diff --git a/src/elexmodel/utils/math_utils.py b/src/elexmodel/utils/math_utils.py
index c501f9a7..5535b2f7 100644
--- a/src/elexmodel/utils/math_utils.py
+++ b/src/elexmodel/utils/math_utils.py
@@ -92,26 +92,25 @@ def compute_error(true, pred, type_="mae"):
     computes error. either mean absolute error or mean absolute percentage error
     """
     if type_ == "mae":
-        return np.mean(np.abs(true - pred)).round(decimals=0)
+        return np.mean(np.abs(true - pred))
     if type_ == "mape":
         mask = true != 0
-        mape = np.mean((np.abs(true - pred) / true)[mask])
+        mape = np.mean((np.abs((true - pred) / true))[mask])
         # if all true values are zero, then race was uncontested and mape doesn't make sense to compute
         if math.isnan(mape):
             return mape
-        return mape.round(decimals=2)
+        return mape
 
 
 def compute_frac_within_pi(lower, upper, results):
     """
     computes coverage of prediction intervals.
     """
-    return np.mean((upper >= results) & (lower <= results)).round(decimals=2)
+    return np.mean((upper >= results) & (lower <= results))
 
 
 def compute_mean_pi_length(lower, upper, pred):
     """
     computes average relative length of prediction interval
     """
-    # we add 1 since pred can be literally zero
-    return np.mean((upper - lower) / (pred + 1)).round(decimals=2)
+    return np.mean(np.abs(np.nan_to_num((upper - lower) / pred)))
diff --git a/tests/utils/test_math_utils.py b/tests/utils/test_math_utils.py
index 2f856dc1..3b6f9757 100644
--- a/tests/utils/test_math_utils.py
+++ b/tests/utils/test_math_utils.py
@@ -114,7 +114,7 @@ def test_compute_mape():
     # if multiple true values are zero
     y_true = pd.Series(np.asarray([0, 1, 4, 0, 5, 3]))
     y_pred = pd.Series(np.asarray([10, 4, 8, 20, 5, 8]))
-    mape = round((abs(1 - 4) / 1 + abs(4 - 8) / 4 + abs(5 - 5) / 5 + abs(3 - 8) / 3) / 4, 2)
+    mape = (abs(1 - 4) / 1 + abs(4 - 8) / 4 + abs(5 - 5) / 5 + abs(3 - 8) / 3) / 4
     assert math_utils.compute_error(y_true, y_pred, type_="mape") == pytest.approx(mape)
 
     # if all true values are zero
@@ -127,7 +127,7 @@ def test_compute_frac_within_pi():
     lower = np.asarray([0, 1, 4, 10, 5, 3])
     upper = np.asarray([10, 4, 8, 20, 5, 8])
     pred = np.asarray([5, 8, 5, 10, 5, 9])
-    assert math_utils.compute_frac_within_pi(lower, upper, pred) == round(4 / 6, 2)
+    assert math_utils.compute_frac_within_pi(lower, upper, pred) == pytest.approx(4 / 6)
 
 
 def test_compute_mean_pi_length():
@@ -135,4 +135,4 @@ def test_compute_mean_pi_length():
     lower = random_number_generator.normal(loc=5, scale=1, size=100)
     length = random_number_generator.lognormal(mean=1, sigma=5, size=100)
     upper = lower + length
-    assert math_utils.compute_mean_pi_length(lower, upper, 0) == np.mean(length).round(decimals=2)
+    assert math_utils.compute_mean_pi_length(lower, upper, 1) == np.mean(length)