We developed the What Are My Chances engine to predict a students odds of getting into college. There are two ways of doing this. The first is to build a statistical model of every college in the us and then run the model against the student profile. The second is our way.
Based on what we learned from colleges across the US we had a pretty good understanding of what would help or hurt a student in the admissions process. The engine weighted all those factors and produced consistently accurate results.
We tested the model by running actual admit/deny data against it. When it said that the chances were good they would be admitted we were usually right. Also, when we said chances were low we were right too.
This was the orginial press release from February 2008. The tool had been up a while, but we waited to formally annouce it. Students loved it.