1) I LOVE this stuff...where thinking human beings pitch in and throw 2 or 3 cents into a pile and kick around ideas...great stuff ppl.
2) And I agree with....
"
Yes, what you're saying does make sense - to a degree. BUT - the FICO algorithms are not based on what is "logical" from a human perspective - that is, they are not based on any sort of hypotheses or theories of human behavior, they are mathematical algoritms designed to predict future risk based on analysis of long term datasets of past performance of millions od people. Sort of like trying to predict the future performance of a given stock, based purely on the past perfomance of all stocks - see where I'm going? The entire system, IMO, is based on the flawed premise of past performance being an adequate indicator of future performance - in fact it is but one peice of the picture
The best we can hope for is to figure out what the "best performers" have in common and emulate that."
All good but the 'flawed premise' thing is IMO a bit sour grapes because let's be HONEST....that is our BEST tool in protecting against 'risk' and 'guessing' towards the future. We as humans DO NOT have a better way of 'trying' the best we can towards making decisions regarding our futures.
Eq: If your child (daughter for example) were to date/marry a guy...you darn well would use a FLAWED PREMISE of what he may or may not have done in
the past...be it hit another woman, leave his last mate with 2 hungry kids or a history of not working....of course you'd use a 'flawed premise' of GUESSING he may NOT be 'the one' for "your baby girl". (Lenders might feel somewhat protective of their money and whom they extend credit to)
3)
"Second, while I understand this is the nature of how credit works, it doesn't make all that much sense to me logically. Let’s say I'm a manager hiring for an important position and I'm interviewing two candidates. The first has nine years’ experience at one single company in the exact same position. The second has three years’ experience at three different companies with their related skill set spread around with good references. I'm going with candidate number two, since he's proven he can handle multi-faceted aspects of the job with different companies and still be dependable. Does that make sense?"
*This is a great analogy and I THANK YOU for asking it in this way...it helps show where you're off track.
This bleeds right back into what the above poster correctly states that credit scoring (which is RISK scoring) isn't the same as ALL human nature...it's an algorithm,
a computer program (an ap, for you youngsters) it simply crunches data and spits out a numericial factor based upon said data...and it does so freshly every time, meaning that yesterday means ZERO to the algorithm, it runs the data brand new every time, not remembering or CARING about yesterday or last week or month or 2 days ago.
The funny thing about your example is when it comes to SCORING ...it's actually kind of a wasted comparision because there is NO WAY based upon just that info to KNOW 'anything'...(which is great because it helps to say to you BLOW UP this way of thinking, all together....otherwise you'll keep the 'work harder' at the solving the wrong problem which kills so many ppl) it's so much better to say NOPE all wrong, wipe it clean and see it another way.
a) 1st off we could debate your premise that 9 years at Google or the NFL is a 'worse' qualifield candidate vs one that hopped or possibly couldn't last past 36 months at any of 3 places and is looking for a 4th within about 100 months...but we'll leave that be and roll WITH your position.
b) It can't be moved to risk scoring because there are too many other factors in play (and this aspect of several moving targets isn't the easy button that our modern culture desires)
On one hand the model does like several accounts being managed over just one BUT said manage of an account for nearly 10 years just might = or have proven a better RISK than how 3 younger accounts may have SO FAR......
Again the 'process' of getting those 3 accounts and their age may or not have garnered a SCORE higher or lower than a profile where the said RISK may according the the model be actually LESS due to a well managed account opened 9 years with ZERO attempts to take on more debt (that is what a HP is...which is required to build the other profile with 3 accounts ...how many HP's did it take to get there again the other guy hasn't even REQUESTED debt let alone mis-managed what he already has)
Way more RISK scoring than just reading a resume...candidates can explain pros and cons on their resume and speak about degrees etc ...credit scoring doesn't give a jack crap about your WHY....it's the numbers stupid...just the numbers.
Which actually makes things EASY once one commits to understanding how the formula works..then any schmuck can duplicate H2O = water for any moron that follows the formula...which is GREAT, no special talent needed, no magic beans, no special carpet, fat, skinny, tall, shortANY schmuck will get like results if they follow the formula it's beautiful...money can't BUY it, poor college students or elderly old ladies both have a shot at a top shelf risk score.....
