diff --git a/docs/conf.py b/docs/conf.py
index 1f7754a..02f2f13 100644
--- a/docs/conf.py
+++ b/docs/conf.py
@@ -31,7 +31,8 @@ MOCK_MODULES = [
     "notmuch",
     "requests",
     "bs4",
-    "novaclient.v2"
+    "novaclient.v2",
+    "dota2py",
 ]
 
 for mod_name in MOCK_MODULES:
diff --git a/i3pystatus/dota2wins.py b/i3pystatus/dota2wins.py
new file mode 100644
index 0000000..9237c24
--- /dev/null
+++ b/i3pystatus/dota2wins.py
@@ -0,0 +1,105 @@
+from dota2py import api
+from i3pystatus import IntervalModule
+
+
+class Dota2wins(IntervalModule):
+    """
+    Displays the win/loss ratio of a given Dota account.
+    Requires: dota2py
+    """
+
+    settings = (
+        ("matches", "Number of recent matches to calculate"),
+        ("steamid", "Steam user to track"),
+        ("steam_api_key", "Steam API key "
+            "(http://steamcommunity.com/dev/apikey)"),
+        ("good_threshold", "Win percentage (or higher) which you are happy "
+            "with"),
+        ("bad_threshold", "Win percentage you want to be alerted (difference "
+            "between good_threshold and bad_threshold is cautious_threshold)"),
+        ("interval", "Update interval (games usually last at least 20 min)."),
+        ("good_color", "Color of text while win percentage is above "
+            "good_threshold"),
+        ("bad_color", "Color of text while win percentage is below "
+            "bad_threshold"),
+        ("caution_color", "Color of text while win precentage is between good "
+            "and bad thresholds"),
+        ("screenname", "If set to 'retrieve', requests for the users's "
+            "screenname via API calls. Else, use the supplied string as the "
+            "user's screename"),
+        "format"
+    )
+    required = ("steamid", "steam_api_key")
+    good_color = "#00FF00"     # green
+    caution_color = "#FFFF00"  # yellow
+    bad_color = "#FF0000"      # red
+    good_threshold = 50
+    bad_threshold = 45
+    matches = 25
+    interval = 1800
+    screenname = 'retrieve'
+    format = "{screenname} {wins}W:{losses}L {win_percent}%"
+
+    def run(self):
+        api.set_api_key(self.steam_api_key)
+        hist = api.get_match_history(account_id=self.steamid)['result']
+        recent_matches = []
+        while len(recent_matches) < self.matches:
+            recent_matches.append(hist['matches'].pop(0))
+
+        player_team_per_match = []
+        # create a list of tuples where each tuple is:
+        # [match_id, bool]
+        # The bool will be true if the player is on Radiant and alse if they
+        # are on Dire.
+        for match in recent_matches:
+            this_match = [match['match_id']]
+            for player in match['players']:
+                # 64bit player ID
+                long_id = player['account_id'] + 76561197960265728
+                if long_id == self.steamid:
+                    if player['player_slot'] < 128:
+                        this_match.append(True)
+                    else:
+                        this_match.append(False)
+            player_team_per_match.append(this_match)
+
+        outcomes = []
+        for match in player_team_per_match:
+            if api.get_match_details(match[0])['result']['radiant_win'] == match[1]:
+                outcomes.append(1)
+            else:
+                outcomes.append(0)
+
+        wins = outcomes.count(1)
+        losses = outcomes.count(0)
+        win_percent = float(sum(outcomes) / float(len(outcomes))) * 100
+
+        if win_percent >= float(self.good_threshold):
+            color = self.good_color
+        elif win_percent <= float(self.bad_threshold):
+            color = self.bad_color
+        else:
+            color = self.caution_color
+
+        if self.screenname == 'retrieve':
+            from urllib.request import urlopen
+            import json
+            response = urlopen(
+                'http://api.steampowered.com/ISteamUser/GetPlayerSummaries/v0002/?key=%s&steamids=%s' %
+                (self.steam_api_key, self.steamid))
+            screenname = json.loads(bytes.decode(response.read()))['response']['players'][0]['personaname']
+        else:
+            screenname = self.screenname
+
+        cdict = {
+            "screenname": screenname,
+            "wins": wins,
+            "losses": losses,
+            "win_percent": win_percent,
+        }
+
+        self.output = {
+            "full_text": self.format.format(**cdict),
+            "color": color
+        }