Matrix question relative to foreclosures/bankruptcies

[Anonymous]

There are now and will continue to be a myriad of delinquencies, defaults, pre-sales, deficiencies, bankruptcies, and foreclosures for many. In concert with this, I have assumed that the matrix scoring was not just subject to our own utilization patterns but subject to the masses. So if say 3 out of 10 consumers are more delinquent on bills (credit cards to mortgages), will that not have a corresponding effect on our scoring relative to others. E.g. not having as much of an impact given increased commonality if I follow suit or conversely if I am without similar challenges my score will be falsely improved without a change in utilization? 

[Anonymous]

The scoring model has not changed-

So if you keep doing the right things- you score will continue to rise.

You score will not be inflated - just because there are more people with lower scores.

People with serious credit issues- (things you mentioned) score will take a big hit- 

The thing is- lenders are tightening standards, increasing score requirements, thus- less qualified people, less lending.

The "Curve" of scores has moved- but the score requirements have gone up- so these curves are moving in different directions.

[RobertEG]

Interesting supposition, Timothy.  I am not questioning it, however, FairIsaac states on its recent webinar (2/22/08, available on the Credit Education tab above) that it constantly tweaks its algorithms as patterns change.  He was not talking about Fico08, but ratther regular tweaks in the software sent to the CRAs.  With both mortgage and CC use trends shifting so dramatically in recent times, are they not changing their risk anaylsis, and thus scoring model to keep abreast of current risk analysis?  So I am not sure that the scoring algorithm does not shift based upon udated credit risk data that pours into FairIsaac constantly.  It may not be as static as we presume?  I dunno, but I also dunno what I dunno.

Message Edited by RobertEG on 04-17-2008 05:17 PM

Message Edited by RobertEG on 04-17-2008 06:30 PM

[cobra19]

Good post, JGZ. I agree with you.  I think it is obvious the scoring algorithms are fluid (and have changed) over time depending on peoples spending habits, amount of debt, and probably many of factors.  I think of it as going back to my college course in thermodynamics. Our first quizz was a toughie, and the highest score was a 68 (which was an A). But a majority of us got scores between 48-56 - which was a C. Our final quizz, some 5 months later, a 99 was an A, and to score a C you had to be between 71-77.  

With FICO you are being compared with others in your "class". Thing is, the "class" keeps changing.

[RobertEG]

Cobra, the only thing I remember from my thermodynamics classes back in 19xx is that heat flows from those bodies who got temperature to those who dont, and exponentially based upon theiir degree of bodily separation. 

I do see a paralled with FICO.!  Heat flows with separation of the bodies... its a scientific fact!

LOL! 

[Anonymous]

Thanks Timothy,

Can you point me to more information on the curves dialog and the lack of shifting in the modeling?  I tend to share this type of information and my colleagues tend to request verifiable sourcing, particularly in interpretational education.  Fortunately, everything else you offered jibes with my current knowledgebase.  Much appreciated.