I discovered that the queries in question were victims because of SQL jobs running causing them to die. Just seems odd that the duration would go so long. I can't make the mental connection between the two.