The source code for two applications might match 89% and yield completely different results
To be clear, I'm saying that genetic match supersedes any list of superficial attributes to determine identity. If it's established that X and Y are an 89% DNA match, even though X tastes and looks like Z, the fact that Y and Z belong in distant genetic branches should be enough to dismiss X as a possible Z.
89% doesn’t really seem like a case closed kind of stat.