diff --git a/src/mlpack/methods/decision_tree/best_binary_numeric_split.hpp b/src/mlpack/methods/decision_tree/best_binary_numeric_split.hpp
index 657df344d9..3f5cd27e65 100644
--- a/src/mlpack/methods/decision_tree/best_binary_numeric_split.hpp
+++ b/src/mlpack/methods/decision_tree/best_binary_numeric_split.hpp
@@ -29,6 +29,10 @@ HAS_MEM_FUNC(BinaryScanInitialize, HasBinaryScanInitialize);
 // we can use with SFINAE to catch when a type has a BinaryStep(...) function.
 HAS_MEM_FUNC(BinaryStep, HasBinaryStep);
 
+// This gives us a HasBinaryGains<T, U> type (where U is a function pointer)
+// we can use with SFINAE to catch when a type has a BinaryGains(...) function.
+HAS_MEM_FUNC(BinaryGains, HasBinaryGains);
+
 /**
  * The BestBinaryNumericSplit is a splitting function for decision trees that
  * will exhaustively search a numeric dimension for the best binary split.
@@ -134,7 +138,9 @@ class BestBinaryNumericSplit
       HasBinaryScanInitialize<FitnessFunction, void(FitnessFunction::*)
           (const ResponsesType&, const WeightVecType&, const size_t)>::value &&
       HasBinaryStep<FitnessFunction, void(FitnessFunction::*)
-          (const ResponsesType&, const WeightVecType&, const size_t)>::value,
+          (const ResponsesType&, const WeightVecType&, const size_t)>::value &&
+      HasBinaryGains<FitnessFunction,
+          std::tuple<double, double>(FitnessFunction::*)()>::value,
       double>::type
   SplitIfBetter(
       const double bestGain,
diff --git a/src/mlpack/methods/decision_tree/best_binary_numeric_split_impl.hpp b/src/mlpack/methods/decision_tree/best_binary_numeric_split_impl.hpp
index 1913fc0868..3d0ebdc078 100644
--- a/src/mlpack/methods/decision_tree/best_binary_numeric_split_impl.hpp
+++ b/src/mlpack/methods/decision_tree/best_binary_numeric_split_impl.hpp
@@ -327,7 +327,9 @@ typename std::enable_if<
     HasBinaryScanInitialize<FitnessFunction, void(FitnessFunction::*)
         (const ResponsesType&, const WeightVecType&, const size_t)>::value &&
     HasBinaryStep<FitnessFunction, void(FitnessFunction::*)
-        (const ResponsesType&, const WeightVecType&, const size_t)>::value,
+        (const ResponsesType&, const WeightVecType&, const size_t)>::value &&
+    HasBinaryGains<FitnessFunction,
+        std::tuple<double, double>(FitnessFunction::*)()>::value,
     double>::type
 BestBinaryNumericSplit<FitnessFunction>::SplitIfBetter(
     const double bestGain,