How to estimate the "New Credit" category

[jello77]

FICO says the "New Credit" category accounts for 10% of your score. It includes newly opened accounts and all "hard" credit inquiries during the past 12 months (except that multiple inquiries from auto or mortgage lenders within a short period of time are treated as a single inquiry).

On several other web sites, I found a formula estimating the effect of this category on your score. I'm not sure how valid it is, but I'm posting it for your consideration.

Here is the "New Credit" estimate formula:

=============================================================

FICO scores range from 300 to 850. This includes a base of 300 plus 550 points that are variable.

"New Credit" accounts for 10% of your score, which is equivalent to 55 variable points (10% of 550).

If you have no new accounts or hard inquiries in the past 12 months, you get all 55 points.

For each new account or hard inquiry in the past 12 months, there is a time-based deduction as follows:

initial deduction for each new account or hard inquiry:  10 points

after 3 months reduced to:  5 points

after 6 months reduced to:  2.5 points

after 12 months reduced to:  0 points

(Maximum total deduction will not exceed 55 points)

=============================================================

This estimate formula seems credible in that it incorporates two factors included in FICO's official Credit Score Estimator as shown below. 

http://www.myfico.com/ficocreditscoreestimator/

-------------------------------------------------------------------------------------------------------

3. How many loans or credit cards have you applied for in the last year?

0

1

2

3 to 5

6 or more

-------------------------------------------------------------------------------------------------------

4. How recently have you opened a new loan or credit card?

less than 3 months ago

between 3 and 6 months ago

more than 6 months ago

-------------------------------------------------------------------------------------------------------

On my new TU and EQ reports, I have two hard inquiries 2 months old and one new account 10 months old. The estimate formula implies that I have deductions of 10  + 10  +  2.5  =  22.5. Compared to my scores a year ago when I had no new accounts or inquiries, my TU score is down 24 points and EQ is down 21 points. So for me, this formula seems to be accurate. 

UPDATE:

After posting this message, I found this article How to improve your FICO score which seems to endorse the concept behind this formula.

[UtahTiger]

Thanks for his info....where are the other sites where you are finding this good stuff!!!!

[RobertEG]

I can agree with the general thinking, and think it is good guide for including the potential affects of both inquiries and new accounts, but I dont buy the rigid point assumptions made.

First of all, the FICO algorithm for scoring new credit history is a proprietary algorithm that is not published, so everything in this "formulat" is, by definition, purely speculative, and based on only anecdotal experience. My experience does not lead to such rigid assumtions.

First, as to the 12-month assumption, yes, FICO has stated that it does not score inquiries after 12-months, but FICO has never said that an account has no scoring impact in this category once it turns a magic 12-month time period.

Second, the formua assumes a straight linear decline in the impact of new inquiries over the 12-months that they are scored.  FICO has never stated this.

Third, the formula assumes an initial impact of each new inq and new account at 10 pts.  This is, from my experience, clearly speculative, and a gross oversmplification.

I think there is much evidence to show that the effect of new inquiries varies based on what credit scoring bucket, or algorithm, you are placed in.  For example, new inquiries seem to have more impact on those in higher scoring buckets.

Fourth, my experience is that second, third, fourth, etc., inquiries are not scoring the same as a first inquiry. 

Under this formula, if you made three inquiries today that led to three new accounts, you would lose all 56 points under this scoring category.

I think the formula is interesting, but I would little faith in the specific numerical assumptions made. 

[jello77]

RobertEG,

Thanks for your feedback. All your points make good sense.

[Anonymous]

---

@RobertEG wrote:

I can agree with the general thinking, and think it is good guide for including the potential affects of both inquiries and new accounts, but I dont buy the rigid point assumptions made.

First of all, the FICO algorithm for scoring new credit history is a proprietary algorithm that is not published, so everything in this "formulat" is, by definition, purely speculative, and based on only anecdotal experience. My experience does not lead to such rigid assumtions.

First, as to the 12-month assumption, yes, FICO has stated that it does not score inquiries after 12-months, but FICO has never said that an account has no scoring impact in this category once it turns a magic 12-month time period.

Second, the formua assumes a straight linear decline in the impact of new inquiries over the 12-months that they are scored.  FICO has never stated this.

Third, the formula assumes an initial impact of each new inq and new account at 10 pts.  This is, from my experience, clearly speculative, and a gross oversmplification.

I think there is much evidence to show that the effect of new inquiries varies based on what credit scoring bucket, or algorithm, you are placed in.  For example, new inquiries seem to have more impact on those in higher scoring buckets.

Fourth, my experience is that second, third, fourth, etc., inquiries are not scoring the same as a first inquiry. 

Under this formula, if you made three inquiries today that led to three new accounts, you would lose all 56 points under this scoring category.

I think the formula is interesting, but I would little faith in the specific numerical assumptions made. 

---

Actually, the scoring decline is non-linear

[haulingthescoreup]

I was told that an inq hurts exactly as much through the 12 months, and then it doesn't hurt at all.

What does fade with time is the newness of the new account. That's why you'll see a gradual fade.

Inqs often don't hurt at all. For instance, the first inq on a given CRA doesn't hurt, and the second one does. Then the third one might not, but the fourth does. And after a certain number (a half dozen? I don't know for sure), they don't count against you for scoring, although lenders will certainly give you the fish eye.

[RobertEG]

The hypothetical formula you cite states that:

________________________________________________________

FICO scores range from 300 to 850. This includes a base of 300 plus 550 points that are variable.  (huh?)

"New Credit" accounts for 10% of your score, which is equivalent to 55 variable points (10% of 550). (it is 10% of score, not “variable points”…)

_______________________________________________________

I don’t follow this logic, at all. 

The suggested “formula” for scoring in the new credit category starts with a basic assumption, which I don’t agree with.  It assumes  that FICO scoring is built upon an initial assignment to you of a “base of 300 points,” and then assigns to your score additional, variable points, potentially as much as 550, based on your scoring in each of the five major FICO scoring categories.  It then assumes that the 550 “variable points” are then weighted, for all consumers, using the generic category weighting factors for each of the five major scoring categories, and then added to your accumulated base of 300.

I don’t think that is how FICO scoring works.  300 may be the lowest score they post, but that is not a free 300-pts, plus bonuses, scoring program.  I have never, even seen the posting by anyone who develops credit scoring of a “base score plus additional points” assessment of credit scoring.  I don’t think it exists.

Rather, this is my view of credit scoring.  You don’t start with points, you start with a credit file. 
Before they (FICO) do a lick of scoring, they must first determine which of their dozen or so different scoring algorithms they are going to apply to your credit file.  These are what most on here call “scoring buckets.”  Which algorithm you are scored under depends on such major credit file considerations as what is called a “clean” or “dirty” credit file (with or without major derogs), “thin” or “thick” file (presence of multiple TLS), or a “lean,” short credit history.

Each algorithm assigns different weightings to each category, and information reported under each category, that does not adhere to a uniform 35% payment history, 30% util of current credit, 15% length of credit, 10% new credit, and 10% credit mix, algorithm. Those are just averages.

Then the specific algorithm you are assigned to reviews your CR, and scores each category, from the bottom up.  It then adds them up, based on their own weighting of the risk of each category within your own bucket, and arrives at a final FICO score.  If the final score is below 300, they will report the minimum of 300. 

So, next month you show a improvement in one category.  But FICO runs their algorithm, and while showing improvement in that one category, still shows a resulting score of under 300.
You would, in my opinion, still get a FICO score of only 300.  It is not an “assumed base points of 300 pts last month, plus further improvement above that base” scoring process for improvement in one category from month to month.  I don’t think that is the way it works.

But just my opinion.

[jello77]

RobertEG,

Thanks again for your feedback. 

Because FICO scores have such a major impact on our financial lives, there is understandably a strong temptation to try to de-construct the complex scoring formula into something more comprehensible.  But it is no doubt a futile exercise.