Can Someone Explain This Scoring @$%&@

[marty56]

---

@DaveSignal wrote:

FICO doesn't always correctly interpret the input data.  Slight changes that result from disputes (such as date reported on a charge-off) that would have no effect on the decision of a human reading the report can cause drastic fluctuations in a FICO score.

---

I understand that you don't like the way FICO scores certain types of data.  I didn't like the fact that late payments stay on my report of 7 years or that my 120 day late prevented me from being in the 800 club for 7 years.  I do like the fact that FICO would do the same to you or anyone else with the same data on their report.

Do you think a human would look at a 7 year old CO and say it is only going to be their a few more months so I will forget it?  Do you think the average CSR would know that? Risk their job over it?

Even if you do find a CSR who would do that, I could find one that won't. So your human formula doesn't always work they way you want it to and it is impossible to predict the correct result with the same input data.

That is the definition of formula producing inconsistent results.

[DaveSignal]

---

@marty56 wrote:

---

@DaveSignal wrote:

FICO doesn't always correctly interpret the input data.  Slight changes that result from disputes (such as date reported on a charge-off) that would have no effect on the decision of a human reading the report can cause drastic fluctuations in a FICO score.

---

I understand that you don't like the way FICO scores certain types of data.  I didn't like the fact that late payments stay on my report of 7 years or that my 120 day late prevented me from being in the 800 club for 7 years.  I do like the fact that FICO would do the same to you or anyone else with the same data on their report.

Do you think a human would look at a 7 year old CO and say it is only going to be their a few more months so I will forget it?  Do you think the average CSR would know that? Risk their job over it?

Even if you do find a CSR who would do that, I could find one that won't. So your human formula doesn't always work they way you want it to and it is impossible to predict the correct result with the same input data.

That is the definition of formula producing inconsistent results.

---

No, It is not that I don't like the way FICO scores the data.  It is that FICO has errors and is inconsistent.  If there was a charge off on 5/2008 reported on 6/2008, it would be scored a certain way.  If there was the EXACT SAME DATA for a charge off on 5/2008 reported today (updated as verified from a dispute), FICO would score it differently than it did the first time.  

What I meant is that a human looking at the data would see the EXACT SAME DATA whether it was reported on 6/2008 or yesterday, it is still a charge off on 5/2008 and SHOULD BE SCORED THE SAME.

FICO is not a human and interpets these two instances of the same data differently, making it inconsistent and unpredictable.

[marty56]

While it may be reasonable to assume that your score should go back to where it was after a dispute if the item was verfied, it would be hard to say why it didn't.  It is quite possible that the FICO scoring program actually gives some weight to when the CO was reported as opposed to when it happend.  It may simply be that FICO doesn't remember disputed items.

[DaveSignal]

I will start a new thread to discuss this.

[lazerz69]

I would like to thank everyone who has responded to this post.  Based on what I have read so far in this thread, it appears that FICO scoring is really a complete mystery.

I still can not determine how or why my score went down 23 points after a dispute was removed when it never went up when the dispute was first placed on my CR.

However, I do understand what everyone has stated here, that there may be more to the scoring process than meets the eye . . .so to speak.

[llecs]

OP, this was on EQ right? Did you pull a new EQ report (from anywhere) to look for any changes?

[lazerz69]

---

@llecs wrote:

OP, this was on EQ right? Did you pull a new EQ report (from anywhere) to look for any changes?

---

llecs, thanks for taking the time to respond to my question.

Yes, this was on EQ.  I did pull a new EQ report, at least I think it was new.  I had a "confirmation number" from the CA that was disputed.

I logged on to EQ and was provided a CR that had the current date.

In the 81 month payment history, the CA added a "CO" code, (code is probably the wrong term), in December 2010.  That was the only change for that CA.

However, I am now really confused, because the updated CR shows that a previously reported CO, with a CO amount over 1500.00 has been deleted.  I would have presumed that with this deletion I would have had a boost in my score.  I guess my AAoA could have taken a hit. 

[llecs]

How old is that deleted account in relation to your AAoA? What kind of CO was it? CC? Loan? Utility? I've lost points by having charge-offs deleted. Each time it was either my oldest accounts (or oldest within CCs or oldest within installments) or older than my AAoA whereby AAoA dropped.

[lazerz69]

---

@llecs wrote:

How old is that deleted account in relation to your AAoA? What kind of CO was it? CC? Loan? Utility? I've lost points by having charge-offs deleted. Each time it was either my oldest accounts (or oldest within CCs or oldest within installments) or older than my AAoA whereby AAoA dropped.

---

The collection account that was deleted was, (which was not the CC that I disputed), about 10 years old.  This is about my AAoA.  Most of my old tradelines are about 10 years old.  My oldest tradeline, (which is a negative account), is twenty years old.

However, reviewing my Score Watch history, it appears that I may have received a small bump in my score when this collection account was removed.

llecs - Thanks for the continued responses and help.  It is greatly appreciated.

[stan_the_man]

---

@DaveSignal wrote:

---

@marty56 wrote:

As a formula, it just crunches data so I would look at the input data.  Since it's based on probability and statistics, it's output may not predict the correct result but it is consistent.

There never could be a perfect scoring formula.  Someone will always be given too much credit so to speak or not enough.

---

FICO doesn't always correctly interpret the input data.  Slight changes that result from disputes (such as date reported on a charge-off) that would have no effect on the decision of a human reading the report can cause drastic fluctuations in a FICO score.

---

Don't kid yourself. The date of a CO is always going to affect how a human reads your report.

Surely you would be more forgiving of someone who had a CO 6 years ago compared to someone who have one 6 months ago -- so would I, and so does the Fico formulas.