Spectre
Our own data, from ranked matches above 7500 MMR.

Adaga Espectral
A Spectre lança uma adaga para criar um Caminho Sombrio , causando dano e reduzindo a velocidade de movimento dos inimigos ao longo do rastro. Unidades atingidas pela adaga também criarão um Caminho Sombrio . Enquanto caminha por ele, a Spectre se move mais rápido e pode passar por terrenos antes intransponíveis.

Andar nas Sombras
Envia uma ilusão incontrolável rumo ao alvo para segui-lo e atacá-lo. A sub-habilidade Realidade pode ser usada para se teletransportar para a ilusão, destruindo-a.

Dispersão
A Spectre dispersa parte do dano causado a ela, refletindo-o contra todos os inimigos próximos. O efeito é reduzido com a distância.

Reality
Spectre assumes the target Shadow Step or Haunt illusion, destroying it.

Desolar
A Spectre e as suas ilusões causam dano puro adicional ao atacar um inimigo que não tenha outra unidade aliada em até 350 unidades de distância.

Assombrar
Cria nêmesis espectrais para assombrar todos os heróis inimigos. Essas ilusões são incontroláveis, ignoram terrenos, sofrem dano adicional e causam dano inferior ao da Spectre. A sub-habilidade Realidade pode ser usada para se teletransportar para uma ilusão, destruindo-a.
Skill order
1
2
3
4
5
6
7
8
9
10- 11
- +212
13- +214
- +215
- 16
- +218
- +220
- +225
Talents
- +4% Dispersion27% · WR 66%25+12% All Spectre Illusion Damage73% · WR 72%
- +15% Spectral Dagger Slow/Bonus24% · WR 70%20+300 Health76% · WR 63%
- +80 Spectral Dagger Damage16% · WR 50%15+1.0s Shadow Step Duration84% · WR 59%
- +12 Desolate Damage92% · WR 53%10-3s Spectral Dagger Cooldown8% · WR 57%
Items
Starting kits




9556%
2

7953%



5547%
Starting items
97%53%
85%53%
80%53%
72%52%
65%52%
53%51%
29%54%
6%45%
5%
6%62%
Core, in the order it usually comes out
min 1587%56%+11.4 pp
min 1785%56%+11.8 pp
min 2188%57%+9.6 pp
min 2481%60%+10.5 pp
min 2849%69%+11.1 pp
min 3243%72%+10.2 pp
Situational
min 799%53%+7.8 pp
min 3436%65%+9.3 pp
min 3525%73%+10.4 pp
min 3624%69%+4.7 pp
min 2822%64%+13.3 pp
min 2921%64%+11.6 pp
min 4012%
min 3817%76%+11.8 pp
min 3916%75%
min 3815%73%
min 4010%
min 2915%
min 3715%66%
Did this build work for you?
The numbers come from the top of the ladder. What they do not say is whether the build holds up in the games played here. Your answer is what adjusts that.
Trend in the 7500+ ladder
510 ranked matches in the last 8 weeks
- Win rate
- 53.3%
- Pick rate
- 3.1%
Rating: 70% win rate (adjusted for sample size) and 30% pick rate, compared with heroes in the same position. S is the top 10%, A the next 20%, B 30%, C 25%, D the rest.
Spectre worldwide
OpenDota, public ranked matches from the last 7 days
Win rate
53.1%
322,938 matches
Pick rate
9.1%
of matches include the hero
Pro scene
2 picks and 2 bans
0.0% win rate when picked
Win rate by rank bracket
- Herald and Guardian54.5% · 26,844 matches
- Crusader and Archon53.1% · 121,308 matches
- Legend and Ancient52.9% · 126,426 matches
- Divine and Immortal52.5% · 22,694 matches
Win rate by match duration
- up to 25 min36.9% · 1,663 matches
- 25 to 35 min43.8% · 4,288 matches
- 35 to 45 min51.7% · 3,330 matches
- over 45 min54.8% · 2,027 matches
Advantage in points over Spectre's usual win rate: negative means trouble. World data comes first (STRATZ and OpenDota); our high-MMR matches take over once the sample is big enough.
Worst counters
Meepo43.5% win rate in 972 games · STRATZ-9.2
Elder Titan45.7% win rate in 442 games · STRATZ-7.0
Primal Beast45.9% win rate in 1,583 games · STRATZ-6.8
Phantom Lancer46.3% win rate in 6,094 games · STRATZin our high-MMR matches: 50.0% in 42 games-6.4
Phoenix46.7% win rate in 2,397 games · STRATZ-6.0
Brewmaster46.6% win rate in 1,098 games · STRATZin our high-MMR matches: 52.2% in 23 games-6.0
Visage47.0% win rate in 728 games · STRATZ-5.7
Lifestealer47.1% win rate in 8,579 games · STRATZin our high-MMR matches: 55.6% in 36 games-5.6
Best synergies
Meepo58.9% win rate in 842 games · STRATZ+6.2
Bloodseeker57.1% win rate in 889 games · STRATZ+4.5
Naga Siren57.0% win rate in 128 games · STRATZ+4.4
Arc Warden56.8% win rate in 2,576 games · STRATZ+4.2
Riki56.5% win rate in 2,315 games · STRATZ+3.8
Outworld Destroyer56.4% win rate in 7,180 games · STRATZin our high-MMR matches: 53.2% in 62 games+3.7
Dawnbreaker56.4% win rate in 5,297 games · STRATZin our high-MMR matches: 54.5% in 44 games+3.7
Winter Wyvern56.4% win rate in 110 games · our high-MMR matches+3.5
Spectre: you fight with your ult, not with Radiance
Spectre is the carry who looks easiest in replay reviews: the student farms well, finishes Radiance, reaches minute 30 with a good net worth. And loses. Across Spectre's 17 lessons in the catalog, the sentence that sums up the hero came up in a Legend review: "you fight with your ult, you don't fight all the time". Spectre has one ability that decides the game and a long cooldown between one use and the next. Almost everything that goes wrong comes from treating that resource as if it were cheap.
A wasted Haunt is minutes of punishment
In the review of a Legend 5 Spectre, Radiance comes out at minute 18 and the first ult worth anything only at 23. In between, the student uses Haunt to secure a Rubick who was already dead. In the coach's words: "now you're two minutes in the penalty box doing nothing, no ult, not helping your team". And later, after the second wasted ult: since the cooldown is long, getting one wrong compromises the next three minutes of the game.
The mistake shows up early too. In an Ancient 3 Spectre game, the player ults at 2:30 onto a Keeper of the Light, arrives alone, ends up in a two versus three and dies. In an Archon game, Spectre ults in before the team arrives and eats focus from all five.
The coach's rule for Haunt fits in three questions:
- Who do I want to kill with this ult? Almost always the support in the back: Dazzle, Keeper of the Light, Winter Wyvern, Crystal Maiden.
- What do I need to kill them? Usually, a stun. "You wait for the stun."
- Is my team going to show up? If not, the ult isn't yours, it's the enemy's.
Radiance is a call to fight, not an item
Radiance is not the end of the preparation, it is the start of the game. The coach's advice for the Legend 5 Spectre is almost literal: finish Radiance at the Secret Shop, open your mic and say "guys, let's fight at the mid wave, I have Radiance". A Radiance that spends five minutes farming jungle without joining a fight is gold sitting still.
Farming too much jungle is the hero's second big mistake. In the review of another Legend 5 Spectre, the student spends 15 minutes going from camp to camp without touching a wave, and only finishes Radiance at 27. "You go to the jungle when you're under pressure and can't farm the lane. But there's a lane for you to farm." The jungle keeps Spectre permanently low on HP, buying salves, and out of the fights where the ult would make a difference.
The same goes for the side of the map. In an Ancient 4 Spectre game, the team does Roshan at minute 18 while she farms near the Tormentor on the other side. Fight lost, and a 51 minute game against Phantom Lancer that did not need to last that long.
Target: the support who saves, not the tank who absorbs
The replays repeat the same wrong focus: Spectre hitting Bristleback, a protected Troll Warlord, Phoenix's egg, Pudge. Meanwhile Rubick, Dazzle or Skywrath Mage unload their damage with nobody on them. In the review of an Ancient 3 Spectre, the coach shows that the right play was to land on the save supports first, force out Shallow Grave and Winter Wyvern's ult, and only then go for the carry.
The item the game asked for
Spectre scales endlessly, so the student tends to buy the item that looks strong, not the one that solves the game.
| Item bought | The problem | What the coach asked for |
|---|---|---|
| Butterfly | The enemy already had MKB or true strike | Skadi, Bloodthorn or Disperser |
| Heart of Tarrasque after Aegis | Takes a slot and doesn't cut enemy healing | Eye of Skadi |
| Radiance disassembled against Phantom Lancer | Loses area damage against illusions | Keep the Radiance |
| No Dust in 40 minutes | Glimmer Cape saving everyone | Dust always in the inventory |
Orchid appears in several reviews as the missing item to solo a slippery hero. And Urn of Shadows comes early, before boots and Radiance, because it gives mana, healing and kill potential in the phase where Spectre suffers the most.
What Spectre doesn't do
Spectre doesn't hit towers. The coach says it in an Ancient 4 review while explaining why few people pick the hero: a two minute cooldown on the ult and no pressure on buildings. That defines the plan. With an item timing, her objective is Roshan, with smoke and the team together, not the tier 2.
And it defines the buyback. In an Ancient 2 review, the student buys back with 20 seconds left on the respawn, gets no kill and goes from 28 thousand to 23 thousand gold. With the lead, Spectre's job is to keep farming the map up to six items, not to pay a fortune to show up 20 seconds earlier.
Before the next Haunt
The test is just one: if you press ult now and nobody dies, what happens over the next two minutes? If the answer is "nothing, I farm", fine. If it is "the enemy team will do Roshan and I'll have no ult", hold it. Spectre wins the game with a few well used Haunts, not with many spent chasing kills.
Locked on exactly what you wanted?
The whole catalog (real match analysis, by hero, position and patch) opens with the subscription.
See plansBuild of the players who play Spectre the most
7500+ ranked matches our radar captured in the last 60 days: accounts with at least 10 matches on the hero, and the build from their matches only. This is not the pro league: that one is in the hero's pros.
5 specialists, 101 matches, 66% win rate on the hero
- Shadow Moses39 matches, 74%
- Specialist 218 matches, 61%
- Specialist 316 matches, 63%
- Specialist 414 matches, 71%
- Specialist 514 matches, 50%
Items
Starting kits




3070%
2

1771%
2


967%
Starting items
97%67%
94%67%
87%69%
76%69%
67%63%
31%71%
31%61%
6%
Core, in the order it usually comes out
min 1586%71%
min 1783%73%
min 2190%70%
min 2483%71%
min 2859%80%
min 3248%83%
Situational
min 7100%66%
min 2727%74%
min 3626%77%
min 2925%72%
min 3522%59%
min 3721%76%
min 3717%65%
min 3817%71%
Skill order
Most played25 matches, 64% win rate
1
2
3
4
5
6
7
8
9
10- 11
- +212
13- +214
- +215
- 16
- +218
- +220
- +225
Highest win rate3 matches, 100% win rate
1
2
3
4
5
6
7
8
9
10- 11
- +212
13- +214
- +215
- 16
- +218
- +220
- +225
Who plays Spectre in pro Dota
Pro leagues since patch 7.33. Recent matches also include 7500+ ranked games our radar watches live.
20 matches shown, 45% win rate
10 min 20 min NW 4.2k 10.8k LH 57 144 DN 15 Lane +0.3k 











2





3 min
5 min
5 min
14 min
18 min
21 min
36 min
10 min 20 min NW 4.4k 11.6k LH 44 146 DN 5 Lane +0.8k 
![Death Prophet (]m[oqpay/o;iago) Death Prophet](/dota/herois/icone/death_prophet.webp)









2



4 min
5 min
6 min
15 min
18 min
21 min
25 min
30 min
34 min
10 min 20 min NW 3.6k 9.2k LH 54 160 DN 24 Lane +1.4k 











2



2 min
3 min
6 min
17 min
20 min
23 min
32 min
39 min
45 min
46 min
47 min
49 min
10 min 20 min NW 3.8k 9.2k LH 63 131 DN 7 Lane +0.6k 









2
4
2





3 min
4 min
5 min
18 min
21 min
24 min
25 min
27 min
32 min
34 min
37 min
10 min 20 min NW 3.6k 8.2k LH 46 118 DN 9 Lane +0.1k 









2
2



7 min
8 min
8 min
19 min
24 min
10 min 20 min NW 4.6k 10.9k LH 53 152 DN 11 Lane +0.7k 

















3 min
5 min
15 min
17 min
20 min
10 min 20 min NW 3.2k 10.3k LH 53 161 DN 3 Lane -0.1k 









2

2


3 min
5 min
8 min
13 min
19 min
22 min
27 min
31 min
10 min 20 min NW 3.3k 8.1k LH 44 114 DN 6 Lane -1.7k 













2



3 min
5 min
7 min
13 min
19 min
28 min
10 min 20 min NW 3.5k 9.5k LH 45 135 DN 11 Lane +0.6k 

















3 min
5 min
8 min
12 min
16 min
19 min
23 min
34 min
42 min
48 min
53 min

















4 min
6 min
10 min 20 min NW 3.8k 9.1k LH 41 104 DN 13 Lane +0.5k 

















4 min
5 min
7 min
13 min
17 min
22 min
30 min
10 min 20 min NW 3.4k 8.8k LH 49 141 DN 8 Lane +1.4k 












3
2


4 min
5 min
6 min
6 min
17 min
20 min
22 min
24 min
27 min
32 min
34 min
10 min 20 min NW 4.5k 10.5k LH 41 115 DN 8 Lane +0.3k 

















2 min
3 min
5 min
16 min
19 min
23 min
29 min
30 min
10 min 20 min NW 3.5k 7.2k LH 33 71 DN 4 Lane -0.1k 
















1 min
3 min
5 min
23 min
27 min
32 min
40 min
41 min
44 min
10 min 20 min NW 3.9k 9.5k LH 46 127 DN 1 Lane -1.4k 









2






3 min
5 min
5 min
9 min
15 min
20 min
27 min
10 min 20 min NW 4.6k 12.5k LH 51 157 DN 8 Lane +2.0k 

















4 min
6 min
15 min
15 min
17 min
19 min
10 min 20 min NW 3.0k 8.9k LH 35 115 DN 4 Lane -0.8k 


















4 min
8 min
18 min
18 min
18 min
21 min
25 min
32 min
40 min
44 min
48 min
10 min 20 min NW 4.0k 10.8k LH 44 131 DN 5 Lane +1.6k 

















4 min
7 min
7 min
16 min
19 min
22 min
10 min 20 min NW 4.0k 9.8k LH 45 168 DN 12 Lane +0.5k 
















2 min
3 min
7 min
8 min
17 min
20 min
23 min
26 min
26 min
29 min
33 min
35 min
10 min 20 min NW 4.3k 9.2k LH 58 136 DN 12 Lane +1.4k 


















3 min
6 min
17 min
21 min
24 min
28 min
33 min
40 min
40 min
42 min
Pro league and 7500+ ranked (seen live by our radar), newest first. In the tree, the lit branch is the talent the player took. In ranked, last hits and denies come from the replay and show up a few hours after the match. Position is estimated from gold per minute within the team.
Patch
Position
Tournaments
Minimum matches
With the item at the end of the match
| # | Player | W x L | Last | |||
|---|---|---|---|---|---|---|
| 1 | dx | 22 | 19 x 3 | 86.4% | 60.4% | 12/14/2023 |
| 2 | exhff | 33 | 20 x 13 | 60.6% | 52.7% | 11/29/2023 |
| 3 | tv/otakerdota | 24 | 15 x 9 | 62.5% | 52.2% | 09/01/2025 |
| 4 | Dronik | 19 | 12 x 7 | 63.2% | 51.6% | 03/05/2025 |
| 5 | WoE | 43 | 23 x 20 | 53.5% | 49.9% | 10/13/2026 |
| 6 | BoomBoom | 19 | 11 x 8 | 57.9% | 49.9% | 09/10/2025 |
| 7 | VW.Snickerdoodle | 15 | 9 x 6 | 60.0% | 49.9% | 01/01/2024 |
| 8 | med | 26 | 14 x 12 | 53.8% | 49.1% | 09/25/2024 |
| 9 | Хёма Чигири 10 место | 125 | 62 x 63 | 49.6% | 48.7% | 10/18/2025 |
| 10 | Bedbug | 24 | 12 x 12 | 50.0% | 47.6% | 10/23/2025 |
| 11 | donk | 16 | 8 x 8 | 50.0% | 47.2% | 12/05/2023 |
| 12 | why | 48 | 23 x 25 | 47.9% | 47.1% | 12/03/2023 |
| 13 | Crystallize | 15 | 7 x 8 | 46.7% | 46.2% | 10/25/2025 |
| 14 | Keu | 67 | 31 x 36 | 46.3% | 46.2% | 10/08/2025 |
| 15 | Bubiza | 33 | 15 x 18 | 45.5% | 45.8% | 09/26/2024 |
| 16 | Pudre | 22 | 9 x 13 | 40.9% | 44.3% | 05/10/2024 |
| 17 | ghfghfghfghfghfghfghghf | 18 | 7 x 11 | 38.9% | 43.9% | 12/12/2023 |
| 18 | Snémaât | 16 | 6 x 10 | 37.5% | 43.6% | 10/09/2025 |
| 19 | 999 | 21 | 8 x 13 | 38.1% | 43.3% | 12/06/2023 |
| 20 | ВСЯ ЭТА КРОВЬ МНЕ | 20 | 7 x 13 | 35.0% | 42.4% | 10/20/2023 |
| 21 | AKID | 25 | 9 x 16 | 36.0% | 42.2% | 08/15/2024 |
| 22 | JAM | 18 | 6 x 12 | 33.3% | 42.1% | 08/21/2025 |
| 23 | tinkywinky | 26 | 9 x 17 | 34.6% | 41.6% | 09/27/2024 |
| 24 | VittoBossG | 24 | 8 x 16 | 33.3% | 41.3% | 09/10/2024 |
| 25 | framoon | 34 | 12 x 22 | 35.3% | 41.1% | 12/02/2023 |
Adjusted pulls the win rate of players with few matches toward the hero's average in these filters (46.1%): 14 wins in 16 does not beat 92 in 134. Position is estimated from gold per minute within the team.