@@ -0,0 +1,30 @@
+import numpy as np
+import timeit
+
+setups = {
+    "hit": """data = np.float64(0.1234)""",
+    "miss": """data = float(0.1234)""",
+}
+
+tests = {
+    "eafp": """
+try:
+    data.dtype
+except AttributeError:
+    pass
+    """,
+    "lbyl": """if hasattr(data, "dtype"): data.dtype""",
+}
+
+results = {}
+for setup_name, setup_stmt in setups.items():
+    for test_name, test_stmt in tests.items():
+        test_name = f"{setup_name}, {test_name}"
+        times = timeit.repeat(
+            setup=setup_stmt, stmt=test_stmt, repeat=10, number=100000, globals={"np": np}
+        )
+        avg_time = sum(times) / len(times)
+        results[test_name] = avg_time
+
+for test_name, result in sorted(results.items(), key=lambda x: x[1]):
+    print(f"{test_name:>18}: {result:03g}")
@@ -0,0 +1,30 @@
+import pandas as pd
+import timeit
+
+setups = {
+    "hit": """data = pd.Series([0.1234])""",
+    "miss": """data = [0.1234]""",
+}
+
+tests = {
+    "eafp": """
+try:
+    data.dtype
+except AttributeError:
+    pass
+    """,
+    "lbyl": """if hasattr(data, "dtype"): data.dtype""",
+}
+
+results = {}
+for setup_name, setup_stmt in setups.items():
+    for test_name, test_stmt in tests.items():
+        test_name = f"{setup_name}, {test_name}"
+        times = timeit.repeat(
+            setup=setup_stmt, stmt=test_stmt, repeat=10, number=100000, globals={"pd": pd}
+        )
+        avg_time = sum(times) / len(times)
+        results[test_name] = avg_time
+
+for test_name, result in sorted(results.items(), key=lambda x: x[1]):
+    print(f"{test_name:>18}: {result:03g}")