Dota 2 Support Warding: Vision Lanes, Dewarding, and Map Control

Wards Answer Questions, They Do Not Decorate the Map

Almost every support learns warding the same way: memorise a list of cliff spots, walk to the nearest one, place an observer, repeat when it expires. That approach produces a map full of glowing icons and almost no useful information, because it treats wards as a chore to be completed rather than as a limited resource spent to answer a specific question.

The correct model is interrogative. Before you place a ward, name the question it answers. Is there an incoming rotation through the river? Is the enemy carry farming their own jungle or ours? Are they sitting in the Roshan pit? Is this camp blocked? Every ward you place should have a question attached, and a ward whose question you cannot state is a ward you should not have placed. This reframing matters because it immediately fixes the two most common vision failures: placing wards in places that show you things you already know, and placing wards that expire without ever seeing anything worth seeing.

Vision also has a lifetime and a context. A ward is worth the most in the sixty to ninety seconds before a specific planned action, and worth almost nothing in a part of the map nobody will walk through during its duration. A ward placed on the enemy high ground two minutes before your team intends to smoke for Roshan is enormously valuable. The exact same ward, placed with no follow-up plan, is a hundred gold of nothing.

After reading this guide you should be able to do four things. Name the question before every ward placement. Execute a complete vision sequence for a Roshan attempt, including the dewarding step that most supports skip. Block and contest neutral camps with wards and understand why spawn blocking works mechanically. And diagnose your own vision failures from a replay, using the enemy's reactions rather than guesswork to locate their wards.

Vision Mechanics: Radii, Durations, and What Blocks Sight

Observer wards provide vision but no true sight. Sentry wards provide true sight, which reveals invisible units and enemy wards, and they also give a small amount of ordinary vision. The radii are strongly asymmetric: an observer sees roughly sixteen hundred units in both day and night, while a sentry's true sight radius is roughly half that, around eight hundred units. That asymmetry is the entire mechanical basis of dewarding: you have to guess the enemy ward's location within roughly eight hundred units, which is why knowledgeable players deward by checking known cliff spots rather than by dropping sentries in open ground.

Durations are long, on the order of six to seven minutes for observers and a comparable window for sentries, though exact values have been retuned across patches and you should confirm the current numbers in the client. The practical consequence is that a ward is a multi-minute commitment, not a momentary one, and you should be tracking when each of your wards expires so that you re-ward on a schedule rather than discovering three dead wards after you lose a fight.

The day and night cycle is the other mechanic that drives vision value. Most heroes see roughly eighteen hundred units during the day and roughly eight hundred at night, so nightfall cuts natural hero vision by more than half. Some heroes are exceptions with extended night vision, and a handful of abilities and items modify it, but the general rule is that at night the map is dark and wards become dramatically more important relative to hero sight. This is why night is the correct time to attempt smoked rotations, and why placing observer wards before nightfall is worth more than placing them during the day.

Terrain blocks line of sight. Trees, cliffs and elevation create dead zones, and a ward placed on the high ground overlooking a low-ground corridor sees the corridor while a ward in the corridor sees almost nothing. This is the reason the community ward spots exist at all: they are specific pixels on specific cliffs where the sight line happens to cover a high-traffic path without being easily spotted. High ground also imposes a miss chance on attackers who lack vision of it, which is why vision around the Roshan pit and around base high ground decides fights outright.

Two more mechanics round out the toolkit. Smoke of Deceit hides you from enemy vision until you come within a short radius, roughly nine hundred units, of an enemy hero or tower, which means you can walk deep into dangerous territory to place a ward and then leave without the enemy knowing a ward exists. And neutral creeps spawn on a fixed cadence, roughly every minute starting around the one minute mark, only if their spawn box is empty of units of any kind, including wards. That last rule is the basis of camp blocking, and it is the highest-value trick in this entire guide.

The Vision Numbers Table

| Quantity | Value | Why it decides your placement |

| --- | --- | --- |

| Observer ward vision radius | roughly 1600 units, day and night | One observer covers a camp-to-lane corridor, not the whole map |

| Observer ward duration | roughly 6 to 7 minutes | Track expiry and re-ward on a timer, not on discovery |

| Sentry true sight radius | roughly 800 units | You must localise an enemy ward to within this radius to deward it |

| Observer cost | free, but limited by a shop stock that refills over time | You cannot buy unlimited wards at once; spend them deliberately |

| Sentry cost | costs gold each | Dewarding has a real economic cost; failed dewards hurt |

| Standard hero day vision | roughly 1800 units | Daytime rotations are spotted by heroes, not wards |

| Standard hero night vision | roughly 800 units | Night halves natural vision, so wards and smokes peak in value |

| Day and night phase length | a few minutes per phase | Time smokes, pushes and Roshan attempts around the flip |

| Neutral creep first spawn | around 1:00, then every 60 seconds if the box is empty | A ward inside the box permanently blocks that camp |

| Lane creep waves | every 30 seconds | Pull and stack timings derive from this cadence |

| Bounty runes | spawn at the start, then every few minutes | Contest timings and a reason to ward the bounty spots |

| River power runes | begin spawning a couple of minutes in, then every 2 minutes | Ward one or both rune spots to control rune fights |

| Deward bounty | a small gold reward for destroying an observer | Successful dewarding partly pays for its own sentries |

| Smoke break radius | roughly 900 units from enemy heroes and towers | Place deep wards from outside that radius |

Three Warding Sequences, Step by Step

Sequence One: The Opening Minute and the Pull Camp War

In the first minute, before creeps meet, your vision plan has three jobs: protect your own pull camp, contest the enemy's, and cover the rune or rotation path your lane is most exposed to. Walk to your safe lane's small camp and place a sentry in or near its spawn box if the enemy support is likely to block or contest it. Then place your observer on high ground where it sees both the lane approach and the path the enemy offlaner uses to contest your pull, so a single ward answers two questions.

Now understand why blocking works and how to use it offensively. Neutral creeps only spawn if nothing occupies their spawn box, and a ward counts as an occupying unit. Placing a ward inside the enemy's small camp permanently denies their support the ability to pull, which forces their lane to stay pushed and gives your carry a much easier time last hitting near your tower. Conversely, if you find your own camp blocked, you must deward it before the next spawn, which is a sentry you should budget for at the start rather than discovering at minute three.

Finally, note the clock. Because camps attempt to spawn roughly every minute, a camp blocked at the start stays blocked for its full duration, so the timing of your blocking ward relative to the first spawn attempt determines whether it works at all. Place it before the attempt, not after.

Sequence Two: The Roshan Setup, Ninety Seconds Out

Roshan attempts are decided by vision, not by damage. Here is the complete sequence. At ninety seconds out, smoke with your team and walk toward the pit using a path that keeps you outside the smoke break radius of enemy heroes and towers. Enter from the side where you already have vision, never blind.

At sixty seconds out, place a sentry at the pit entrance and inside the pit area on the side the enemy is likely to approach from, and check for existing enemy wards. Deward first, always, because placing an observer into an area where the enemy has vision simply tells them you are coming. Then place your observers: one on the high ground behind the pit so you can see the enemy approaching through their own jungle, and one on the approach corridor that their carry would farm toward. The purpose of the second ward is not to see Roshan; it is to tell you whether their carry is farming or rotating, which tells you whether you can commit or should back off.

At thirty seconds out, stop placing wards and start holding the corridor. Your wards have told you who is coming and from where. If you see one enemy approaching alone and your team is smoked, you now have a free kill and a free Roshan. If you see four enemies rotating, you leave and take a tower on the other side of the map, and the vision paid for itself by preventing a throw. Either outcome is correct, and both were purchased by the two wards you placed ninety seconds earlier.

Sequence Three: Playing From Behind with Defensive Vision

When your team has lost the outer towers, warding the river and the enemy jungle is not brave, it is wasted, because your team cannot walk there safely and the vision will expire unused. Instead, invert your entire plan. Your vision should now cover the entrances to the part of the map you still own: your jungle, your triangle, your shrine area, and the paths the enemy must walk to reach them.

Place observers on your own high ground at the junctions the enemy must pass through, and sentries beneath them to keep the area dewarded, because a defensive ward that the enemy can see is worse than no ward at all, it tells them where your team is comfortable. Then do the single most important thing: place a sentry at each of your own camp spawn boxes you suspect they have blocked, and clear their blocking wards. Camp blocking against a team that is already behind is devastating, because farming their own jungle is their only path back into the game.

The goal of defensive vision is not map control, it is information density in a small area. You want to know, within a few hundred units, exactly when the enemy enters your remaining space, so your team can fight five against three on your own terms or smoke out through a gap you can see is empty. Use the team scan to confirm which half of the map is empty before you commit to a smoke, and only then move.

Ward Purpose Matrix

| Ward purpose | Where it goes | When it is worth it | Typical failure |

| --- | --- | --- | --- |

| Rotation detection | high ground above a river or jungle corridor | Any time a lane is pushed past the river | Placed too far forward, so it sees the rotation too late to react |

| Farm protection | your own jungle entrances, on high ground | When your carry is the win condition | Placed in the jungle itself, where enemies see it and deward |

| Objective control | pit high ground plus one approach corridor | 60 to 90 seconds before a planned attempt | Placed during the attempt, when it is already too late |

| Aggressive pickoff | enemy jungle paths between camps | When you have a smoke and a hero who can solo kill | Placed unsmoked, so the enemy knows and dewards instantly |

| Dewarding | known cliff spots and common ward pixels | Whenever an enemy reacts suspiciously fast | Dropped in open ground, far outside true sight range |

| Camp blocking | inside an enemy neutral camp spawn box | Against supports who rely on pulling, or versus jungling carries | Placed after the spawn attempt, so it does nothing |

| Rune control | one or both river rune spots | When your mid laner or roamer can contest | Both spots warded while nobody can actually contest them |

| Base defense | your own remaining jungle junctions | After losing outer towers | Warding forward instead, where the vision expires unused |

Dewarding, Camp Blocking, and the Advanced Toolkit

Dewarding is a guessing game with very poor odds unless you reason it out. Because sentry true sight is roughly eight hundred units, dropping sentries in open ground wastes gold. Instead, treat dewarding as a checklist of known spots. Every cliff and tree gap that gives useful vision has a small number of exact placement pixels, and players overwhelmingly use the same ones. Walk to the highest-value suspect spots in order, and drop the sentry only when you are within range of the pixel you suspect.

Better still, use behaviour to localise the ward before you spend anything. If an enemy support instantly reacts to your carry stepping forward, or a hero who had no vision of you suddenly backs off, there is a ward nearby, and you can usually narrow it to the two or three spots that cover that specific piece of ground. Think backwards from what they saw, not forwards from a list of spots. This is also the right way to diagnose your own vision failures in a replay: pause on every death and ask which spot would have had to contain an enemy ward for them to know what they knew.

Camp blocking deserves its own paragraph because it is underused. Any unit inside a camp's spawn box prevents the camp from spawning, and wards are units. One ward can deny an enemy camp for its entire duration, which is minutes of denied farm for their carry or denied pull plays for their support. Check periodically for enemy blocking wards in your own camps, and reciprocate. In lane, blocking the enemy small camp forces their lane to stay pushed toward your tower, which is a quiet, permanent win for your carry that never shows up on the scoreboard.

Ward economy is a real skill. Observers are free but gated by a shop stock that refills over time, so you cannot buy unlimited numbers at once, and sentries cost gold, meaning failed dewards are a genuine economic setback for a support who is already poor. Prioritise: never spend two sentries on a guess when a single scan or a moment of behavioural reasoning would localise the ward first. Conversely, never skimp on the sentry before an objective, because an attempt into enemy vision loses the game.

Placement discipline separates good supports from adequate ones. Place wards while smoked when going deep, so the enemy cannot see you place them and cannot immediately deward. Place your observer after you deward, never before, or you hand them a target. Do not stack an observer and a sentry on the exact same pixel, because a single enemy sentry removes both at once; offset them by several hundred units. And spread your wards rather than clustering them, because overlapping vision is wasted vision.

Mistakes and Their Specific Fixes

| Mistake | Why it happens | The specific correction |

| --- | --- | --- |

| Warding the same three spots every game | Habit and a memorised list | Name the question before you place; if you cannot state it, do not place it |

| Placing observers before dewarding an area | Impatience | Always sentry first, then place the observer once the area is clean |

| Stacking observer and sentry on one pixel | Efficiency reflex | Offset them by several hundred units so one enemy sentry cannot kill both |

| Sentrying open ground | Assuming true sight is large | Only sentry near known ward pixels, within roughly 800 units |

| Warding deep while visible | Walking in without smoke | Smoke before any deep placement, and break the smoke outside enemy radius |

| Letting wards expire unused | No expiry tracking | Note placement times and re-ward on a timer; check the minimap icons each minute |

| Not blocking or checking the pull camp | Focus on lanes only | Ward the enemy small camp early and check your own every minute |

| Warding forward when behind | Copying the winning team's ward map | Invert to defensive junctions covering the space you still own |

| Buying sentries only after dying to invisibility | Reactive spending | Buy proactively before objectives and before nightfall |

| Overlapping two observers | Fear of missing something | Place the second ward where it answers a different question |

Practice Plan and Replay Diagnosis

Spend your first practice block in demo mode learning terrain rather than playing games. Pick one side of the map and learn twelve useful spots: four lane-protection spots, four jungle entrance spots, two rune-area spots and two aggressive deep spots. For each, note what question it answers. This takes an evening and it permanently raises the quality of every ward you place, because you stop walking to the nearest cliff and start walking to the spot that answers your question.

Second block, play five games with one rule: before placing any ward, type the question it answers in chat. This feels tedious for about two games and then becomes automatic. Third block, play five games where you track ward expiry explicitly, noting the game time of each placement and re-warding at expiry rather than on discovery. Fourth block, play five games where your first purchase includes a sentry for your own pull camp and a ward for the enemy pull camp, and check both every minute.

For replay diagnosis, watch with the enemy vision perspective enabled, which is the fastest way to learn. Look at three things. First, count how many of your wards expired without ever revealing an enemy hero, because that is your wasted ward percentage, and anything above roughly half means your placement questions are wrong. Second, pause on each of your deaths and identify which enemy ward gave them the information to kill you, then note whether you could have dewarded it with one sentry beforehand. Third, check the sixty seconds before every failed team fight or lost objective: in most cases you will find that your team walked into an area with no vision while the enemy had one or more wards covering it.

Track one number over ten games: enemy observers destroyed. A support who consistently dewards three or more observers per game is winning the vision war outright, because every destroyed observer is enemy gold spent for nothing and a window of map darkness your team can exploit. Pair it with deaths-while-smoked or deaths-in-your-own-jungle, because those are the deaths that vision was supposed to prevent.

Closing

Vision is the one resource in Dota that converts directly into every other resource. It decides which fights happen, which farm is safe, which smokes succeed, and which Roshan attempts end in a throw. Treating warding as a checklist produces a lit-up minimap and no information; treating it as a set of questions you ask the map, asked at the right moment and in the right place, produces the feeling that you always know where the enemy is. Learn the spots, learn the durations, deward before you ward, block camps, and re-ward on a timer. Do that and you will be the reason your team never walks into darkness again.