Maddening score inconsistencies

[Anonymous]

First I want to thank the moderators and members of this forum for your invaluable assistance in my credit repair journey. You guys and gals all rock. Now the rant.

For the past year I have been monitoring my progress and how my stats change based on things which I am cognizant of doing and puzzling over changes which appear to happen for reason which cannot be explained. I am beginning to thing the whole CRA industry is a huge shell game which holds itself out to be fair, objective, transparent and unbiased. Frankly none of those words apply. Here is why I have some doubts.

I have subscribed to Myfico and monitor changes as they happen along with subscribing to each of the CRA products so I know exactly what they are reporting. No offense Myfico. Trust but verify is the operative motto.

As of today, Myfico 8 scores are EQ 718, TU 730 and EX 723.

Pulling today’s score reported by the CRA’s today are EQ 630, TU 623, and EX 723.

I won’t bore everyone with the Myfico scores for the other models none of which bear any resemblance to the scores reported by the actual CRA.

The standout exception is EX which appears to routinely report either identical or very close match up to Myfico 8 scores.

I realize TU uses their maddeningly different Vantage model which appears to be an outlier to the other two. I have questioned equally maddening and criminally irresponsible EQ about what model they use and the response I got was “it is an internal proprietary model”. EX seems to be the only CRA who is routinely transparent and my tracking if stats over a year pretty much track exactly to FICO 8 as reported here.

I look at the time delay on reporting wrinkle and for the most part now the CRA’s are pretty much reporting using identical data points. EQ is routinely the most delinquent in timely reporting which often has them lagging at least 30 days. Clearly a company with huge internal process problems over and above their criminally irresponsible data breaches of which I was a victim.

So, final conclusion seems clear. This whole process of tracking and reporting credit scores is not only an imperfect science it is arbitrary, devoid of transparency and appears to be intentionally manipulated by “secret” formulas or processes.

Of all the CRA’s EX seems to standout as the only one of these bandits who actually appear to operate responsibly and a year of data tracking confirms they utilize known models accurately.

It is clear to me this industry needs a dramatic transparency overhaul and needs to be monitored much closer to ensure that they are providing a correct, consistent and non-gamed result to both creditors and those on whom they report.

I be interested to hear other Myficoers opinions.

Caveat Emptor.

[Anonymous]

There are only 3 factors that go into generating a credit score:

1 - The scoring model used (FICO 08, VS 3.0, etc)

2 - The bureau data used (EX, TU, EQ)

3 - The point in time that the first 2 pieces of data were used

If you are speaking about your MF FICO 08 scores, you're answer to #1 above is "FICO 08"

If you're getting all 3 scores, you answer is all for #2... so you're getting your EX FICO 08, TU FICO 08 and EQ FICO 08.

The time should be attached to the generation of any score, found somewhere on the screen where any score is shown.

When you say pulling scores from the "CRAs" I'm not sure what you mean.  Obviously the CRA you are getting the score from answers #2, but what scoring model are you speaking of?  #1 is extremely important here. 

Any two scores generated at the same time with the same scoring model and same bureau data will be identical.  If you're seeing FICO 08 scores provided by MF that are different than other scores you are seeing, they either aren't FICO 08 scores, or are being pulled at different times where big changes are taking place (based on the large variance between your scores).  Since the chances of that are pretty slim, I'd say the scoring models are different... or, #1 from my list above.  If that's the case, you're comparing apples to oranges when looking at your MY FICO 08 scores compared to the other scores that are 70-100 points different.

[Anonymous]

I am very aware of how he system works. What is maddening is the use of “non transparent proprietary models” ie what EQ does. After many calls they refused to tell me what scoring model they use other than the what I mentioned above. I am a relentless data tracking nerd and there is no explanation for the inconsistencies as large as they are if they are using a FICO model. As far as your question about what I mean about pulling CRA data, I subscribe to their individual products so I know exactly what they report on any given day. With the exception of EX which is routinely accurate to what I see on Myfico the others are in a land of their own.

[Remedios]

He's comparing Vantage scores to Fico scores.

[Anonymous]

You're over thinking this, of the 3 CRAs, only Experian provides FICO 8, that's why it tracks the EX FICO 8 score from myfico. TU and EQ do not provide FICO 8 scores, that's why they don't match your TU and EQ FICO 8 scores from myfico. It's that simple, no CRA conspiracies.

Ignore the scores from TU and EQ, you're paying for the daily reports and the ability to lock/unlock your profile.

[Anonymous]

---

@AnonymousI am very aware of how he system works.

---

Then you are very well aware of points 1-3 that I made in my first reply. 

If you're seeing large inconsistencies between scores you're comparing, at least one of those factors is not the same that went into generating that score.  9 times out of 10 that's going to be the scoring model.

As for FICO scores, there are a ton of different versions; I think MF provides you with 28 or so?  There are some members on here such as SJ for example that have referenced a 100-110 point variance between two FICO scores from different models... that's not even considering something else like VS 3.0 which of course isn't even FICO.

[Anonymous]

It would be interesting if there was a way to have an equivalence table (i.e. VS 2.0 is equivalent FICO X). 

[Anonymous]

The only thing like that is an equivalence of names.   E.g....

          EX FICO 98

                = Experian FICO Score 2  
                = Experian FICO Risk Model v2

are all different names for the same model.

But if two models were actually developed by different teams of people (e.g. FICO 8 Classic vs. FICO 8 BE) then they won't be equivalent, aside from possibly having the same range.

[Anonymous]

---

@AnonymousIt would be interesting if there was a way to have an equivalence table (i.e. VS 2.0 is equivalent FICO X). 

---

That wouldn't be possible, because different scoring models give weight to different profile factors. 

Using your scoring models above, Jill may have a 700 VS 2.0 score and a 740 FICO X score, but Steve may have a 740 VS 2.0 score and a 700 FICO X score.  The point here is that there wouldn't be any correlation between VS 2.0 and FICO X that could be drawn up to create an equivalence table due to different profiles and different strength factors depending on the scoring model. 

[Anonymous]

Ya I see your point. I guess my ocd kicks in when there are variations based on identical data. Thanks for the insight. But I still do think EQ is a criminally irresponsible poorly managed organization based on their extreme carelessness with sensitive data and being persistently behind in in data updates compared to the other CRAs. I will end my rant with that comment.