DotA 2: Understanding Captain's Mode Draft in Competitive Games
Fri 24th Aug 2012 - 4:58pm
Sitting down in front of your computer you open up your internet browser and navigate to your choice DotA 2 youtube channel. It’s time for you to catch up on some games from your favorite teams. You have a cup, glass, or perhaps pitcher of your favorite drink be it coffee, tea, beer, wine, or otherwise. Arriving at the youtube channel there are many games to choose from, so you pick one from the team you planned on watching.
The video hasn’t even begun to load before you are looking for the timelink posted by a fellow spectator that skips the 10-15 minute drafting phase and gets right to the meat of the match. If you justify doing this by saying that the pros only pick the same heroes from the pool anyways, you would be right. Well, mostly.
Nothing to see here, right? You've seen these heroes countless times before.
While it is true that the pros tend to pick from a limited array, it is the specific mixture of a team's picks versus the opposing side’s medley that indicates at the very start of the match how the rest of the game may proceed. Now, there is no right or wrong way to enjoy a DotA match, but in this guide I hope to be able to provide you with some seeds of knowledge that will allow you to eventually see the underlying game of DotA that plays out beyond last hitting creeps, buying items, and winning the game with practiced skill usage.
Know Your Enemy
"If you know your enemies and know yourself, you will not be imperiled in a hundred battles... if you do not know your enemies nor yourself, you will be imperiled in every single battle."
-Translation from The Art of War by Sun Tzu
Before we dive into drafting, we need to gain an understanding of perspective from the eyes of a team captain. Prior to the game even launching there are pieces of information that a captain uses to gain a slight advantage over their opponent. A disciplined captain will have knowledge of their opponent’s strengths and weaknesses on both the team and individual player level.
They will have a general idea of who is likely to play which hero and in what lanes. They will know what strategies and playstyle the opposing team tends to use. They will also be intimately aware of their own team’s limitations. With this information a captain has general idea of their options and strategies going into a match. This is similar to how any other competitive team, electronic or otherwise, approaches their opponent. Know your opponents and don’t underestimate them.
Mechanics are crucial to fully understanding the draft, and they are fortunately straightforward. First is a phase of bans where each team alternate selecting three heroes which neither of them will be able to play. The Radiant always has the first turn in each phase. Following that is the first of two picking phases where the team will be able to select the heroes they are going to play with the Radiant picking one hero first.
If each team selected only one hero at a time then the Radiant would always have a bit of an advantage in the draft. In order to counteract this the Dire are given two picks at once. To make sure the Dire doesn’t gain the upper hand, the Radiant next get to pick two in a row. The round of picks is evened out with one final pick from the Dire.
The following two ban and pick phases return to single alternating selection. For each of these turns the team has 40 seconds to make a decision and a bank of 90 seconds to use as reserve time for the entirety of the draft. The one exception to the time rule is when a team has to pick two heroes in a row they still only have 40 seconds to make those two picks, the clock does not reset between each pick.
Now, keeping in mind the pieces of intel that the captain holds against his opponents, we can take a look at the first stage of bans. The captain has essentially a blank sheet of paper and must make the first marks on the canvas to be able to see how the rest of the drafting will progress. Captains start by making choices based on their own strengths and weaknesses as well as their opponent’s.
Usually heroes that will be banned at this point are those that the captain has a feeling their opponent wants to use (perhaps taking into account previous games in a best of 3 or 5 series) or that they don’t want to play against. At this stage it is too early to be able to make meaningful predictions about the rest of the draft; there are too many unknown factors.
Setting the Stage
Tier 2 tower at 5 minutes, but will they survive teamfights at 45 minutes?
One common misconception about the picking phase is that, although there may be trends as to what heroes are picked up first, there are absolutely no set rules as to the selections a captain should make. There are no heroes that must be banned or picked 100% of the time. That being said, typically a captain will pick heroes early that have a variety of uses to avoid giving their opponent concrete information on their strategy.
However, if a hero gets through the ban phase that the team does not want to risk giving away, even if it makes their strategy predictable, they will pick that hero and incur the subsequent risk. At the end of the first stage of picks there is usually enough information for the captain to begin to see their opponent’s strategy, however it is still possible to mislead or hide intentions with the first three picks and change direction with the last two. This is a strategy in drafting that is often employed. To understand how this works knowledge of general strategic options is necessary.
A game of DotA will not reach an end if neither team crosses the river. Taking down towers is known as pushing. To win a team must push all the way down through the lanes and into the base to take down the ancients. An optimal time to push down towers, be it in the early, middle, or late game is tuned by the type of heroes that make up a team composition.
For the most part players kill in groups of two or more, known as a gank. One on one kills do occur, but with less frequency. While theoretically it might be possible to end a game of DotA without a single hero dying, the likelihood of this occurring is very small (if you know of a match where this happened be sure to post it in the comments). Practically speaking, a team will have to kill their opponents in order to gain an advantage, however the number of kills does not always indicate which team is ahead.
A team has to have a sufficient number of heroes at a high enough level to be able to remain alive during fights and push into the base long enough to destroy towers, barracks, and ultimately the ancients. In public matchmaking games players will often pick more carries than necessary to accomplish this goal, but in competitive games teams will usually optimize their carrying potential by selecting one or two heroes to take the majority of gold and experience while the rest are used as support in one form or another.
These are the basic elements of a game of DotA and a team will combine them to varying degrees to win a match. Put simply, you cannot win if you don’t push down towers, if you don’t kill the enemy, and if you don’t have heroes strong enough to sustain an attack on the enemy base. The application of these elements produces the actual strategies that a team uses in competitive match-ups. There are no limits to the number of strategies a team can develop, but there are plenty of commonly used strategies that can be identified from the draft.
Increase your game knowledge by reading abilities. Know what heroes are best suited for what roles and why.
Four Pushing to Protect One
This strategy involves picking heroes that are able to gain map control by taking down towers as early as possible, and then hold their advantage. The goal is to make it less safe for the opponent’s team to collect gold and experience, while at the same time making it safer for your carry to do the same. The timing of when the carry will join teamfights depends on its peak effectiveness and the overall flow of the game.
Four Ganking to Protect One
In a similar vein to the previous strategy this one requires that the team kill early on and have the ability to withstand 4v4 and sometimes 4v5 fights in the mid-game while the carry farms. Kills in this strategy are prioritized over towers, but still require that the carry have sufficient experience and gold before joining the teamfight.
The split-push requires that a team has two or more heroes that can push creep-waves to towers and take them down on their own or with little help. Typically a team that plans on split pushing will try to get defensive kills while their carries are able to gain the experience and gold required to advance across the river towards the enemy’s ancients.
When a team cannot take a full head-on teamfight to get a tower or barracks they hunt around the map looking to pick off a hero. While that player is down they advance on their target objective with the full team knowing that they have an advantage.
Usually relies on heroes with big ultimates plus a carry to finish off anyone that was able to survive. This strategy works well once the crucial heroes hit level 6, but can fade as the ultimate's effectiveness tapers off in the late game.
Ranged Tower Push
This is where a team will look to take down towers from a safe location and try to avoid full teamfights where they risk losing heroes. This type of pushing usually comes in the form of a split push, but may end up with all five heroes in one lane instead of split up between the lanes. They will try to maintain pressure in all lanes as long as possible.
Level One Roshan
Not often used as it can be extremely risky and oftentimes obvious with the drafting of heroes capable of taking down Roshan at level one. This strategy, if successful, will result in an early advantage in both gold and experience which can help translate into winning the laning phase. The major downside to this strategy is the cost required to take down Roshan which may involve extra salves or a specific team composition that is weak without finishing Roshan.
As I said, there are an endless amount of strategies a team can utilize, and I do not proclaim to know them all. Feel free to add anything I didn’t touch on below in the comments!
Returning to the draft, the final stage of bans allow the captain to either take guesses as to what two final heroes will round out their opponent’s strategies, ban heroes with known synergies, or ones that would counter the captain’s plan. At this stage the captain should have at least some idea of the possible strategies their opponent will use and what heroes would be an asset to them.
The final picking phase is about making sure that you either reveal your strategy at the right time when it is less likely that the opponent will be able to pick effective counters. Waiting until your last pick, however, to grab a key hero to your team composition can prove risky as the enemy team may pick that hero.
This is also the part of the picking phase where you are most likely to see a less common hero picked. Since the hero is less common the likelihood that the hero would be either picked or banned beforehand is relatively low, so to maximize the effectiveness of the draft the captain will wait on selecting the unconventional hero. The captain will have a repertoire of strategies with various heroes that he can pick from if he feels he is in a position where the unconventional team composition will throw the enemy team off course, or catch them off guard.
A team will not pick a hero they are truly uncomfortable using in a serious match. Keep in mind that every hero does have a use in a certain situation in DotA and it isn’t always safe to assume that the enemy is going to remain using the popular heroes dictated by the current meta-game. Part of the reason for the rise and fall of hero popularity is due to captains trying out something new to usurp the reigning meta-game strategy. The more a certain strategy is used the longer teams will have had time to contemplate counter strategies. Those who only follow the rest will reach limited success due to their lack of creativity and subsequent predictability.
Would you put Broodmother and Lion in a lane?
Lion protects Broodmother and Broodmother helps Lion jungle.
Next is arguably the most important part of winning the early game for a team in DotA, lane selection. There are three lanes: hard, safe, and mid plus the jungle. There are any number of ways that heroes can be divided and it is up to the entire team to try and predict their opponents lane choices, as a mistake here could prove costly.
Predicting lanes, however, is a matter of knowing both specific team tendencies as well as the strengths of specific heroes and is therefore outside of the scope of this guide. With the knowledge in this guide you hopefully have gained some new insight to drafting that should help demystifying the decisions made, which in turn will allow you to understand lane selection as you watch more competitive games.