cancel
Showing results for 
Search instead for 
Did you mean: 

How accurate is CreditXpert?

tag
KSK1912
Valued Contributor

Re: How accurate is CreditXpert?


@jdxprs1 wrote:

@KSK1912 wrote:

@ALHondaGuy wrote:

Unless they are pay for deletes AND your only baddies, I don't see how that's accurate. Paying a charge-off or collection will not improve your scores. It would only make you look good in manual underwriting.


+1 Especially a very old one such as 6 years. You might have some luck calling CRA's and asking for Early Exclusion and get them removed without even paying. It's worth the try.


A charge off that isn't sent to a collection agency, can still show a balance. If it still shows a balance it's being added to your utilization. That's why I gained so many points after paying. 


That is good to know. I did not know that. Thanks.


6/14/15: Equifax 576 Experian 542 Transunion 571
8/17/15: Equifax 626 Experian 619 Transunion 622
9/28/15: Equifax 655 Experian 636 Transunion 641
11/11/15: Equifax 698 Experian 687 Transunion 672
Short Term Goal Score: 720 Across the board


Take the myFICO Fitness Challenge


Are you free Tuesday..?
Message 21 of 27
jdxprs1
Frequent Contributor

Re: How accurate is CreditXpert?

I didn't know either. That's why I started this topic. Couldn't believe it but the software was correct.
Last app 1/2/2017.
In the garden until at least 6/2/2017.
Message 22 of 27
Anonymous
Not applicable

Re: How accurate is CreditXpert?

Credit Xpert is super accurate.

Message 23 of 27
Thomas_Thumb
Senior Contributor

Re: How accurate is CreditXpert?

I pulled my CreditXpert 3B scores and credit report summary through a $1 PrivacyGuard 14 day trial yesterday. The credit report summary was nicely done. The interactive report has a great feature that compares CRA score differences and summarizes findings into High impact, moderate impact and low impact differences. 

 

In my case, the large range in CreditXpert scores suggests CRAs have the ability to unduly tweak the scoring model (just like they do with Mortgage Fico scores) -  and the CreditXpert model applied to EQ is a bit skitish toward # of cards reporting balances (very similar to what I see with EQ Fico 04 used for mortgages). EQ and TU mortgage Ficos are based on the Fico 04 algorithm. The EX Fico mortgage algorithm is actually based on an even older Fico 98 model. 

 

Side note score range for CredtXpert is 350 to 850.  The range for mortgage Ficos is 300 to 850. For a complete list of credit scores and there ranges click on the below shortcut.

 

http://ficoforums.myfico.com/t5/Understanding-FICO-Scoring/The-many-flavors-of-FICO-Editions-version...

 

The results I see with CreditXpert model suggest a high degree of similarity with Fico 04 scores.This is not surprising as CreditXpert is not a new model. The newer Fico 8 model/scores and VantageScore 3.0 model/scores are quite different from the older Fico 04 and Fico 98 algorithms used for the majority of mortgage reviews.

 

The amount of variation in CrXp scores I see between the CRAs is rather disconcerting but, I see that with Fico mortgage scores as well. Fico has taken steps to reduce variability among the CRAs with their Fico 9 model and VantageScore has done this with VS 3.0. The best way to judge "accuracy" of a credit scoring model - IMO is: How good is the model at providing repeatable and consistent scores among the three major CRAs - given data among the CRAs is nearly the same. The older Fico mortgage models and CreditXpert are quite lacking in consistency between CRAs.

 

I hope these older models get displaced with more consistent "modern" alternatives as part of mortgage application reviews.

 

CreditXpert scores.jpg

 

CreditXpert is significantly more reflective of Fico mortgage scores (Fico 04 model) than VantageScore 3.0 or Fico 8 - at least for my profile.

 

During my trial period I did some what if testing of the CreditXpert Simulator. I like how PrivacyGuard has set up the user interface for the CreditXpert simulator. Results from my scenarios are shown below

 

CreditXpert Simulation Factor DescriptionEQTUEX
Reduce credit card debt (both $1000 & $1500)2703
New Auto installment loan (both $10k and $25k)-8-30-13
New credit card ($10k CL, $100 balance)-14-24-24
Reduce mortgage by $50k (currently at $70k)2300
Max out all revolving credit cards-61-91-87
One 30 day late on a revolving account-56-55-80
One 30 day late on all revolving accounts-99-72-85
One 90 day late on one revolving account-106-175-178

 * AAoA = 16 years, Oldest account 33 years, youngest 6 years. Aggregate CC balance $2200 prior to simulation paydown.

 

The CrXp simulator is a bit flawed for reduction in credit card debt and reducing mortgage.

 - Reducing CC debt from current 2% utilization to 0.5% UT would have no impact on score. However, reducing # cards reporting balances from 5 to (3 or 2) does improve my EQ mortgage score from 788 to 809.

- Reducing mortgage debt based on my current B/L of 30% would have no impact on score - 23 point improvement on EQ is inaccurate.

- A new auto installment loan would help my credit mix and result in a score increase or no change on Fico 8 or VS 3.0 (going from 0 to 1 hard inquiries in Fico 8 and VS3 has no impact on my score. However going from 1 to 0 inquiries did improve my EQ Fico 04 mortgage score 13 points. Net impact on Fico mortgage scores might be -5 to -10 points. The CreditXpert simulator is a bit off but, not too much

- Adding a new CC account with $10k CL and $100 balance would again result in a hard inquiry. Again, this would have no impact (positive or negative) on my Fico 8 or VS 3.0 scores. However, the inquiry could drop my Fico 04 mortgage scores up to 10 points. The CreditXpert simulator over predicts impact by 100% or more.

 

Impact of max out and lates would be severe as suggested by the simulator. No plans to test these to determine actual magnitude Smiley Wink

 

Conclusion:

 

The CreditXpert model as provided through PricacyGuard calculated scores that are a close approximation to my Fico 04 scores. The simulator is flawed (as are all simulators due to their built in simplicity). Nonetheless, the simulator is a useful tool that more often than not provides results that are "in the ballpark"

 

I would highly recommend considering signing up for PrivacyGuard's 14 day trial offer for $1.

Fico 9: .......EQ 850 TU 850 EX 850
Fico 8: .......EQ 850 TU 850 EX 850
Fico 4 .....:. EQ 809 TU 823 EX 830 EX Fico 98: 842
Fico 8 BC:. EQ 892 TU 900 EX 900
Fico 8 AU:. EQ 887 TU 897 EX 899
Fico 4 BC:. EQ 826 TU 858, EX Fico 98 BC: 870
Fico 4 AU:. EQ 831 TU 872, EX Fico 98 AU: 861
VS 3.0:...... EQ 835 TU 835 EX 835
CBIS: ........EQ LN Auto 940 EQ LN Home 870 TU Auto 902 TU Home 950
Message 24 of 27
medicgrrl
Valued Contributor

Re: How accurate is CreditXpert?


@Thomas_Thumb wrote:

I pulled my CreditXpert 3B scores and credit report summary through a $1 PrivacyGuard 14 day trial yesterday. The credit report summary was nicely done. The interactive report has a great feature that compares CRA score differences and summarizes findings into High impact, moderate impact and low impact differences. 

 

In my case, the large range in CreditXpert scores suggests CRAs have the ability to unduly tweak the scoring model (just like they do with Mortgage Fico scores) -  and the CreditXpert model applied to EQ is a bit skitish toward # of cards reporting balances (very similar to what I see with EQ Fico 04 used for mortgages). EQ and TU mortgage Ficos are based on the Fico 04 algorithm. The EX Fico mortgage algorithm is actually based on an even older Fico 98 model. 

 

Side note score range for CredtXpert is 350 to 850.  The range for mortgage Ficos is 300 to 850. For a complete list of credit scores and there ranges click on the below shortcut.

 

http://ficoforums.myfico.com/t5/Understanding-FICO-Scoring/The-many-flavors-of-FICO-Editions-version...

 

The results I see with CreditXpert model suggest a high degree of similarity with Fico 04 scores.This is not surprising as CreditXpert is not a new model. The newer Fico 8 model/scores and VantageScore 3.0 model/scores are quite different from the older Fico 04 and Fico 98 algorithms used for the majority of mortgage reviews.

 

The amount of variation in CrXp scores I see between the CRAs is rather disconcerting but, I see that with Fico mortgage scores as well. Fico has taken steps to reduce variability among the CRAs with their Fico 9 model and VantageScore has done this with VS 3.0. The best way to judge "accuracy" of a credit scoring model - IMO is: How good is the model at providing repeatable and consistent scores among the three major CRAs - given data among the CRAs is nearly the same. The older Fico mortgage models and CreditXpert are quite lacking in consistency between CRAs.

 

I hope these older models get displaced with more consistent "modern" alternatives as part of mortgage application reviews.

 

CreditXpert scores.jpg

 

CreditXpert is significantly more reflective of Fico mortgage scores (Fico 04 model) than VantageScore 3.0 or Fico 8 - at least for my profile.

 

During my trial period I did some what if testing of the CreditXpert Simulator. I like how PrivacyGuard has set up the user interface for the CreditXpert simulator. Results from my scenarios are shown below

 

CreditXpert Simulation Factor DescriptionEQTUEX
Reduce credit card debt (both $1000 & $1500)2703
New Auto installment loan (both $10k and $25k)-8-30-13
New credit card ($10k CL, $100 balance)-14-24-24
Reduce mortgage by $50k (currently at $70k)2300
Max out all revolving credit cards-61-91-87
One 30 day late on a revolving account-56-55-80
One 30 day late on all revolving accounts-99-72-85
One 90 day late on one revolving account-106-175-178

 * AAoA = 16 years, Oldest account 33 years, youngest 6 years. Aggregate CC balance $2200 prior to simulation paydown.

 

The CrXp simulator is a bit flawed for reduction in credit card debt and reducing mortgage.

 - Reducing CC debt from current 2% utilization to 0.5% UT would have no impact on score. However, reducing # cards reporting balances from 5 to (3 or 2) does improve my EQ mortgage score from 788 to 809.

- Reducing mortgage debt based on my current B/L of 30% would have no impact on score - 23 point improvement on EQ is inaccurate.

- A new auto installment loan would help my credit mix and result in a score increase or no change on Fico 8 or VS 3.0 (going from 0 to 1 hard inquiries in Fico 8 and VS3 has no impact on my score. However going from 1 to 0 inquiries did improve my EQ Fico 04 mortgage score 13 points. Net impact on Fico mortgage scores might be -5 to -10 points. The CreditXpert simulator is a bit off but, not too much

- Adding a new CC account with $10k CL and $100 balance would again result in a hard inquiry. Again, this would have no impact (positive or negative) on my Fico 8 or VS 3.0 scores. However, the inquiry could drop my Fico 04 mortgage scores up to 10 points. The CreditXpert simulator over predicts impact by 100% or more.

 

Impact of max out and lates would be severe as suggested by the simulator. No plans to test these to determine actual magnitude Smiley Wink

 

Conclusion:

 

The CreditXpert model as provided through PricacyGuard calculated scores that are a close approximation to my Fico 04 scores. The simulator is flawed (as are all simulators due to their built in simplicity). Nonetheless, the simulator is a useful tool that more often than not provides results that are "in the ballpark"

 

I would highly recommend considering signing up for PrivacyGuard's 14 day trial offer for $1.


Ok, I've got my sister and brother in law trying this out for their refi.  I'll update once we get to the point in real life that we simulated.  Probably won't be until the beginning of August, unless I feel generous and pay off some of their debt.  Smiley Very Happy



EQ 778 EXP 782 TU 729
Message 25 of 27
Anonymous
Not applicable

Re: How accurate is CreditXpert?

I just signed up for the PrivacyGuard membership..thanks TT!

 

One question, my Equifax and Experian show a score of 582. However, my Transunion shows 625. If CreditXpert is based off the FICO 04 model, this is what myFICO uses for the TU mortgage score so can I safely assume that my TU mortgage score is now sitting at 625?! 

 

If so..yay me!! My CC payments haven't hit my reports yet, and the simulator says my Experian score will go up 31 points once my CC balances update. 

Message 26 of 27
Thomas_Thumb
Senior Contributor

Re: How accurate is CreditXpert?

A few points of clarifications regarding mortgage credit scores:

 

1) The Fico scores used by lenders for mortgages are

 - Equifax score 5 (Fico 04 model)

 - Transunion score 4 (Fico 04 model)

 - Experian score 2 (Fico 98 model - which is older than Fico 04). Experian score 3 uses the Fico 04 model. That score is included in MyFico 3B reports under the bankcard section. However, it is not the score mortgage lenders look at. Why? Lender databases are populated with EX Fico 98 scores so they would lose that comparison basis if they switched to a different model for new applicants.

 

2) CreditXpert is not the same as Fico 04. CreditXpert uses a credit scoring model they developed. One attribute of that model is it can closely approximate mortgage Fico scores and the simulator has a proven track record in providing mortgage applicants with accurate feedback on how to improve scores along with good ballpark estimates on score impact.

Fico 9: .......EQ 850 TU 850 EX 850
Fico 8: .......EQ 850 TU 850 EX 850
Fico 4 .....:. EQ 809 TU 823 EX 830 EX Fico 98: 842
Fico 8 BC:. EQ 892 TU 900 EX 900
Fico 8 AU:. EQ 887 TU 897 EX 899
Fico 4 BC:. EQ 826 TU 858, EX Fico 98 BC: 870
Fico 4 AU:. EQ 831 TU 872, EX Fico 98 AU: 861
VS 3.0:...... EQ 835 TU 835 EX 835
CBIS: ........EQ LN Auto 940 EQ LN Home 870 TU Auto 902 TU Home 950
Message 27 of 27
Advertiser Disclosure: The offers that appear on this site are from third party advertisers from whom FICO receives compensation.