No credit card required
Browse credit cards from a variety of issuers to see if there's a better card for you.
In a very simple way to help understand it, FICO considers people with similiar situtions together (bucket them) For instance if you have a BK then you may be comapred against others that have had a BK and can score higher if you are being more responsible with your new payments as compared to the group.
Then FICO would move you as soon as the BK is no longer on your CR - they would re-bucket you with others that don't have a current BK - and many acutally see a point drop here. While you were the top of the pack with other BKs you may be an average consumer with those that don't have a BK.
That being said, no one that I'm aware of can tell you what group you are being compared with and when you're moved, when score differences are based on buketting or re buketting, etc.
We have some fairly strong evidence of some buckets....the precise definitions of each bucket is not publicly known though.
In general:
Major Lates, Charge Offs, Collections, Bankruptcies, and Public Judgement form the "bad buckets" ie you are compared to other people with bad credit behavior in the files.
The other buckets seem to be largely driven by age, though there is likely a bucket for people with thin credit histories (ie few accounts).
0-2
2-5
5-12ish
12-19ish
19+ish
(All of the above in years)
Are some of the buckets we surmise are out there (though there may be an average age of accounts factor to them as well.
"New Accounts" can also play a role in scoring buckets.
@Anonymous wrote:We have some fairly strong evidence of some buckets....the precise definitions of each bucket is not publicly known though.
In general:
Major Lates, Charge Offs, Collections, Bankruptcies, and Public Judgement form the "bad buckets" ie you are compared to other people with bad credit behavior in the files.
The other buckets seem to be largely driven by age, though there is likely a bucket for people with thin credit histories (ie few accounts).
0-2
2-5
5-12ish
12-19ish
19+ish
(All of the above in years)
Are some of the buckets we surmise are out there (though there may be an average age of accounts factor to them as well.
Fair Isaac's official term for what we call "buckets" is "scorecards"
http://www.fairisaac.com/ficx/Products/scoring/FICO-Scores.html
@Anonymous wrote:
I keep hearing about being in a certain bucket.....what the heck does that mean exactly ??
I can't do as eloquent job of others but to put it simply and I mean gravely simplified, you are compared to your cohorts, broadly defined as people who match your specific credit makeup and when something that put you in that bucket is no longer on your report, you are switched to a bucket that is similar to your configuration but does not include that item. This can make you go up or down depending on what it is and a ton of other factors.
When you take the MCAT, you are compare to all the ones who took the test to go med school, not to the ones who took the LSAT who are compared to their own, this would be a simple, very simple, idea behind buckets.
Here is a 40-page document all about "bucketing", or scorecards, as Fair Isaac calls it:
It's very mathematical and highly theoretical (no actual examples of the buckets that are used), but it's interesting to scan through it.
Mike14 wrote:Here is a 40-page document all about "bucketing", or scorecards, as Fair Isaac calls it:
It's very mathematical and highly theoretical (no actual examples of the buckets that are used), but it's interesting to scan through it.
Message Edited by Mike14 on 01-25-2009 10:17 PM
I have read this a few times. It's quite easy to comprehend.
@Anonymous wrote:Here is a 40-page document all about "bucketing", or scorecards, as Fair Isaac calls it:
It's very mathematical and highly theoretical (no actual examples of the buckets that are used), but it's interesting to scan through it.
Message Edited by Mike14 on 01-25-2009 10:17 PM
Thanks, it looks fascinating. I've never worked in the financial industry, but am actually somewhat familiar with the statistical techniques used because I do a lot of data mining and statistical analysis in my own work, pharmaceutical research.
FICO wants to predict how likely somebody is to default on a loan based on credit history data. We want to predict how likely a patient is to have various outcomes based on medical data. FICO wants to know which items in a credit profile best predict getting into credit trouble. We want to know which medical factors best predict how effective our drugs will be in treating a patient.
So different data, different domains, but similar sorts of questions and very similar statistical tools. In fact there are many scoring systems used in medical practice. For example, google framingham risk score to learn about an algorithm for estimating the odds of somebody getting a heart attack. This score is used by doctors to identify people at high risk so they can do things like prescribe drugs that lower blood pressure and bad cholesterol.
@MattH wrote:For example, google framingham risk score to learn about an algorithm for estimating the odds of somebody getting a heart attack. This score is used by doctors to identify people at high risk so they can do things like prescribe drugs that lower blood pressure and bad cholesterol.I really appreciate learning that particular piece of information. The framingham risk score is definitely more important to me than my FICO score.
It is amazing how a person can post something "on topic" about consumer credit health and also share info that can really impact a person's physical well being.
I have always ignored advice about what I should or shouldn't do to avoid a heart attack. I didn't know that there was a "FICO (type) score" for avoiding heart attacks. Now I am ready to pay attention.
![]()
Many thanks.