diff --git a/modyn/evaluator/internal/metrics/accuracy.py b/modyn/evaluator/internal/metrics/accuracy.py
index 540b7e27a..5c881d8df 100644
--- a/modyn/evaluator/internal/metrics/accuracy.py
+++ b/modyn/evaluator/internal/metrics/accuracy.py
@@ -24,9 +24,6 @@ def _batch_evaluated_callback(self, y_true: torch.Tensor, y_pred: torch.Tensor,
         self.total_correct += labeled_correctly
         self.samples_seen += batch_size
 
-        self.total_correct += labeled_correctly
-        self.samples_seen += batch_size
-
     def get_evaluation_result(self) -> float:
         if self.samples_seen == 0:
             self.warning("Did not see any samples.")
diff --git a/modyn/tests/evaluator/internal/metrics/test_accuracy.py b/modyn/tests/evaluator/internal/metrics/test_accuracy.py
index 379611c77..a197cb953 100644
--- a/modyn/tests/evaluator/internal/metrics/test_accuracy.py
+++ b/modyn/tests/evaluator/internal/metrics/test_accuracy.py
@@ -67,6 +67,7 @@ def test_accuracy() -> None:
     accuracy.evaluate_batch(y_true, y_pred, 6)
 
     assert accuracy.get_evaluation_result() == pytest.approx(1.0 / 3)
+    assert accuracy.samples_seen - accuracy.total_correct == 0 + 6 + 4
 
 
 def test_accuracy_invalid() -> None: