do you think this is a common danger for profiling if so how would you control this risk

One criticism of psychological profiling is the tendency for police to believe the expert, even if the profile is mistaken on key variables, which may divert them from meaningful leads. This was exemplified in the search for the D.C. snipers in 2002. Do you think this is a common danger for profiling? If so, how would you control this risk?