Information on the Internet says Apophis will come within 20,000 miles of earth.  That's less than one-tenth of the distance to the Moon, which is 238,000 miles.  Plus, some satellites of the GEO type are 36,000 km (or 22,500 miles) above the earth, and some HEO satellites are higher than this.  This is definitely too close for comfort!  According to Tom Horn's books, a more accurate approximation for pi predicts the asteroid will crash.  Were two different approximations used for pi, and the one less accurate used for the path of the asteroid that was made public?  There is an infinite series representation for pi that converges slowly; nevertheless, I have heard  pi has been calculated to accuracy within 2,000 decimal places.  (I'm retired from teaching college mathematics.)