More sports
Dota 2 team ratings and series chances
Check live Dota 2 team rankings, pick two teams and see the chance to win one map and a Bo1, Bo3 or Bo5 series, plus how accurate the ratings have been.
OpenDota team ratingsUpdated 2026-10-28 19:06 UTC
Rating gap 28 points: Team Yandex wins a single map 54.0% of the time, and the Bo3 about 56 times in 100.
Bo1 vs Bo3 vs Bo5
Exact map score
| Map score | Chance | Fair price |
|---|---|---|
| Team Yandex 2–0 | 29.1% | 3.43 |
| Team Yandex 2–1 | 26.8% | 3.73 |
| Team Spirit 1–2 | 22.9% | 4.37 |
| Team Spirit 0–2 | 21.2% | 4.72 |
Top active teams by rating
| # | Team | Rating | W–L30 days |
|---|---|---|---|
| 1 | Team Yandex | 1,583 | 24–7 |
| 2 | Team Spirit | 1,555 | 20–10 |
| 3 | PARIVISION | 1,518 | 21–8 |
| 4 | Aurora Gaming | 1,404 | 9–7 |
| 5 | Natus Vincere | 1,336 | 2–4 |
| 6 | Team Liquid | 1,327 | 8–13 |
| 7 | GamerLegion | 1,324 | 11–7 |
| 8 | LGD Gaming | 1,311 | 7–9 |
| 9 | OG | 1,301 | 17–11 |
| 10 | Team Resilience | 1,298 | 9–2 |
| 11 | Yellow Submarine | 1,255 | 30–8 |
| 12 | Xtreme Gaming | 1,252 | 4–4 |
| 13 | Klim Sani4 | 1,218 | 0–0 |
| 14 | Team Nemesis | 1,214 | 3–5 |
| 15 | HULIGANI | 1,185 | 16–8 |
| 16 | Yakult Brothers | 1,161 | 8–9 |
| 17 | Team Synapse | 1,157 | 18–14 |
| 18 | Pipsqueak + 4 | 1,150 | 0–0 |
| 19 | PlayTime | 1,145 | 5–5 |
| 20 | Royal Bear | 1,140 | 11–1 |
| 21 | LEGION | 1,133 | 17–13 |
| 22 | Estar Backs | 1,126 | 0–0 |
| 23 | Team Cake | 1,124 | 5–1 |
| 24 | CyberHero | 1,123 | 21–16 |
2 active teams named after a betting brand are left out of this list.
How accurate is the rating?
Checked out of sample: for every map we rebuilt the ratings both teams had before it was played, so the forecast never saw the result. The higher-rated team won 61.7% of single maps. Its strongest favourites won less often than it said.
Bar: how often the favourite actually won. Red mark: the chance the rating gave.
- Bo3: the higher-rated team won 65.1% of 315 series (the rating gave it 67.6% on average).
- Bo5: the higher-rated team won 66.7% of 9 series (the rating gave it 63.6% on average).
How it is calculated
OpenDota rates every pro team with an Elo rating per map: the chance that A beats B in one map is 1 / (1 + 10^((rating B − rating A) / 400)), and after each map the winner takes 32 × (1 − its chance) points from the loser. We use exactly that chance – no extra tuning.
Series: each map is treated as independent with that chance; a best of three is won 2–0 or 2–1, a best of five 3–0, 3–1 or 3–2.
Accuracy check: walking back through recent pro maps we undo each rating change, which gives the ratings every team had before each map. Those pre-match chances are compared with the results. The same undo reproduces the last rating change OpenDota reports for each team, which confirms the formula.
The rating knows only results. It does not see roster changes, patches, stand-ins or which side a team plays, so treat it as a starting point.
Series and map handicap calculator →CS2 Major math →League of Legends series math →Margin and no-vig calculator →
Data: OpenDota – team ratings and recent pro matches, updated automatically.
How the Elo rating works
- Results feed the ratingRatings are built from pro map results in the OpenDota public API and update by themselves.
- A win takes pointsAfter each map the winner gains points and the loser loses the same amount – never more than 32 points per map.
- Equal teams split evenlyWhen both teams have the same rating, the winner gains 16 points.
- Upsets move more pointsIf the lower-rated team wins, it gains more than 16. An expected win by the favourite moves fewer points.
- The gap becomes a chanceMap chance = 1 ÷ (1 + 10^(−gap ÷ 400)). A gap of 100 gives 64.0%.
Rating gap to series chance
| Rating gap | One map (Bo1) | Bo3 series | Bo5 series |
|---|---|---|---|
| 50 | 57.1% | 60.6% | 63.2% |
| 100 | 64.0% | 70.5% | 74.9% |
| 200 | 76.0% | 85.5% | 90.6% |
Chances for the higher-rated team. Each map is treated as independent with the same chance. More on Bo1, Bo3 and Bo5.
What the rating sees and misses
Sees: map results
Who won each pro map and against whom. Nothing else goes into the number.
Sees: opponent strength
Beating a higher-rated team moves more points than beating a lower-rated one.
Misses: rosters and stand-ins
The rating does not know about roster changes or stand-ins; it only changes through map results.
Misses: patches
A new game patch can change which teams are strong. The rating only catches up through results.
Misses: sides and drafts
Radiant or Dire side, picks and bans are not part of the rating.
Rating chance vs a sportsbook price
Rating chance
- Built from map results only
- Has no margin inside
- Ignores roster news and patches
- Same method for every match
Sportsbook price
- Includes the bookmaker's margin
- Remove it first with the margin calculator
- Turn odds into % with the odds converter
- Can react to news the rating does not see
Questions
Where do these Dota 2 rankings come from?
From the OpenDota public API. The tool builds an Elo rating from recent pro map results and updates by itself; the label above the table shows the update date.
Does this page have data for The International 2026?
No. The International is the best-known event in Dota 2, but this page has no event data. Check the official event pages for dates and format.
Why is a Bo3 chance higher than the one-map chance?
The stronger team needs only two maps of three, so one bad map matters less. For example, 60% per map gives 64.8% in a Bo3 and 68.3% in a Bo5.
Why is a team missing from the list?
The table shows only active teams. Teams named after a betting brand are left out.
Is Dota 2 betting legal in Canada?
The answer depends on the laws of your province; this page is not legal advice. See esports betting in Canada for what to check.