@RobertEG wrote:
Fair Isaac has stated that the difference in scoring, with the same data, should not vary by more than approx 20-30 pts between the scoring models used by the big 3 CRAs. So that is a rough guideline. If the variance is more than that, there is most likely some difference buried in the data cranked into their models.
One situation that could be atypical is having a small positive balance on most or all cards, which may bring about a particularly low EQ score while TU swings less. In exercising a monthly cycle of such balances at this time with the EQ already 20 points lower than the TU, I have noticed that the scores may be impacted as well by some cards considered previously dormant by FICO as Amex does not report non-positive balances. It's not a pretty picture.
