Welcome to the ultimate Dota 2 guide – a complete list of all major ways to improve in this game. Wherever possible, we’ve provided resources where the topic is discussed more in-depth. With time we’ll fill in the list of reasources.
You win at Dota by destroying the enemy Ancient.
So, Dota boils down to:
All ways to get better below are in some way connected to this simple truth. If what you're doing doesn't lead to better map control and a resource advantage, you're doing something wrong. If you get lost and don't know what to do and where to go, think back about this rule.
You are fighting to kill the Ancient, not the enemy heroes. Always prioritize pushing towers over getting kills.
You need to become the richest hero in the game in order to secure the victory mid/late game.
Most Greedy & Selfish: Your resource development is your number 1 priority. Make use of the best/safest farming space, don't sacrifice yourself for your teammates, etc.
Most risk-averse: every single death increases your chances of losing a great deal. Don't take almost any risks, leave this to your play-makers.
You need to be a carry, yet you also need to be a play-maker. Focus on your own resource growth, but when you hit your timing and you become one of the strongest heroes on the map (usually after the laning stage) use it to be proactive and to force aggressive fights in order to win map control for your team and free farming space for your pos. 1 carry.
The sacrificial core. Take the hardest farming spots on the map to leave the safer space to your pos. 1 & 2 heroes. Play with your teammates and participate in all possible fights to secure map control. Build the utility items your team needs.
A mixture between a utility core and support. Depending on the hero, you might have the opportunity to be very mobile in the laning stage and help out in different lanes. Help out with supporting if needed, but be greedier than the hard support and try to get some key items (usually Blink, etc.) which will increase your fight impact.
Your job is to make sure your cores have a good start. Trade in the lane and secure space for your partner to farm in the lane. You also need to win the Vision wars (ward/deward) throughout the game - with vision you can make more educated moves around the map. Farming is a very low priority - shadow your active heroes and help out in fights.
Frontliner: usually a tanky farmed core, you are the hardest hero to kill on your team. You stay in front of your allies and want to draw the attention of enemies in fights. You are usually the one hitting the tower in pushes. It's best if your team has save mechanisms to help you out if you get in trouble.
Common mistake: playing too defensively. Enemies will ignore you and pick off your teammates.
Damage-dealer: you follow-up the initiation of your teammates. You don't enter the fight first because if you get killed early the chances of your team winning the fight decrease drastically. Once the fight starts, you commit in order to deal as much damage as possible and secure a victory for your team.
Common mistake: too aggressive positioning - you get picked off too early by the enemy initiators, which becomes a disaster for your team, especially if they commit to the fight anyway.
Initiator: you search for an opportunity to start a favorable fight for your team. This means staying out of vision in order to use the element of surprise on your side.
Common mistake: showing yourself needlessly - try to always move in the fog of war. You are the main pick-off danger on the map. If enemies see you, they will know where on the map they can play aggressively and where they cannot.
Counter-Initiator: usually a hero with a big team-fight spell. Your job is to wait for the fight to happen. Once the enemy team commits and ideally clumps-up, you jump in and get a multi-hero disable.
Common mistake: Impatience. Multi-hero counter-initiations are much easier to do in the middle of the fight rather than in its beginning. If you're impatient, you won't be able to find those opportunities.
Support with a save: usually supports with defensive spells and items. Your positioning needs to be very defensive, behind your cores, ideally out of vision. You need to keep out of danger so that you can bail out your cores at the exact moment they need it. You need to be in-and-out of the fight to optimize your spell usage. Don't commit hard - your right-clicks don't matter and it's too risky.
Common mistake: Bad positioning. Overextending in fights will get you killed in an instant at the beginning of the fight and you won't have any impact.
Execution: minor mistakes and good plays can snowball to a big advantage and secure games. Becoming proficient with a small hero pool and a particular role is a good idea because this way you can refine your play to the smallest details. Strategy: the meta matters much more in this bracket than in the others. Use it to your advantage.
The most basic thing you need to learn to gain a resource advantage is to farm quickly and efficiently.
As mentioned, the key thing you need in order to out-farm your opponents is space (access to resources on the map). This means that Dota can boil down to two things - creating and using space. Using space if self-explanatory (you farm the resources you have access to). There are different ways, however, to create space (more details in the tactics section).
In order to make good decisions, you need information. You gain information through vision. If you're having vision troubles, take care of them yourself even if you're not the hard support - but the benefits far outweigh the drawbacks.
Where do you need vision? Generally speaking, there are two kinds of wards:
Locations: where exactly you want the wards depends on the game (you want them where the action's taking place), but the basic idea of the ward is to give vision over a strategically important area. Those are junctions and chokepoints (ramps, river), important areas in the jungle (Ancients, farm triangle), Roshan, etc. The designated ward locations (ward high grounds) are generally speaking the most efficient vision-wise, but the most predictable and easy to deward, so make sure you're not seen when you place wards there.
Resources: Purge - Basics of Vision & Warding;
Which areas are safe to farm and which aren't changes throughout the game. For example, during the early laning stage, your safe lane and your own jungle are generally speaking the safest places to farm. After the lane progresses, however, it's very common for the enemy team to try to move into your jungle and safe lane to take control of that area. At this point, the T1 safe lane tower actually becomes one of the most dangerous locations on the map.
The Dead Lane: A few years ago BSJ introduced the concept of the dead lane to describe this. When you're a farming carry, make sure to think where the dead lane is and to avoid it at all costs while farming on your own. If you want to take back control of the dead lane, move there by force (with your whole team, using smokes, etc.).
Resource: BSJ Dead Lane Guide;
The map features are crucial when it comes to fighting and ganking:
High-grounds: don't walk onto high-grounds without vision if the dangerous enemies are alive. That's the easiest way to lose your life needlessly. Try to avoid team-fights where you are the one pushing onto a high ground without vision - obviously, impossible when you're pushing the enemy base, but easy to avoid in most other situations. Use Observer Wards to make your life easier in that regard.
Chokepoints: ramps, the river Fag, the Rosh pit; choke points force heroes together and make them more vulnerable to AoE. If you're fighting against a team with powerful AoE abilities, it's best to take skirmishes and fights in more open areas where the fight can be more spread-out and vise-versa.
Juke spots: Most juke spots are openings in the trees that allow you to walk there hidden from enemy vision. The obvious way to use them is to juke in order to survive a gank and buy time for your teammates to react, but trees and juke spots are equally important for staying outside of vision in the actual fight - very valuable for initiators and counter-initiators.
The most important places on the map are the map features (choke points, ramps, high grounds, etc.) leading to important resources - Roshan, Ancients, farm triangle, outposts, runes, etc; Make sure to think about those areas if you are the one warding or the one making map movement calls.
The resources and space you have on the map are limited and in order to secure an advantage, you need to realize which hero needs farm the most at what point. You need to think about the resource gain of your whole team and how to optimize it, not only of your own hero.
This usually means creating space and protecting your farming cores. Sometimes, however, it might mean allowing a support/utility hero to take some safe space in order to farm an important item like a Blink Dagger, etc.
Not every hero has the potential to farm all available locations and in order to gain the maximum amount of resources you need to utilize all available space.
E.g. let’s say you have a farming Jugg and a farming SF. The enemy team has no disable through BKB. Because of this, the Jugg can easily farm the more dangerous locations – if he gets ganked he can Spin and TP out. This will allow your SF to take the safer farm and as a result, both of your cores will have the needed space to grow.
Another example: you have a safe lane Lycan. At some point in the laning stage, he’ll have the needed levels and items to farm the jungle very fast. If you have another hero that can make good use of the safe lane (offlaner or support), it’s a good idea to move Lycan into the jungle and give the lane to this hero. This isn’t mandatory, however. Another option you have is to start pushing out the lane as fast as possible and to flash-farm the nearby jungle camps until the enemy team pushes the lane back in your direction, thus maximizing Lycan’s GPM.
Dota is quite predictable in the laning stage, a bit more unpredictable in the mid game, and extremely unpredictable in the late game. This means that the later the game goes, the more the game is decided by random mistakes/plays (who gets the jump, randomly over-extensions and deaths, not saving for buyback, etc.).
The single best tactical tip you need is to try to get objectives after every successful fight, most importantly - Roshan followed by the enemy base. If you manage to win before the late game, you minimize the chances that random mistakes will cost you the victory.
Choosing the right time to fight is one of the most important decisions you can take. Your team isn't equally strong at all stages of the game, and the same applies to your enemy team. Obviously, you want to fight when your lineup hits a strong timing and avoid fights when your opponents hit a peak.
For example: you have a lineup with Bristleback/Jakiro/Abaddon. You don't scale amazingly well, but you are stronger early on. This doesn't mean that you should play aggressively right from the start of the game because over-aggression in the laning phase could ruin your mid-game timing. A good game plan would be:
Getting an important item on a hero is a good indicator to make a move. The most obvious example is Blink Dagger - make sure your opponents don't see it when you buy it and use the element of surprise to your advantage to gank successfully.
Another good example is BKB on your cores. If they need BKBs, it's a good idea to play passively until they get them and to force a fight as soon as possible once they do. The same principle applies to most other major items.
Pretty self-explanatory: don't take team fights if your important spells are on CD. Ravage, Chronosphere, Black Hole, etc., are all spells whose huge impact dictates the pace of the game.
At the same time, try to take fights when the important spells of enemies are not available and make sure to keep track of the clock to know when the spells will become available once again.
So far we've been talking about when to make a move. You also need to decide, however, what move to make exactly. That, of course, depends on what your opponents are doing:
Resource: Dota 2 Tactics Guide
Grouping up as 5 has two big implications:
This means that you need to have a particular reason to group as 5. Sticking as 5 for long periods of time without achieving anything is a sure way to get out-farmed and lose the game in the long run. Generally speaking, you want to group as five to be as strong as possible in team fights around objectives.
The goal is to utilize all available resources on the map. The advantage is that you are growing in resources very fast. The big disadvantage is that you are vulnerable to fights because you're always in a numbers disadvantage. This means you need to avoid engagements as much as possible.
In case you don't have good vision of the enemy heroes, make sure your more vulnerable heroes (easy-to-kill farmers) are only farming the safer, more defensible and well-warded farming areas, closer to teammates and towers. The dangerous areas need to be farmed only by the very hard to catch heroes.
In case you have good vision of the enemy (they are gathering for a push/Rosh/gank), use this information to push out the lanes aggressively and even split-push in order to force reactions.
Stall the game - constantly out-push the waves, anti-push and split-push during enemy aggression to buy time for your GPM/XPM advantage to start kicking in and to allow you to dominate fights.
The art of finding kills on heroes when they are away from their teammates. If you are a strategy that wants to fight, but your opponents don't want to take the 5v5 engagements (they are split-farming, spit-pushing, etc.), your best option is to gank the vulnerable farming enemy heroes.
The king of ganking is vision - a gank will never work if it's not surprising.
Your initiators should stay out of vision as much as possible, especially before they make a move. Initiating from an angle is also useful.
You should get aggressive observer wards to increase your chances of finding a target and you should deward the traditional ward spots in the areas you want to gank if you intend to do so without Smoke.
Last but not least - the best way to find a fight (a gank or a full 5v5) is by smoking up.
The art of baiting the enemy team into taking an unfavorable fight.
Baits are hard to organize in pubs, but they could be game-winning and they are the best way to turn the game against a ganking team.
The reason is simple. The counter to ganking is 5-man Dota. Yet, the enemy wouldn’t attempt a gank if they know your whole team is prepared to take the fight. So, you need to trick them.
The usual way to perform a bait is to leave one hero visible in a location where it seems vulnerable (usually pushing out a lane alone) and to hide the rest of the team (or part of it) nearby in the fog of war (or even better - smoked) so that the whole team can respond once the enemies initiate on the supposedly vulnerable target.
Winning the laning stage is obviously quite important as an advantage in Dota can easily snowball.
In a sense, you can think of the lanes in terms of map control. Generally speaking, unless your opponents are misplaying or have a terrible draft, it's hard to control (and win) all three lanes, but it's always a good idea to plan to win at least two. So, how will you achieve that?
Lane swaps are a great tool to help you avoid bad matchups. It's advisable to use this tactic every time your opponent pics a specific lane counter. Also, don't swap only at the beginning of the landing stage. You can swap throughout the laning stage to adjust to changing circumstances.
You can also rotate heroes to reinforce weak lanes or to grow your advantage in strong ones. Early on the supports do all the rotations because cores don't want to miss out on lane creeps. Later on, however, rotating even cores (strategically, with a plan in mind) is a great idea to keep ahead of your opponents and win map control.
Tactics are team-activities, you cannot be successful if your teammates don’t participate. Use voice chat as often as possible, tell your allies what you want to do and ask them to help if necessary. It is better to do the wrong tactic together as a team than to do the right tactic alone. Everything could work if you execute it as a team and nothing works if you and your teammates are on different pages.
Much more on this in the TEAMWORK section below.
Always have a game plan before you start playing. How will you win the game? The plan typically consists of three parts:
Of course, the plans can change. For example, if the game is not going well and you cannot execute your late game plan (how to close the game), it can change from the one above to something different: E.g. 4 people will stall the game and try to delay pushes while our pos. 1 will split-push to force reactions and buy time. We reach the super late game where the resource advantage the enemy team has doesn't matter as much and we close the game when an important enemy doesn't have a buyback.
Resource: DotaAlchemy planning to win a Dota Game
It's useful to approach item builds in two steps.
One: what is the standard build for that hero in that position in the current meta? You should understand it and use it as your starting point. It's the default build for a reason. If there is more than one standard build, usually this entails more than one way to play the hero. For example, PA can easily go for a quick Battlefury OR for faster-paced aggressive items (fast Deso, etc.). The BF build means that PA plans to farm for longer, while the aggressive build means she plans to join fights fast and create space. Each build is better in specific situations, and you need to choose the right one for the game (that works best with your game plan).
Two: does the build need adjustments based on the specific game? E.g. the standard build goes for a BKB as a 2nd big item, but in this game, you're facing a lot of magic damage and disables. It might be a good idea to get BKB as a first big item if you plan to fight. Another example - the standard build goes for a Daedalus in the late game, but in this game, your lineup has plenty of damage and not enough disables. Maybe it's a better idea to get Abyssal/Scythe, etc.
The logic is the same as above:
What makes any combination of 5 heroes a good or a bad draft? Theoretically, you can find a viable game plan with any random 5 heroes, but following a few basic rules of thumb can make your life a lot easier. Here is a checklist for you to follow:
Logical lanes: what lane setup will you go for? Who will be supporting in the lanes, who will be farming, how will you deal with the strong lanes of your opponents, how will you punish the weak lanes?
1-5 Positions filled out: there isn't enough space for 5 farming heroes on the map. Some heroes need to have higher farming priority, others to spend most of their time creating space and supporting them in fights. So, do you have a hero who can do decently the job of each of the five positions - from hard farmer to hard support?
Damage: Winning fights is almost impossible without a good source of damage. Make sure that you have enough damage to win fights in the early and mid game, not only in the late game once you have luxury items. Moreover, burst damage and sustained damage output both have their uses. Nukes are great for securing a quick kill on a target, but might not be enough to win a prolonged 5v5 fight. Ideally, you need a bit of both.
Pushing and Roshan potential: How will you take the objectives? Do you have heroes who deal good damage to towers/Roshan? Do you have a frontline hero who can push the tower without fear of being instantly blown-up by the enemy initiation?
Wave clear: a large part of the battle for space in a game of Dota is actually about pushing the creep waves into enemy territory. If you have a lot of single-target heroes who kill creep waves slowly, you'll have trouble keeping the creep waves pushed out and securing map control.
Control: if killing the enemy heroes is a part of the game plan, then you need to have control, otherwise the enemy team will be able to disengage unwanted fights easily. Disclaimer: killing heroes as a goal is often part of the plan, but doesn't have to be in all cases. E.g. if you plan to force fights and take map control by pushing towers, you don't need control that desperately - you can be successful at claiming the objectives regardless if enemy heroes die or disengage - they will have to commit to a fight at some point if they don't want to lose their base. That said, control is still very useful in fights for a variety of reasons - keeping opponents from kiting you, etc.
Initiation & Counter Initiation & Saves: who starts the fights on your team? If you don't have an initiator, you'll have trouble starting a fight on your terms. Like the example with control, you can still force fights by going for objectives and forcing the enemies to fight into you, but this by definition means that they will start the fight on their terms. In such situations, it's very, very helpful to have a counter-initiator with some form of AoE control and a mechanic that allows you to save the teammate(s) which the enemy team focuses on (ideally your tanky frontline hero).
Resource: Dota Alchemy Drafting in any meta
Even the highest level pros have a pool of comfort picks. Forcing a player on your team to play a hero they aren't familiar with just because it fits the game well theoretically is usually a very bad idea.
Your pool of heroes for a game doesn't actually consist of all heroes in Dota. It consists of the heroes your teammates can play well.
Lane synergy is how well your lane partner heroes work together. For example, a mid OD can work very well with certain roaming supports because he can set up for a gank with Astral Imprisonment. A core Ursa works very well with a support that can provide good control - slows and disables help Ursa stick to his target for as long as possible and increases his kill potential greatly.
Lane counters are all about lane matchups. For example, if the enemy team has a safe lane Ursa, drafting an offlane Dark Seer on your team will make Ursa's lane much more difficult. Ursa has trouble out-pushing the lane against the Ion Shell pressure unless it invests more points in Earth Shock which is unusual, and Surge counters Ursa's high kill potential because it disallows him to stick to his target. (Tip: after the picks are made, use the lane-swapping tactic to avoid bad lane matchups.)
Resource: DotaAlchemy laning stage drafting
Spell synergies are generally speaking powerful spell combinations that increase the impact of your individual picks.
Spell counters follow the same logic, but instead of increasing the impact of specific abilities, they decrease it.
Much more abstract but equally important as the other kinds of synergies and counters. On this level, however, we're not talking about individual hero interactions, but the interactions of the game plans of the two teams.
The tactical triangle is a useful tool here as well, but you need to think about it in the context of the game plan. For example, if you know that the enemy team will try to gank and find pick-offs, it's a great idea to have a draft that can play 5-man Dota in order to invalidate the ganking game plan of the enemy team and to force them to play your game instead. The worse mistake you can make if the enemy team has strong gankers is to pick passive heroes that need a lot of space and time to farm in the early-mid game (i.e. they want to split-farm, and the enemy ganks will prevent them to do so).
Not all game plans are made equal. Some require a lot of coordination and clutch execution, while others are quite straightforward. This is true on two levels - tactics and fight execution.
Hard to execute fights: Try to minimize the riskiness of not landing your spells. This means two things in general:
Hard to execute tactics: 5-man Dota Wins Pubs, Rat Dota loses pubs
In pubs, especially low to mid-level games, relying on split-farming and split-pushing as your default game plan is a bad idea. The reason is that it requires your whole team to avoid fights for long periods of time, and even one trigger-happy teammate can mess up your execution. Moreover, one bad over-extension by your split-pusher(s) can be very costly.
On the other end of the spectrum, 5-man Dota is quite easy to execute. You gather as 5 and start pushing towers together and fighting together. It doesn't require coordination between your teammates all over the map and it doesn’t require that much map awareness. It only requires good positioning and spell-usage in the actual fights, which is more intuitive.
Being unpredictable is a huge benefit in mid-high level Dota as it makes your strategy much harder to counter. This usually means two things
Position flexibility: heroes who could theoretically play different positions are a good way to confuse your opponents. E.g. Mirana is a great example of a hero who reveals very little about your strat - she can literally be played in all 5 positions. Of course, the metagame will make one or two positions much more likely, but you can still use her flexibility to throw a curveball at your opponents and e.g. to last pick an unexpected new carry hero.
Lane flexibility: we mentioned lane swaps a couple of times so far - a great way to avoid lane counters. Having cores who can do well in different lanes helps a lot in this regard. E.g. PA is usually played as a safe lane carry in dual or trilanes, but she is entirely capable of playing solo mid.
Tactical Freedom: a great way to avoid tactical counters and to change your game plan when needed. E.g. if your team is good at nothing else but 5-man Dota, you'll have trouble dealing with split-farming teams. That's why a good lineup is able to play all three play styles with some reasonable degree of efficiency – it gives you the opportunity to adapt to the game. Keep in mind that as the game goes late, items give you the opportunity to fill in the gaps in your lineup and make it flexible anyway. That said, having some flexibility before this point is quite valuable and will allow you to play around timings. E.g. split-farming until you get some important levels/items, 5-man Dota afterward, etc.
Resource: DotaAlchemy Flex Picks
Very generally speaking, having a synergistic lineup that your players are comfortable with is 80% of what makes a draft good. All the intricacies of drafting and high-level counters are the other 20%. It doesn't make sense to sacrifice the 80% for the 20%. Make sure the 80% is covered, and only then think about the 20%.
Quite frankly, execution (or simply put - how well you play) is what makes the biggest difference. Take a look at smurfs as an example - strategy, counters, and all the rest rarely matters in games with a smurf in them. The higher level player will simply out-play the enemy team, gain a huge advantage, and carry his teammates to victory in most cases.
Because of this, improving your own performance will, in the long run, make the biggest difference.
Mechanical skill and execution are mostly about muscle memory and the game sense (intuition about the game) you develop with practice. It's very hard to gain it only by reading/watching, and you will gain it slower if you play a wide variety of heroes/roles/strategies. Spamming a very limited hero pool (even only one hero if you can bear it) is the quickest way to develop great muscle memory and game sense because you'll be exposed to the same situations much more often and you'll develop an instinct for them. You won't have to pull out this document in the middle of the match to see which tactic you need to undertake, you'll simply feel where on the map you need to be and why and how you need to use your skills, etc.
With this in mind, here are some concepts to allow you to understand some of the execution mistakes people make and help you avoid them.
Resource: Pvgna: spamming heroes to gain MMR?
One thing that helps a lot with efficient execution is always keeping in mind your main job in the game and role on the team. You cannot really do everything on your own, you'll spread yourself too thin. If you do one thing extremely well, however, you can get great results.
E.g. as an Anti-Mage your major job is to out-farm everyone. You do this by split-farming and split-pushing. Actively fighting to create space comes secondary - your minor job. Doing it is great, but you should under no circumstances sacrifice your main job for the minor one. Dying in order to help out in a fight as AM is a huge mistake. In contrast, dying to help out in a fight as a pos. 3 Tide is usually fine.
Another example - you are Silencer support. Your major job is to cast your ultimate at the right time in fights, your minor job is to deal some damage with your other spells and right-clicks. If you play aggressively in fights to deal right-click damage, you will be exposed to danger and you risk dying before you're able to use your Global Silence. Even though you might do more damage like that, your overall impact in the game will be lower. The right approach is to play safely and wait for the right moment to cast Global, and afterward when you are certain your major job is done - enter the fight and contribute in other ways.
So, always concentrate on your main job in the game first and foremost. It makes execution much easier. Minor jobs rarely dictate the outcome of the game.
When you are reacting to the movements of your opponents, you have only seconds (or even less) to make the right decisions. When you are choosing and doing the moves yourself, you have time to think and communicate with your teammates. Obviously, when you are reacting to your opponent you are more likely to make mistakes.
So, try to be the first one that makes moves and force your opponents to react to you rather than the other way around. (That's much easier to do when you have the more aggressive lineup as a whole, and there is a strong argument to be made that offensive strategies are better than defensive ones in pubs.)
Most inexperienced players make the mistake that they try to do too much during the laning stage and as a result - make mistakes and die. Winning the laning stage doesn't necessarily mean killing your opponents. It means controlling the lane and taking more resources out of it. So, it boils down to two simple things:
If this is the current situation, everything is going well and you shouldn't do anything fancy. If it's not, it's up to you to make some moves to change the status quo.
Split-pushing and split-farming, ganking with part of the team, going on warding expeditions by yourself - all of these are viable moves. Yet, they’re all risky because your team is exposed to an unfavorable fight (5v4 or worse). This means you need to have a good reason to do any of these moves, and when doing them - to do them as fast as possible.
If you don't have a concrete plan in your head, just stick as five and make moves together. This way even if you stumble onto a surprising fight (which will happen sooner or later), you'll be in full force and won't be caught unprepared, which will drastically decrease your amount of throws as a whole.
This is a very basic tip, but still worth mentioning for beginners as it improves execution.
When you stun a target, a disable bar shows on top that displays for how long the target gets disabled. Use this information to make sure you get the maximum duration out of your disables by not overlapping them. An extra second of stun duration is often what you need to secure a kill and could make a big difference in the fight/game.
Another basic tip, but quite important.
Flashy snipes with a skill shot look great on a highlights vid, but they are very hard to pull off consistently and consistency is what wins games, not flashiness. Because of this, it's always better to use your skill shots when you have the maximum chance of landing them - this is usually when you are able to use other disables to set up your skill shots.
Another skill shot tip - using hard-to-land skill shots when visible decreases your chances of landing them greatly because your target will anticipate them and try to dodge. Use them outside of vision to utilize the element of surprise. When not possible, you could try animation canceling to bait a reaction, or to shooting when the target is in a bottleneck and doesn't have a lot of space to move and dodge.
Generally speaking, you want to use your spells as much as possible to get as much value out of them over the course of the game. This, however, has some implications:
Short CD Spells: usually you want to use them at the beginning of the fight because there is a good chance you will be able to use them again (even multiple times) within the same fight if you don't die. If you save them for the kill, you'll get a lot less value.
Long CD Spells: high-impact long-CD spells share the same philosophy. Waiting for the perfect opportunity and never using the spell is a bad idea. 10 average spell uses are much better than 1 great spell use and 9 missed opportunities. The big difference is that while you need to search for opportunities to use the high-CD spell (force fights when it's ready, gank, etc.), you need to be patient inside fights. Don't throw it at the first enemy you see. Most big-CD spells are AoE, and giving the fight some time to develop will usually allow your opponents to clump-up, which will present a better opportunity (ofc. sometimes this is not possible, e.g. when your big CD spell is your team's only way to initiate the fight/gank). This ofc., doesn't apply for single-target high CD spells.
Communication is obviously vital in Dota, but as you probably know, relatively few people use this tool well in (non-party) pubs.
Start using voice chat. It will improve your success rate. It's as simple as that. Of course, for what you use it matters, but simply using it is a good-enough start because you need to develop a habit of talking to your teammates.
Games are very often lost due to positioning mistakes or wrong tactical decisions. If you talk to your allies (and say the right things) you could prevent these mistakes from ever happening no matter the skill level. Let your carry know he is in danger of getting ganked, and he might listen to you and avoid a needless death. Tell your teammates to wait for important timings (BKB, BB, CD) before engaging, etc.
Most people would gladly listen to you because:
Of course, there are people who wouldn't listen, but this doesn't mean it's not worth trying.
The only occasion where you shouldn't take leadership of the team is when someone else is already doing it actively. On these occasions try to help that person rather than challenge them even if you don't fully agree with their calls. Often, doing the "wrong" thing together as a team is much more likely to give you success than half of your team doing the "right" thing, and half of your team doing the "wrong" team. That's true because usually right or wrong is debatable and every move has pros and cons. The only truly 100% wrong thing is the lack of teamwork.
Resource: GameLeap communication hacks
Equally important to making calls, being a leader means taking responsibility for your team's morale and emotions. Putting people down for their mistakes only increases your chance of losing. This doesn't simply mean not flaming people yourself - it means actively pulling people together and managing any friction in the team to the best of your ability.
Use voice chat to empower your teammates and they will carry you to victory. You'll be surprised how huge of an impact a few nice words and compliments can have in an environment where everyone is used to being flamed.
Bonus: There is a certain type of Dota players who are mechanically very good for their bracket, but their temper (tiltability, toxicity, etc.) means they lose a lot more games than they should. If you are able to chill such players out, you vastly increase your chances of winning - on a good game, they will over-perform (and in bad games they won’t simply under-perform, they might even game-ruin).
Resource: True Sight 2018
The simplest tip to implement, but still impactful. There is no "surrender" button in Dota for a reason.
Don’t give up until the Ancient is destroyed even when you are certain the game is hopeless. You are playing against people and people make mistakes (true even in high MMR). If you take advantage of those mistakes you could win even theoretically unwinnable games - take those games as a gift from your opponents.
It costs nothing to tryhard till the very end, and it will improve your win rate (and improve it a lot especially in the lower skill brackets).
Be open to new ideas and opinions - you are not perfect, you can always improve. You will win consistently only if you are better at the game than your opponents, but getting better is an endless process. Blaming others is a natural psychological defensive mechanism, but it's detrimental to your growth. Don't use the mistakes of your teammates to shift the blame away from yourself. Instead, use the good and bad things your teammates do to learn and improve your own play for the next game.
Tool: A Dota journal. The best way to do this is to take two minutes to write down something important that you have learned after every single game of Dota. It's not only useful for learning, but it's a good method to get a short break and not to carry any bad emotions into your next match.
You can do this any way you want, but a great tool to use is roamresearch.com because it lets you link topics. For example, every hero name can become a link and this way you'll be able to find all the notes you wrote down that include a specific hero/move/strategy, etc.
Viewing games to pinpoint mistakes (and good plays/calls) is a very useful practice because a bird’s eye view of the game can give you a perspective you don't have while playing. That said, you need to learn to do that very efficiently rather than waste hours analyzing every little detail. You need to go into the reply with something specific you want to search for:
Disclaimer: not all replays are worth studying. If you had a straightforward victory and everything went according to plan, you are unlikely to find major lessons in it.
Obviously, one of the best ways to improve fast is to copy better players. A true MMR gaining cheat coming from Cookie Dota: is imitating boosters. Watching pros or very high-level games is not as efficient because these people play in a very different environment. If you are not good enough, you are very likely not even to notice the mistakes and plays that determine the outcome of the game.
Boosters at your MMR, however, play in the same environment as you do. Watching their gameplay lets you copy their plan step-by-step. How do they win the lane? What moves do they make after the laning stage and when? What builds do they go for? How do they finish the game? Imitating everything even without fully understanding it will probably give you good results, and with experience, you'll start understanding the game plan better.
In "Coordinate your team" we said "they'll gladly conform because it shifts the blame from them to the call maker". This is true because deep down most people are willing to reduce their chances of winning to protect their ego. In order to truly improve at this game and win pubs, you need to totally eradicate this kind of thinking.
Your carry overextended, died without buyback, and you lost the game? Well, the right way to think about this is that this is 100% your fault because that’s the only productive mindset in the long run.
You could have saved their life with a simple call to back-off, so it's your fault that you didn't foresee the outcome of the situation and manage your teammate.
You tried to do it and the teammate didn't listen anyway? Well, it's your fault for not finding an effective way to reach this person for the duration of the game. Maybe if you were nicer to them and established a connection from earlier, they would be much more inclined to listen because they'd know you're on their side.
Try to find a lesson in such situations, don’t try to shift the blame to the bad play of your teammates to protect your own feelings.
You'll play only one game with this person, but you play all of your games with yourself on the team (cheesy but true). Blaming your teammates or even attempting to teach them is useless in (non-party) pubs. 100% of your energy needs to be inwards focused.
Resource: Extreme Ownership, Jocko Willink; The talk has nothing to do with Dota, yet it has everything to do with teamwork and leadership in a high-intensity environment; even though the stakes in Dota are much lower than the stakes in real war, the extreme ownership principles apply to all team-based competitive games;
Spamming 1 hero gives the best results in terms of MMR, but in order to become an overall better Dota player, you need to learn at least one hero of each playstyle below. This way you will be more versatile, but more importantly – you’ll have a deeper understanding of Dota as a whole.
Great heroes to learn the ultimate power in Dota - gold, and to learn to have your long-term priorities guide your short-term decisions. Most true pos. 1 hard carries involve a lot of split-farming, often split-pushing, and the fights they take need to be low-risk because deaths are extremely punishing.
A greedier, more selfish, safer, long-term oriented playstyle.
Priorities: Out-farming the enemy team and using your resource advantage to win in the mid/late game.
Fighting before you are ready and risking unnecessary deaths.
Over-extending while farming. (Such deaths are much, much more costly on such a hero than on any other in Dota. You need to learn immaculate positioning on the map in the mid/late game - to avoid death, but to maximize your farm and pressure the enemy team if possible.)
Farming for too long. In good situations, once you hit your timings, you need to have the confidence to lead the game and force fights. This is especially important for 5-man hard-farming carries. In bad situations, you need to notice moments in which your team simply cannot afford to play 4v5 anymore and to start contributing.
Great heroes to learn that farming cores could be powerful space-creators, not only space-takers. Most of those heroes need to find a good balance between fighting and farming, which usually means exploiting fight timings as best as you can.
Priorities: Playing very actively, while at the same time never stopping to farm. You create your own space (with the help of the team) and you use it efficiently.
Common mistakes: Playing passively as if you are a hard-farming pos. 1 hero. You can afford to do this only for short periods of time in order to get important levels/items. Playing too passively with such heroes usually means you'll fall behind the enemy carries. Only a viable option if you are the only real carry on the team and if you decide to do it, you need to build items to accelerate your farm (Battlefury, Maelstrom, items with sustain, etc.)
Players to Watch: Matumbaman, Nikobaby, Ana, K1Hector, Nisha, Agressif
Carries aren’t all about farming or fighting – your actual playstyle in fights also differs. Glass cannon carries are great heroes to learn correct positioning and timing in fights.
Priorities: Immaculate fight positioning and timing - join the fight at the right time, stay behind your frontline heroes, and disengage before things turn bad.
Common mistakes: Over-eagerness. You cannot start fights by yourself, and overextending in fights will cost you your life.
Players to Watch: W33, Qojqva
The opposite of glass cannons. Great heroes to be proactive on and to lead your team. Good heroes to learn offensive positioning in fights.
Priorities: Forcing fights at the right time; leading pushes while your teammates stand behind you. Coordinating your positioning with the teammates with saves, counter-initiation, etc.
Playing too defensively in fights: You need to be the wall that stands before your vulnerable allies and your opponents. You are the unfavorable hero to focus down in a fight, so you want to prevent enemies from focusing your teammates first.
Over-extending in fights: you need to stay close to your team. Chasing too deep will either get you or your more vulnerable teammates killed.
Players to Watch: Miracle-, Matumbaman, Fata, Somnus M
Active pos. 1 heroes often move together with the team, while some active pos. 2 heroes have the opportunity to make moves on their own. Such heroes allow you to learn to use your personal resource lead and strength to dictate the pace of the game and create space for your team without neglecting your own resource growth.
Priorities: To use your early-mid game level and item timings to create space and pressure as much as possible the enemy farming cores while at the same time farming efficiently.
Common mistake: Over-aggression. Most of those heroes (with exceptions) are versatile enough to adapt well into different play styles. Trying too hard to gank when there are no good ganking opportunities is a good way to make mistakes, die, and cripple yourself in the game.
Players to Watch: Topson, Abed, mid Sumail, mid MidOne
Great heroes to learn to use timings to force 5-man fights. Such heroes are rarely active on their own, but their strong fight timings make them perfect to be aggressive and push together with their team.
Priorities: Being aware of the important level and item timings and hitting them as safely and quickly as possible; exploiting said timings to take objectives with your team and win map control.
Players to Watch: Topson, Somnus M, mid Matumbaman and mid Fata, NoOne
Those heroes aren’t usually active in terms of making moves – instead, they apply a huge amount of pressure to the enemy team by constantly pushing a lane and forcing enemy rotations while at the same time being very hard to catch and kill.
Priorities: Dominating and constantly pressuring your part of the map while forcing enemy reactions;
Common mistakes: Being too static on the map: even though you spend most of your time dominating and pushing a particular part of the map, you still need to learn when you are needed in fights or when you need to go and apply pressure to another location;
Players to Watch: Zai, Mind Control, Matumbaman, Qojqva, Topson, NoOne
Two absolutely vital skills for team fights: initiation and counter-initiation; One requires map awareness and game sense, the other patience and fight awareness;
Players to Watch: Mind Control, Yapzor, GH-, Ceb, Jerax, Cr1t, FY
A greedy support is a good way to learn that resources are not only valuable on carries and that some heroes can aim to get important core items and hit powerful timings without being a dedicated farmer from the start of the game.
Priorities: a balance between the regular support duties and finding the needed space and time to get some farm of your own and hit your important timings;
Common mistakes: Too selfish or too selfless play: either extreme is usually bad on such heroes. One means you aren’t contributing enough to creating space and controlling the map, the other – that you don’t get enough resources and that your impact in the mid/late game will be much lower than expected.
Players to Watch: GH-, Yapzor, Cr1t
A playmaking support is arguably the most impactful hero during the early game. Learning to play one is a great way to learn to rotate efficiently and to always find ways to be useful around the map without the need for a lot of resources.
Players to Watch: FY, Yapzor, Cr1t, Jerax, GH, Rodjer
Unlike the playmaker supports who are great for initiating gank rotations, the 5-man supports are great at walking with the active players on their team and forcing skirmishes or full-on team-fights. This is probably the most typical support playstyle and one that doesn’t require a lot of intense mechanical play, which leaves you a lot of free time to invest in looking all over the map and leading your team on voice chat.
Players to Watch: Puppey, Kuroky, Fly, Yapzor, n0tail
The perfect role to learn great defensive positioning in fights and perfect spell timing.
Priorities: great positioning and timing in fights: to stay safe in the fight, but close enough in order to be able to instantly react and save your teammates
Common mistakes: bad positioning: you are a vulnerable support, and dying too early in the fight means you won’t be able to use your saving mechanisms which your team relies on. If you have sustain, it’s important that you don’t die in the fight at all in order to be able to give this sustain to your teammates after the engagement and keep the aggression going to claim objectives.
Players to Watch: Fly, Puppey