. Note that Lich isn’t picked here simply because his ultimate is sort of countered by Chen.

Second phase: I ban Enchantress, since I don’t want to face dual woods/roaming with Enchantress and Venge combo, but also because I’ve grabbed Chen, and Enchantress somewhat counters Chen. They ban CM and Tidehunter since they figure that I need a support, but also because Tidehunter counters Weaver.

First grab is Clockwork to counter the inevitable ES and possible Windrunner. Besides, he also does great against Weaver. They respond with Void, which is the only hero left in the pool, who can stop BKB Lycan. This was fine with me, since now it is pretty clear that they are going to turtle, and I will have little to no problems pushing the towers. For this reason I grab Undying, figuring I will put Clock top vs. their trilane and push bot lane. The ES is sort of self-explanatory.

More scenarios will be added.

Thank you for reading.

Back to top
Written by 14553

Misc guide
Author: 14553
Map Vers.: 6.72f


Drafting in -cm for intermediate

Date Posted: 08/15/11
Last Comment:10/11/2012
Total Votes: 146
Current Rating: 9.20
Views: 177880

Login to post a comment