diff --git a/newsfragments/4879.fixed.md b/newsfragments/4879.fixed.md
new file mode 100644
index 00000000000..2ad659b0786
--- /dev/null
+++ b/newsfragments/4879.fixed.md
@@ -0,0 +1 @@
+ * fixed spurious `test_double` failures.
diff --git a/pytests/tests/test_othermod.py b/pytests/tests/test_othermod.py
index ff67bba435c..f2dd9ad8fd2 100644
--- a/pytests/tests/test_othermod.py
+++ b/pytests/tests/test_othermod.py
@@ -3,11 +3,16 @@
 
 from pyo3_pytests import othermod
 
-INTEGER32_ST = st.integers(min_value=(-(2**31)), max_value=(2**31 - 1))
+INTEGER31_ST = st.integers(min_value=(-(2**30)), max_value=(2**30 - 1))
 USIZE_ST = st.integers(min_value=othermod.USIZE_MIN, max_value=othermod.USIZE_MAX)
 
 
-@given(x=INTEGER32_ST)
+# If the full 32 bits are used here, then you can get failures that look like this:
+# hypothesis.errors.FailedHealthCheck: It looks like your strategy is filtering out a lot of data.
+# Health check found 50 filtered examples but only 7 good ones.
+#
+# Limit the range to 31 bits to avoid this problem.
+@given(x=INTEGER31_ST)
 def test_double(x):
     expected = x * 2
     assume(-(2**31) <= expected <= (2**31 - 1))