Dota 2 seasonal rank distribution based on the data of millions of players. The MMR by medal is an estimate and is constantly adjusted. Updated to season 2.
I have been asked numerous times about the current rank distribution in Dota 2. We don’t have official and specific info from Valve, but thanks to the work of the community and some tools, it is possible to find a reasonable answer.
The following dataset has the purpose to create a plausible MMR distribution by collecting data from as many players as possible.
Unfortunately, a fraction of the player base is not available as you can’t gather data from Dota 2 profiles that keep their info hidden. Still, with over 4 million (season 2) and 5 million (season 1) players analyzed, we can consider this distribution very realistic.
Rank distribution - Season 2
The table is responsive and sortable. If you are using a phone, you may have to turn it sideways for a full view. Otherwise, zoom in or click the grey area to swipe and sort it.
|Herald 1||0.02||Crusader 1||26.16||Legend 1||75.79||Divine 1||98.67|
|Herald 2||0.32||Crusader 2||30.68||Legend 2||80.24||Divine 2||99.05|
|Herald 3||1.49||Crusader 3||35.41||Legend 3||84.15||Divine 3||99.31|
|Herald 4||3.40||Crusader 4||40.35||Legend 4||87.48||Divine 4||99.50|
|Herald 5||5.66||Crusader 5||45.39||Legend 5||90.24||Divine 5||99.63|
|Guardian 1||8.29||Archon 1||50.61||Ancient 1||92.60||Immortal||100.00|
|Guardian 2||11.23||Archon 2||55.84||Ancient 2||94.41|
|Guardian 3||14.47||Archon 3||61.02||Ancient 3||95.81|
|Guardian 4||18.02||Archon 4||66.08||Ancient 4||96.87|
|Guardian 5||21.91||Archon 5||70.97||Ancient 5||98.17|
According to these statistics, the average player is placed between Crusader 5 and Archon 1, while an Archon 5 is already way above the average hitting the 70 percentile.
Just a minuscule portion of the player base reached the top ranks, and there is also a huge skill difference between players belonging to the same medal but with a different number of stars.
Only 7% of the players are spread from Ancient 1 to Immortal, with Divine starting from above the 98 percentile. Many high ranked players often complain about the current state of the matchmaking, but with less than 1.3% of the player base in Divine, and even split in multiple servers, matching together individuals with similar skills isn’t an easy task.
Only 0.4% of the players have a rank higher than Divine 5! Pro players are a restricted elite group. I am not surprised by their long queue times and that the matchmaking is forced to place them in team with less skilled players.
I will update the data monthly, so we will have an idea on how the rank distribution evolves over time, and if there is any change.
|Herald 1||0.04||Crusader 1||27.27||Legend 1||76.87||Divine 1||98.76|
|Herald 2||0.60||Crusader 2||31.79||Legend 2||81.23||Divine 2||99.11|
|Herald 3||2.16||Crusader 3||36.55||Legend 3||84.97||Divine 3||99.35|
|Herald 4||4.23||Crusader 4||41.47||Legend 4||88.15||Divine 4||99.53|
|Herald 5||6.53||Crusader 5||46.52||Legend 5||90.81||Divine 5||99.65|
|Guardian 1||9.19||Archon 1||51.76||Ancient 1||93.04||Immortal||100.00|
|Guardian 2||12.16||Archon 2||56.96||Ancient 2||94.75|
|Guardian 3||15.44||Archon 3||62.12||Ancient 3||96.07|
|Guardian 4||19.05||Archon 4||67.18||Ancient 4||97.06|
|Guardian 5||22.98||Archon 5||72.07||Ancient 5||98.27|
|Herald 1||0.12||Crusader 1||28.36||Legend 1||77.92||Divine 1||98.83|
|Herald 2||1.09||Crusader 2||32.87||Legend 2||82.13||Divine 2||99.15|
|Herald 3||3.00||Crusader 3||37.64||Legend 3||85.74||Divine 3||99.38|
|Herald 4||5.14||Crusader 4||42.57||Legend 4||88.81||Divine 4||99.55|
|Herald 5||7.49||Crusader 5||47.62||Legend 5||91.33||Divine 5||99.67|
|Guardian 1||10.20||Archon 1||52.85||Ancient 1||93.44||Immortal||100.00|
|Guardian 2||13.20||Archon 2||58.03||Ancient 2||95.06|
|Guardian 3||16.51||Archon 3||63.16||Ancient 3||96.30|
|Guardian 4||20.13||Archon 4||68.26||Ancient 4||97.25|
|Guardian 5||24.09||Archon 5||73.16||Ancient 5||98.36|
Differences between July and June
The percentage of players increased slightly at lower ranks, and decreased at higher ones. Overall, we are talking about 1% to 3% adjustments, but they have an impact considering the vast number of players in the game.
The average Dota 2 player (50 percentile) moved down from high Archon 1 to high Crusader 5.
Note that the data may not be final as we are talking about a very fresh season.
|Herald 1||0.54||Crusader 1||24.29||Legend 1||74.11||Divine 1||98.19|
|Herald 2||1.67||Crusader 2||28.36||Legend 2||78.73||Divine 2||98.64|
|Herald 3||3.17||Crusader 3||32.78||Legend 3||82.78||Divine 3||99.00|
|Herald 4||4.79||Crusader 4||37.46||Legend 4||86.24||Divine 4||99.26|
|Herald 5||6.66||Crusader 5||42.39||Legend 5||89.19||Divine 5||99.45|
|Guardian 1||8.81||Archon 1||47.55||Ancient 1||91.60||Immortal||100.00|
|Guardian 2||11.29||Archon 2||52.79||Ancient 2||93.52|
|Guardian 3||14.05||Archon 3||58.17||Ancient 3||95.05|
|Guardian 4||17.09||Archon 4||63.61||Ancient 4||96.25|
|Guardian 5||20.53||Archon 5||68.96||Ancient 5||97.56|
Differences between June and Season 1
The distribution changed at lower ranks. Herald 5 increased from a 2.96 percentile in season 1 to 6.66. Guardian, Crusader, and Archon have been affected by small adjustments as well, but overall there is only a 4% difference between the start of the old and new Legend medal.
Rank distribution - Season 1
|Herald 1||0.05||Crusader 1||23.51||Legend 1||75.30||Divine 1||98.78|
|Herald 2||0.24||Crusader 2||27.25||Legend 2||79.47||Divine 2||99.10|
|Herald 3||0.80||Crusader 3||31.23||Legend 3||83.23||Divine 3||99.32|
|Herald 4||1.74||Crusader 4||35.37||Legend 4||86.29||Divine 4||99.48|
|Herald 5||2.96||Crusader 5||39.63||Legend 5||88.80||Divine 5||99.97|
|Guardian 1||6.30||Archon 1||48.60||Ancient 1||92.95|
|Guardian 2||8.42||Archon 2||53.11||Ancient 2||94.49|
|Guardian 3||10.85||Archon 3||57.61||Ancient 3||95.72|
|Guardian 4||13.59||Archon 4||62.05||Ancient 4||96.81|
|Guardian 5||16.61||Archon 5||66.41||Ancient 5||97.87|
|Herald 1||0.07||Crusader 1||24.60||Legend 1||76.74||Divine 1||98.92|
|Herald 2||0.34||Crusader 2||28.41||Legend 2||80.87||Divine 2||99.20|
|Herald 3||1.03||Crusader 3||32.45||Legend 3||84.41||Divine 3||99.40|
|Herald 4||2.06||Crusader 4||36.67||Legend 4||87.27||Divine 4||99.54|
|Herald 5||3.36||Crusader 5||40.98||Legend 5||89.64||Divine 5||99.96|
|Guardian 1||6.87||Archon 1||50.04||Ancient 1||93.57|
|Guardian 2||9.07||Archon 2||54.56||Ancient 2||94.99|
|Guardian 3||11.61||Archon 3||59.05||Ancient 3||96.13|
|Guardian 4||14.45||Archon 4||63.51||Ancient 4||97.10|
|Guardian 5||17.55||Archon 5||67.90||Ancient 5||98.06|
|Herald 1||0.09||Crusader 1||25.66||Legend 1||78.20||Divine 1||99.04|
|Herald 2||0.50||Crusader 2||29.54||Legend 2||82.26||Divine 2||99.28|
|Herald 3||1.31||Crusader 3||33.65||Legend 3||85.52||Divine 3||99.46|
|Herald 4||2.41||Crusader 4||37.96||Legend 4||88.21||Divine 4||99.58|
|Herald 5||4.18||Crusader 5||42.33||Legend 5||90.47||Divine 5||99.96|
|Guardian 1||7.44||Archon 1||51.46||Ancient 1||94.17|
|Guardian 2||10.14||Archon 2||55.99||Ancient 2||95.47|
|Guardian 3||12.36||Archon 3||60.48||Ancient 3||95.61|
|Guardian 4||15.29||Archon 4||64.96||Ancient 4||97.36|
|Guardian 5||18.47||Archon 5||69.39||Ancient 5||98.24|
|Herald 1||0.15||Crusader 1||26.55||Legend 1||79.49||Divine 1||99.14|
|Herald 2||0.71||Crusader 2||30.48||Legend 2||83.36||Divine 2||99.35|
|Herald 3||1.59||Crusader 3||34.68||Legend 3||86.40||Divine 3||99.51|
|Herald 4||2.76||Crusader 4||39.04||Legend 4||88.97||Divine 4||99.62|
|Herald 5||4.19||Crusader 5||43.44||Legend 5||91.21||Divine 5||99.96|
|Guardian 1||7.96||Archon 1||52.66||Ancient 1||94.66|
|Guardian 2||10.33||Archon 2||57.18||Ancient 2||95.86|
|Guardian 3||13.03||Archon 3||61.69||Ancient 3||96.82|
|Guardian 4||16.02||Archon 4||66.24||Ancient 4||97.57|
|Guardian 5||19.25||Archon 5||70.67||Ancient 5||98.36|
MMR distribution by Medal - Season 1
Comparing the old MMR to the new medals is a more difficult job. Again, we don’t have any info from Valve. We have access to the estimated MMR distribution pre-medals, but honestly that data has almost no value now.
After the release of the new system, clarinetEX held a survey, gathering data from over one thousand Dota 2 players about their calibration matches, the old MMR, and the medal achieved. Below, the final results.
Herald: 1 - 839
Guardian: 840 - 1709
Crusader: 1710 - 2579
Archon: 2580 - 3449
Legend: 3450 - 4319
Ancient: 4320 - 5189
Divine: 5190 - 6055+
The MMR by medal is an estimate and is constantly adjusted.
You must have noticed that your medals changed in the past months even when not gaining MMR - and sometimes even when losing it! I believe that Valve re-arranged the system over time to offer a more balanced division.
Most probably, medals aren’t based on a strict MMR, but on a percentile: If you are a Legend and remain stable on your rank while most Legend players are losing theirs, then you deserve a better one. But, this is just a theory.
Dota 2 Gamepedia displays a slightly different distribution, but both are based only on a rough estimation as only Valve has access to the data of all players.
Herald: 1 - 779
Guardian: 780 - 1559
Crusader: 1560 - 2369
Archon: 2370 - 3149
Legend: 3150 - 3929
Ancient: 3930 - 4709
Divine: 4710 - 5360+
Which distribution should you trust? None.
They are helpful to give you an idea of your actual position and the approximate MMR necessary to advance to the next medal, but overall they all have a margin of error.
It is also very hard to understand how/if the combination of solo and party MMR affects the medal. Hopefully we will have more info directly from Valve at the end of the season.
Note: the first ranked season ended, and Valve has not revealed any official data.
MMR distribution by Medal - Season 2
The second Dota 2 ranked season introduced several changes, such as the Immortal rank and a new number of stars per medal; from 1 to 5 instead of 0 to 5.
Learn more: Season 2 changes.
Also, for the first time ever, a Valve developer revealed publicly what influences our rank during the recalibration.
As soon as the new season commenced, many players gathered data to figure out the impact of the win rate in the placement games, the medal received with a specific MMR, and to compare the old MMR with the new one.
The first distribution after about 24h of data:
Herald: 1 - 899
Guardian: 900 - 1749
Crusader: 1750 - 2649
Archon: 2650 - 3349
Legend: 3350 - 4249
Ancient: 4250 - 5099
Divine: 5100 - 5780+
Unfortunately, all that data has almost no value now because Valve released a patch to readjust all the matches.
Dota 2 Update - June 8th, 2018
In the first 30 hours since recalibration, the matchmaker was overly sensitive on wins and losses and it resulted in players calibrating with bigger gains and losses than intended. We are processing all matches again and will be correcting the rank gains and losses throughout the day. Medals will be appear as undefined for a bit during this process.
Before the patch: you could earn or lose a huge quantity of MMR during the placement games. Players who won 8-9 games out of 10 boosted their rank by 500-800+
After the patch: the placement matches drastically reduced their impact and assigned just a bit more MMR than standard ranked games.
It helped those who were defeated in most of the games, reducing their MMR loss, but simultaneously punished all players who were extremely successful and secured a place in a higher medal.
Honestly, I am not sure if the patch has been caused by a real sensitivity mistake, or if the devs have been forced to revert the changes due to the community outrage. In fact, alongside the players happy for their new MMR, there were also thousands of negative comments from those who lost their hard-earned medals.
Below, the temporary season 2 distribution used by the Dota 2 wiki:
Herald: 1 - 799
Guardian: 800 - 1599
Crusader: 1600 - 2399
Archon: 2400 - 3199
Legend: 3200 - 3999
Ancient: 4000 - 4799
Divine: 4800 - 5440+
Header image, stats, and seasonal rank graphs: dota.rgp.io
Vincenzo is an esports writer with five years of experience. Former head editor for Natus Vincere, he has produced content for DreamHack, FACEIT, DOTAFire, 2P, and more. Follow him on @SkulzDota.