diff --git a/stestr/commands/failing.py b/stestr/commands/failing.py
index 87818a01..9d4bb399 100644
--- a/stestr/commands/failing.py
+++ b/stestr/commands/failing.py
@@ -113,7 +113,7 @@ def failing(repo_type='file', repo_url=None, list_tests=False, subunit=False,
         case.run(result)
     finally:
         result.stopTestRun()
-    failed = not summary.wasSuccessful()
+    failed = not results.wasSuccessful(summary)
     if failed:
         result = 1
     else:
diff --git a/stestr/commands/last.py b/stestr/commands/last.py
index 46f7ce9d..2fd2540c 100644
--- a/stestr/commands/last.py
+++ b/stestr/commands/last.py
@@ -154,7 +154,7 @@ def last(repo_type='file', repo_url=None, subunit_out=False, pretty_out=True,
             case.run(output_result)
         finally:
             output_result.stopTestRun()
-        failed = not summary.wasSuccessful()
+        failed = not results.wasSuccessful(summary)
     else:
         stream = latest_run.get_subunit_stream()
         failed = subunit_trace.trace(stream, stdout, post_fails=True,
diff --git a/stestr/commands/load.py b/stestr/commands/load.py
index 673b57c3..8d6304ba 100644
--- a/stestr/commands/load.py
+++ b/stestr/commands/load.py
@@ -219,7 +219,7 @@ def make_tests():
     if pretty_out and not subunit_out:
         subunit_trace.print_fails(stdout)
         subunit_trace.print_summary(stdout, elapsed_time)
-    if not summary_result.wasSuccessful():
+    if not results.wasSuccessful(summary_result):
         return 1
     else:
         return 0
diff --git a/stestr/commands/run.py b/stestr/commands/run.py
index 8a5df531..5266f6ae 100644
--- a/stestr/commands/run.py
+++ b/stestr/commands/run.py
@@ -29,6 +29,7 @@
 from stestr import output
 from stestr.repository import abstract as repository
 from stestr.repository import util
+from stestr import results
 from stestr.testlist import parse_list
 from stestr import user_config
 
@@ -346,7 +347,7 @@ def run_tests():
                         stream.run(summary)
                     finally:
                         summary.stopTestRun()
-                    if not summary.wasSuccessful():
+                    if not results.wasSuccessful(summary):
                         result = 1
                 if result:
                     return result
@@ -495,7 +496,7 @@ def run_tests():
                         stream.run(summary)
                     finally:
                         summary.stopTestRun()
-                    if not summary.wasSuccessful():
+                    if not results.wasSuccessful(summary):
                         result = 1
                 if result:
                     return result
diff --git a/stestr/results.py b/stestr/results.py
index 13e9d249..5e630271 100644
--- a/stestr/results.py
+++ b/stestr/results.py
@@ -17,6 +17,11 @@
 from stestr import output
 
 
+def wasSuccessful(summary):
+    return not (summary.errors or summary.failures or
+                summary.unexpectedSuccesses)
+
+
 class SummarizingResult(testtools.StreamSummary):
 
     def __init__(self):
diff --git a/stestr/subunit_trace.py b/stestr/subunit_trace.py
index 1461a859..a1f8a72c 100644
--- a/stestr/subunit_trace.py
+++ b/stestr/subunit_trace.py
@@ -29,6 +29,7 @@
 import testtools
 
 from stestr import colorizer
+from stestr import results
 
 # NOTE(mtreinish) on python3 anydbm was renamed dbm and the python2 dbm module
 # was renamed to dbm.ndbm, this block takes that into account
@@ -388,7 +389,7 @@ def trace(stdin, stdout, print_failures=False, failonly=False,
     if count_tests('status', '^success$') == 0:
         print("\nNo tests were successful during the run")
         return 1
-    return 0 if summary.wasSuccessful() else 1
+    return 0 if results.wasSuccessful(summary) else 1
 
 
 def main():
diff --git a/stestr/tests/files/failing-tests b/stestr/tests/files/failing-tests
index 78efc938..5d9dbf0b 100644
--- a/stestr/tests/files/failing-tests
+++ b/stestr/tests/files/failing-tests
@@ -21,3 +21,7 @@ class FakeTestClass(testtools.TestCase):
     def test_pass_list(self):
         test_list = ['test', 'a', 'b']
         self.assertIn('fail', test_list)
+
+    def test_unexpected_pass(self):
+        self.expectFailure("we are sad",
+                           self.assertEqual, 1, 1)
diff --git a/stestr/tests/files/passing-tests b/stestr/tests/files/passing-tests
index a55cb1b5..3d9b2728 100644
--- a/stestr/tests/files/passing-tests
+++ b/stestr/tests/files/passing-tests
@@ -21,3 +21,7 @@ class FakeTestClass(testtools.TestCase):
     def test_pass_list(self):
         test_list = ['test', 'a', 'b']
         self.assertIn('test', test_list)
+
+    def test_xfail(self):
+        self.expectFailure("we are sad",
+                           self.assertEqual, 1, 0)
diff --git a/stestr/tests/test_return_codes.py b/stestr/tests/test_return_codes.py
index d4e9bafb..6ae99c22 100644
--- a/stestr/tests/test_return_codes.py
+++ b/stestr/tests/test_return_codes.py
@@ -127,6 +127,12 @@ def test_parallel_passing_bad_regex(self):
     def test_parallel_fails(self):
         self.assertRunExit('stestr run', 1)
 
+    def test_parallel_passing_xfail(self):
+        self.assertRunExit('stestr run xfail', 0)
+
+    def test_parallel_fails_unxsuccess(self):
+        self.assertRunExit('stestr run unexpected', 1)
+
     def test_parallel_blacklist(self):
         fd, path = tempfile.mkstemp()
         self.addCleanup(os.remove, path)