Would systems that use trace trees (e.g., TraceMonkey) have the same sort of issues? I have a feeling that they would notice repeated calls to the same if-stat

Would systems that use trace trees (e.g., TraceMonkey) have the same sort of issues? I have a feeling that they would notice repeated calls to the same if-statement and optimize accordingly. They may still have the same sorts of issues, but would the used benchmark (looping and querying the if-statement n times) have completely different results in a JIT system?