diff --git a/helpers/refractor_test_data.py b/helpers/refractor_test_data.py
new file mode 100644
index 0000000..eef30ec
--- /dev/null
+++ b/helpers/refractor_test_data.py
@@ -0,0 +1,43 @@
+"""Pure helper functions for the /dev refractor-test command.
+
+Builds synthetic game data to push a card over its next refractor
+tier threshold with the minimum number of plays.
+"""
+
+import math
+
+# Batter: value = PA + (TB * 2). A HR play: PA=1, TB=4 → value = 1 + 8 = 9
+BATTER_VALUE_PER_PLAY = 9
+
+# Pitcher: value = IP + K. A K play: outs=1 (IP=1/3), K=1 → value = 1/3 + 1 = 4/3
+PITCHER_VALUE_PER_PLAY = 4 / 3
+
+
+def calculate_plays_needed(gap: int, card_type: str) -> dict:
+    """Calculate the number of synthetic plays needed to close a refractor gap.
+
+    Args:
+        gap: Points needed to reach the next tier threshold.
+             A gap of 0 means the card is exactly at threshold — we still
+             need 1 play to push past it.
+        card_type: One of "batter", "sp", "rp".
+
+    Returns:
+        dict with keys:
+            num_plays: int — number of plays to create
+            total_value: float — total refractor value those plays will add
+            value_per_play: float — value each play contributes
+    """
+    if card_type == "batter":
+        value_per_play = BATTER_VALUE_PER_PLAY
+    else:
+        value_per_play = PITCHER_VALUE_PER_PLAY
+
+    num_plays = max(1, math.ceil(gap / value_per_play))
+    total_value = num_plays * value_per_play
+
+    return {
+        "num_plays": num_plays,
+        "total_value": total_value,
+        "value_per_play": value_per_play,
+    }
diff --git a/tests/test_refractor_test_data.py b/tests/test_refractor_test_data.py
new file mode 100644
index 0000000..e146e6f
--- /dev/null
+++ b/tests/test_refractor_test_data.py
@@ -0,0 +1,49 @@
+import math
+
+import pytest
+
+from helpers.refractor_test_data import calculate_plays_needed
+
+
+class TestCalculatePlaysNeeded:
+    """Test the pure function that computes how many synthetic plays
+    are needed to push a card's refractor value over the next tier
+    threshold. The formulas are:
+      - batter: each HR play = 9 value (1 PA + 4 TB * 2)
+      - sp/rp:  each K play  = 4/3 value (1/3 IP + 1 K)
+    """
+
+    def test_batter_exact_threshold(self):
+        """When the gap is exactly divisible by 9, no extra plays needed."""
+        result = calculate_plays_needed(gap=27, card_type="batter")
+        assert result["num_plays"] == 3
+        assert result["total_value"] == 27
+        assert result["value_per_play"] == 9
+
+    def test_batter_rounds_up(self):
+        """When gap isn't divisible by 9, round up to overshoot."""
+        result = calculate_plays_needed(gap=10, card_type="batter")
+        assert result["num_plays"] == 2  # ceil(10/9) = 2
+        assert result["total_value"] == 18
+
+    def test_batter_gap_of_one(self):
+        """Even a gap of 1 requires one play."""
+        result = calculate_plays_needed(gap=1, card_type="batter")
+        assert result["num_plays"] == 1
+        assert result["total_value"] == 9
+
+    def test_sp_exact_threshold(self):
+        """SP: each K play = 4/3 value."""
+        result = calculate_plays_needed(gap=4, card_type="sp")
+        assert result["num_plays"] == 3  # ceil(4 / (4/3)) = 3
+        assert result["value_per_play"] == pytest.approx(4 / 3)
+
+    def test_rp_same_as_sp(self):
+        """RP uses the same formula as SP."""
+        result = calculate_plays_needed(gap=4, card_type="rp")
+        assert result["num_plays"] == 3
+
+    def test_zero_gap_returns_one_play(self):
+        """If already at threshold, still need 1 play to push over."""
+        result = calculate_plays_needed(gap=0, card_type="batter")
+        assert result["num_plays"] == 1