diff --git a/pyproject.toml b/pyproject.toml
index 02d4f00..e911c0d 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -18,7 +18,7 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
-dev = ["pytest>=8.0", "ruff>=0.5", "mypy>=1.10"]
+dev = ["pytest>=8.0", "ruff>=0.5,<0.16", "mypy>=1.10"]
 
 [build-system]
 requires = ["hatchling"]
diff --git a/src/trinity/adapters/drop.py b/src/trinity/adapters/drop.py
index 1314803..07fcbd4 100644
--- a/src/trinity/adapters/drop.py
+++ b/src/trinity/adapters/drop.py
@@ -130,9 +130,12 @@ def _maybe_unbox(segment: str) -> str:
     return boxed if boxed is not None else ""
 
 
-#: Surrounding punctuation stripped from a token, EXCLUDING the signs ``+``/``-`` — a
-#: leading sign is part of a number's value, not wrapping noise.
-_STRIP_EDGE = "".join(c for c in string.punctuation if c not in "+-")
+#: Surrounding punctuation stripped from a token, EXCLUDING the signs ``+``/``-`` and
+#: the decimal point ``.`` — a leading sign or decimal point is part of a number's
+#: value, not wrapping noise. A genuinely-trailing ``.`` (sentence period) is handled
+#: by the dedicated rstrip retry in :func:`_normalize_token`, which can tell it apart
+#: from a value-bearing leading point; a blanket edge-strip cannot.
+_STRIP_EDGE = "".join(c for c in string.punctuation if c not in "+-.")
 
 
 def _normalize_token(raw: str) -> str:
@@ -147,20 +150,27 @@ def _normalize_token(raw: str) -> str:
     dropped the ``-`` and left commas to break ``float()``) did not deliver.
 
     A token that is ALREADY a number is recognised before any punctuation is
-    stripped: the edge-strip set includes ``.``, so a leading-decimal token like
-    ``".5"`` would otherwise lose its point and normalize to ``"5.0"`` — equal to
-    a gold ``"5"`` (false positive) and unequal to the value-identical gold
-    ``"0.5"`` (false negative). The official DROP ``_remove_punc`` tests
-    ``_is_number`` first and leaves numbers untouched for exactly this reason."""
+    stripped: a leading-decimal token like ``".5"`` must not lose its point and
+    normalize to ``"5.0"`` — equal to a gold ``"5"`` (false positive) and unequal
+    to the value-identical gold ``"0.5"`` (false negative). The official DROP
+    ``_remove_punc`` tests ``_is_number`` first and leaves numbers untouched for
+    exactly this reason (issue #423). The float-first path alone only covers the
+    *bare* token: with ``.`` in the edge-strip set, wrapped forms like ``"$.5"``
+    and ``".5."`` still lost the leading point on the second-chance path. So the
+    edge strip excludes ``.`` entirely, and a genuinely-trailing period (``".5."``,
+    ``"16.."``) is retried with an explicit ``rstrip(".")`` — right-side dots are
+    sentence punctuation, left-side dots are value."""
     try:
         return str(float(raw.replace(",", "")))
     except ValueError:
         pass
     core = raw.strip(_STRIP_EDGE)
-    try:
-        return str(float(core.replace(",", "")))
-    except ValueError:
-        return _PUNCT.sub("", raw)
+    for cand in (core, core.rstrip(".")):
+        try:
+            return str(float(cand.replace(",", "")))
+        except ValueError:
+            continue
+    return _PUNCT.sub("", raw)
 
 
 def _split_internal_hyphens(token: str) -> list[str]:
diff --git a/src/trinity/orchestration/reward.py b/src/trinity/orchestration/reward.py
index 7192ef8..6395e08 100644
--- a/src/trinity/orchestration/reward.py
+++ b/src/trinity/orchestration/reward.py
@@ -505,6 +505,14 @@ def extract_last_number(text: str) -> str | None:
     candidates: list[tuple[int, str]] = []
     for _start, end, term in _iter_latex_frac_sqrt_spans(text):
         candidates.append((end, term))
+    # Binomial coefficients: whole ``\binom{n}{k}`` / ``{n \choose k}`` so the
+    # lower index digit is not taken alone (issue #489).
+    for m in re.finditer(
+        r"\\binom(?![a-zA-Z])\s*\{[^{}]*\}\s*\{[^{}]*\}"
+        r"|\{[^{}]*\\choose\s*[^{}]*\}",
+        text,
+    ):
+        candidates.append((m.end(), m.group(0)))
     # Match a simple fraction a/b FIRST (so "1/2" is kept whole, not read as "2"),
     # then decimals/integers like -1,234.56 or 42 or .5 ; require a digit somewhere.
     brace = r"\{(?:[^{}]|\{[^{}]*\})*\}"
@@ -521,7 +529,9 @@ def extract_last_number(text: str) -> str | None:
         candidates.append((m.end(), m.group(0).replace(",", "").replace(" ", "")))
     if not candidates:
         return None
-    candidates.sort(key=lambda item: item[0])
+    # Latest end wins; on a tie prefer the longer span so ``\binom{5}{2}`` beats
+    # the trailing index digit that ends at the same index (issue #489).
+    candidates.sort(key=lambda item: (item[0], len(item[1])))
     return candidates[-1][1]
 
 
@@ -820,6 +830,16 @@ def normalize_math_answer(ans: str | None) -> str:
     # The [dt]? family — \frac, \dfrac and \tfrac — all render the same value.
     s = _unwrap_latex_fractions(s)
     s = re.sub(r"\\[dt]?frac\s*(\d)\s*(\d)", r"\1/\2", s)
+    # Binomial coefficients → sympy ``binomial(n,k)`` so ``\binom{5}{2}`` equals
+    # gold ``10`` (issue #489). Also TeX ``{n \choose k}`` / bare ``n \choose k``
+    # (outer-brace peel may remove the wrapping ``{...}`` before we get here).
+    s = re.sub(
+        r"\\binom(?![a-zA-Z])\s*\{([^{}]*)\}\s*\{([^{}]*)\}",
+        r"binomial(\1,\2)",
+        s,
+    )
+    s = re.sub(r"\{([^{}]*)\\choose\s*([^{}]*)\}", r"binomial(\1,\2)", s)
+    s = re.sub(r"(-?\d+)\s*\\choose\s*(-?\d+)", r"binomial(\1,\2)", s)
     # \sqrt{x} -> sqrt(x), and the unbraced single-token \sqrt2 -> sqrt(2). Like the
     # \frac handling just above, the braced and bare forms render the SAME value, so
     # they must normalize identically (else \sqrt{2} is a false negative against a
diff --git a/tests/test_reward_binom.py b/tests/test_reward_binom.py
new file mode 100644
index 0000000..4dd581f
--- /dev/null
+++ b/tests/test_reward_binom.py
@@ -0,0 +1,23 @@
+"""Binomial ``\\binom`` / ``\\choose`` must not grade as the lower index (issue #489)."""
+
+from trinity.orchestration.reward import extract_last_number, normalize_math_answer, score_text
+
+
+def test_extract_binom_as_whole_term():
+    assert extract_last_number(r"\binom{5}{2}") == r"\binom{5}{2}"
+    assert extract_last_number(r"{5 \choose 2}") == r"{5 \choose 2}"
+
+
+def test_normalize_binom_to_sympy():
+    assert "binomial(5,2)" in normalize_math_answer(r"\binom{5}{2}").replace(" ", "")
+    assert "\\binom" not in normalize_math_answer(r"\binom{5}{2}")
+
+
+def test_unboxed_binom_grades_value_not_index():
+    assert score_text("math500", r"answer is $\binom{5}{2}$", "10") == 1.0
+    assert score_text("math500", r"answer is $\binom{5}{2}$", "2") == 0.0
+
+
+def test_boxed_binom_and_choose():
+    assert score_text("math500", r"\boxed{\binom{5}{2}}", "10") == 1.0
+    assert score_text("math500", r"\boxed{{5 \choose 2}}", "10") == 1.0