More than a half century ago, Bill Fair and Earl Isaac- the statisticians behind the original credit scoring formulation- reviewed a set of over 1 million consumers who opened loans at the same time. The credit profiles of these consumers, particularly those who defaulted on the loans, were examined to identify common variables they exhibited at the time they applied for the loan.
They then built statistical models that assigned weights to each variable. These variables were combined to create a credit score.
Today, loan model builders strive to identify the best set of variables from your credit history that most effectively predicts future credit behavior.
In determining credit scores, you’re placed in a risk category that compares you to a large number of consumers with similar credit histories. This allows lenders to compare "apples to apples," ensuring that your credit behavior is judged in a context that’s relevant and fair.
For example, consumers with brief credit histories and only a few accounts aren’t compared to consumers with long-established credit histories. Rather, these consumers are compared to other consumers who also have brief credit histories.
These categories are called “scorecards”. The premise for segmenting consumers this way is to optimize the model’s performance for all different consumer credit file types. If the credit scoring model only had one scorecard, then some groups of consumers would benefit and other groups would suffer. The better the developer is at defining a unique population, the more accurately predictive the credit score becomes. Currently the FICO scoring system has 10 scorecards (12 for FICO 08).
The following three components all reside within the scorecards:
Characteristics: A characteristic is simply a question the model asks your credit report. Some of the questions might include:
- How many inquiries have you had in the last 6 months?
- What is your revolving utilization?
- How old is your oldest account?
Each scorecard has a different set of characteristics, but many of the same characteristics reside across multiple scorecards.
Variables: If the characteristic is best described as a “question” then the variable is best described as “the answer.” i.e. “I have had 2 inquiries in the past 6 months”, etc.
Weights: This is where the final score starts coming together. The weight is the point value given to each variable, and you can either gain or lose points depending on the variable. So point values are different for different scorecards.
Just because a person is on one scorecard at any given time doesn’t mean he or she will stay on that one forever. In fact, certain changes to your credit report can cause you to ‘move’ scorecards. This is called a scorecard hop. A scorecard hop can either help or hurt your score. Remember- it’s all about how you compare to other people on your scorecard, and how you manage your credit relatively.
Sometimes the reasons for a scorecard hop can be subtle, but a late payment, collection, or bankruptcy can drastically increase your chance of switching scorecards.
Sure, scorecards can be confusing, but building solid credit doesn’t have to be. If you focus on paying your bills on time and keeping your credit profile free of delinquencies, you’ll be pleasantly surprised.