topic Re: AAOA Scoring Method in Understanding FICO® Scoring

AAOA Scoring Method

cem13 — Tue, 17 Jan 2017 22:33:22 GMT

Hi Guys/Gals,

I found something very interesting that may or may not be true. I have cleaned up my file since my BK7 in JAN2011. Each year, though GW or other methods, I have slowly and I mean SLOWLY; but surely, improved my file. The kicker: my scores have NOT improved.

My 3B yearly cycle in in March of each year. This year on New Year's Day, I pulled my Annual Credit Report just to compare. In MAR16, I was showing 1-30 day late and 1-60 day late from 2009. It aged off this year without any changes in scoring. This confirms what the TU Supervisor told me a few years ago. That is, if an account is flagged as negative, it is scored for 2 years since the DOFD, then it is no longer scored as negative. I have been range bound on my scores for about 4 years now (+/- 8 points) depending on my CC utilization.

My overall AAOA has increased steadily as the closed accounts prior to BK7, have aged off. My new accounts since BK7 have improved month over month. In other words, as 1 or 2 accounts dropped at the 10 year level, my 10 new CC and intallments accounts are aging.

I looked at the factors and my reported AAOA is 4 years. So I put a spreadsheet together and they are tracking my OPEN accounts. My closed accounts are not factored into my reported AAOA.

So my question is, what evidence does anyone have that the scoring model uses closed accounts in the AAOA? It is possible that the CRAs report the AAOA of open accounts while using the closed accounts for scoring purposes. However, I am starting to think that the closed accounts may NOT factor into the scoring.

I have read through about 20-30 threads on the subject here and on other baords and the conventional wisdom is that all accounts listed on the report is averaged for scoring purposes. However, I could not find one single piece of evidence of this being true.

I am now starting to think that only the open accounts AAOA is used for scoring. Thoughts?

Re: AAOA Scoring Method

Anonymous — Tue, 17 Jan 2017 22:38:22 GMT

Mine uses closed accounts.....so not sure.....maybe u are in a different bucket

Re: AAOA Scoring Method

rooeez — Tue, 17 Jan 2017 22:47:13 GMT

My FICO 08 scores use open and closed accounts. My Vantage 3.0 scores use only open.

http://www.myfico.com/credit-education/whats-in-your-credit-score/

States average of "all" accounts.

Re: AAOA Scoring Method

Anonymous — Tue, 17 Jan 2017 22:55:28 GMT

When you say....

I looked at the factors and my reported AAOA is 4 years. So I put a spreadsheet together and they are tracking my OPEN accounts. My closed accounts are not factored into my reported AAOA.

... what tool are you using to check your AAoA? If it is Credit Karma, they are well known for having an incorrect summary page for age related scoring factors.

FICO and Vantage are the two dominant players in the credit scoring world. Both count closed and open accounts together in their age related factors, whether AAoA (Average Age of Accounts) or Age of Oldest Account.

Re: AAOA Scoring Method

rooeez — Tue, 17 Jan 2017 23:09:15 GMT

@Anonymous wrote:

FICO and Vantage are the two dominant players in the credit scoring world. Both count closed and open accounts together in their age related factors, whether AAoA (Average Age of Accounts) or Age of Oldest Account.

I do remember someone stating that Vantage 3.0 scores do take into account open and closed accounts as well, but for some reason Credit Karma uses an outdated algorithim that still only uses open accounts. I checked Credit Sesame, also Vantage 3.0) and it is in line with my FICO AAoA, so I stand corrected. Thanks.

Re: AAOA Scoring Method

cem13 — Tue, 17 Jan 2017 23:31:26 GMT

@Anonymous wrote:
When you say....

I looked at the factors and my reported AAOA is 4 years. So I put a spreadsheet together and they are tracking my OPEN accounts. My closed accounts are not factored into my reported AAOA.

... what tool are you using to check your AAoA? If it is Credit Karma, they are well known for having an incorrect summary page for age related scoring factors.

FICO and Vantage are the two dominant players in the credit scoring world. Both count closed and open accounts together in their age related factors, whether AAoA (Average Age of Accounts) or Age of Oldest Account.

I am not using any tool except for MS Excel to calculate the average age of account based on my CRs. I do not count the month that the account was opened regardless of what date it is opened. In other words, I age an account open on the 1st of the month the same as an account opened on the 31st of that same month.

This might not be how it is actually calculated but my manually calculated AAOA. However I think it is a mute point becuase the AAOA as reported by the CRAs is the same. The difference is that the AAOA that is reported by the CRAs match my AAOA of my OPEN accounts, not ALL accounts.

So the AAOA as reported to consumers may not be the AAOA calculated in the scoring.

Re: AAOA Scoring Method

cem13 — Tue, 17 Jan 2017 23:34:23 GMT

@rooeez wrote:

I do remember someone stating that Vantage 3.0 scores do take into account open and closed accounts as well, but for some reason Credit Karma uses an outdated algorithim that still only uses open accounts. I checked Credit Sesame, also Vantage 3.0) and it is in line with my FICO AAoA, so I stand corrected. Thanks.

This supports my theory. The scoring by FICO is calulated by what the CRAs gather in the file and report. In other words, FICO is the formula based on what inputs are given to them.

Re: AAOA Scoring Method

cem13 — Tue, 17 Jan 2017 23:37:08 GMT

@rooeez wrote:
My FICO 08 scores use open and closed accounts. My Vantage 3.0 scores use only open.

http://www.myfico.com/credit-education/whats-in-your-credit-score/

States average of "all" accounts.

Thanks for the link. I read it and I am certainly not convinced by the wording. Maybe I am reading too much into it, like a lawyer, but it does not explicitly say all accounts OPEN AND CLOSED.

If they only used open accounts, the statement could still hold true. For example, "Yes, we use all accounts,.........that are open"

Re: AAOA Scoring Method

cem13 — Tue, 17 Jan 2017 23:40:06 GMT

The other point that is true for me is that any account reported as negative, only calculates as negative for the first 2 years. Then after the 2 years, it is not judged as negative for scoring.

I am not certain if such an account is calculated into the AAOAs yet.

Re: AAOA Scoring Method

rooeez — Tue, 17 Jan 2017 23:43:49 GMT

@cem13 wrote:
@rooeez wrote:

I do remember someone stating that Vantage 3.0 scores do take into account open and closed accounts as well, but for some reason Credit Karma uses an outdated algorithim that still only uses open accounts. I checked Credit Sesame, also Vantage 3.0) and it is in line with my FICO AAoA, so I stand corrected. Thanks.
This supports my theory. The scoring by FICO is calulated by what the CRAs gather in the file and report. In other words, FICO is the formula based on what inputs are given to them.

Yes, each CRA can only calculate a FICO based on the specific information/accounts in their report. For instance, TU always reports the open date as the 1st of the month, whereas EX and EQ report the actual day. So, the AAoA is off, but all three still use all accounts, not just open.

Re: AAOA Scoring Method

Anonymous — Wed, 18 Jan 2017 00:31:00 GMT

You write:

"... the AAOA that is reported by the CRAs match my AAOA of my OPEN accounts, not ALL accounts."

But the CRAs don't report AAoA. That is, they don't compute an average age of accounts. That is a computation that is made by FICO (or Vantage, etc.) after the scoring algorithm obtains your raw data, based on your date opened. The CRAs are simply a database containing a listing of all your accounts, which will have a date opened, a date closed, etc.

You seem aware of this, because in your next post you write:

"The scoring by FICO is calulated by what the CRAs gather in the file and report. In other words, FICO is the formula based on what inputs are given to them."

Yup, that's right. AAoA is calculated by the scoring algorithm (FICO, Vantage, etc.) based on the raw data from the CRA.

I also am hearing that you are doing your own AAoA calculation, using a spreasheet, wherein you count only open accounts. But I am still unclear why you think it is likely that FICO's scoring algorithm is using your method of open only rather than considering both open and closed.

Re: AAOA Scoring Method

cem13 — Wed, 18 Jan 2017 01:17:50 GMT

@Anonymous wrote:

I also am hearing that you are doing your own AAoA calculation, using a spreasheet, wherein you count only open accounts. But I am still unclear why you think it is likely that FICO's scoring algorithm is using your method of open only rather than considering both open and closed.

Since I have all my accounts loaded into MS Excel, I can average them any which way I want.

1) Average age of all accounts,

2) average age of closed accounts,

3) Average age of open accounts,

4) Average age of accounts opened on Tuesday

5) Average age of accounts opened after a Michigan Victory

6) Average age of accounts opened and then my wife bit^&es at me for doing so.

The latter is my lowest average age because she seems to not like many open accounts. She is old school and has 2 cards with $5,000 on each.

The reason I question the scoring is that I do not have a baseline to test. I have no basis either way actually.

Either way, AAOA does not seem to affect me in my bucket when I still have the BK7 reporting.

Re: AAOA Scoring Method

Anonymous — Wed, 18 Jan 2017 01:28:57 GMT

I especially like the last two methods.

Re: AAOA Scoring Method

Anonymous — Wed, 18 Jan 2017 01:37:04 GMT

Here are a few more articles about AAoA and closed accounts from respected sources outside the MF forums:

https://thepointsguy.com/2016/04/how-closing-card-impacts-fico-score/

http://www.thecreditsolutionprogram.com/the-truth-about-closing-accounts-that-are-in-good-standing/2/

http://creditcardforum.com/blog/cancelling-credit-card-without-hurting-credit-score/

Re: AAOA Scoring Method

Revelate — Wed, 18 Jan 2017 03:01:26 GMT

@cem13 wrote:
The other point that is true for me is that any account reported as negative, only calculates as negative for the first 2 years. Then after the 2 years, it is not judged as negative for scoring.

I am not certain if such an account is calculated into the AAOAs yet.

Honestly, neither of these assertions is accurate. I don't mean to be antonistic in debunking this but too much misinformation.

1) AAOA is open and closed accounts, as stated by FICO and the bureaus (regarding Vantage); also I have a confirming datapoint of an AAOA breakpoint at 2 years, and I assure you closed accounts were factored into my whopping 4 point gain. What you see on Credit Karma and even myFICO with the exception of the scores (and their changes) and the reason codes is unfortunately without meaning: it's just a 3rd party interpretation of the data in the file, that's it, absolutely no algorithm reference from any of them; access to the algorithm is strictly controlled. Sad but true.

2) 60 day lates certainly count the entire time, this whole 2 years thing may be a complete myth and a TU Supervisor unless he's a member here or similarly knows far less than the majority of members who visit this particular board: they do not have access to the algorithm, and I would suggest credit education of CSR's at the bureau is barely better than the average consumer. Unfortunately they're going to say things to get you off the phone, and that's true of any call center anywhere in the world... betrayed by human nature and incentive programs =/.

Anyway I have reason codes explicitly stating deliquency as my #1 reason on Experian on all models when my tax lien was removed and my last negatives of any type were a 30/60 day late from 2010, over 6 years from the pull. Ergo, full monty 7 years till exclusion the 60D factors. I'm less certain on 30 day lates, unfortunately my tax lien on TU just got another year tacked on it so I'm not going to be able to see what my Oct '15 30 day late does at the 2 year mark on an otherwise clean file, but presumably if it still counts when the lien comes off in 2018 there will still be deliquency marks in my reason codes... or if I go to pretty much the same score as EX/EQ with it's being 3 years old then presumably it doesn't count (or at least not for much).

Re: AAOA Scoring Method

cem13 — Wed, 18 Jan 2017 03:48:04 GMT

@Revelate wrote:
2) 60 day lates certainly count the entire time, this whole 2 years thing may be a complete myth and a TU Supervisor unless he's a member here or similarly knows far less than the majority of members who visit this particular board: they do not have access to the algorithm, and I would suggest credit education of CSR's at the bureau is barely better than the average consumer. Unfortunately they're going to say things to get you off the phone, and that's true of any call center anywhere in the world... betrayed by human nature and incentive programs =/.

Hi Revelate,

Thanks for chiming in on this subject.

I will agree that a TU Supervisor is not the best expert on FICO scoring. Although I would hope he knows what he is talking about.

For my situation, I did not see any score change when my 30 day late and 60 day late accounts aged off. This occured on all 3 CRAs during the same month this year (OCT). So for my bucket, it did not change my score. Now, it could be that my bucket is holding be back and that if I was in a different bucket, I would see a score increase.

I have been maxed out on my scores for 2 years now. I will agree that my AAOA has dropped by at least a year due to my APR16 app spree when I opened 3 more accounts. Again, no significant score loss. My bucket may not allow me to test. I had several old accounts; some clean some not-so-clean, age off. Again, no significant increase or decrease in score.

For some reason EQ has been the most consistant score. For the past 3.5 years my low was 699 and the high was 712. The 712 when I was carrying 2000 on a 25,000 card with $200,000 overall CL. The only baddies were the 30/60 late and BK7.

The low of 699 came when I purchased (APR) 10 acre wooded, hunting property on a cash advance. I hit it for 18K on a 25K card. I quickly did a BT to 0% for 12 months. The score started to improve at the 90% and 30% level.

I also purchased some lumber to build a small cabin so that 2000 increased to $15K on a $25K CL. I will continue to monitor these utilizations. I stand at 703 EQ and I hope to get back to 712 later this year when I pay these cards down to $0.

My plan to to pay the minimum on one card and $1500 per month on the other. I will be monitoring the scores at the 50% (no change expected) and 30% (2-3 point increase expected).

One more point is that with all of my CCs and the BT offers; I bought a $20K land and spent another $15K for a cabin: all without a mortgage and little interest paid. I did get hit with $20 for the few days I had the cash advance balance. I saved about $2000 in mortgage fees etc.

PenFed 0% BT and Barclays 0% BT helped make this dream a reality.

Re: AAOA Scoring Method

Revelate — Wed, 18 Jan 2017 04:59:31 GMT

What else is on your file?

AAOA breakpoints aren't on individual years, i.e. there's one at 2 years, there's not one at 3 based on my data, theorized breakpoints by various people here are around 5 and 8 years. I didn't see in the thread what you suggested your AAOA was before and after.

If you have other dirty bits on your file, you're not likely to get a significant change if any, and if you're still around 700 I would assert you have other issues on it.

Making assertions needs fuller disclosure of data .

Re: AAOA Scoring Method

Anonymous — Wed, 18 Jan 2017 12:57:49 GMT

With the presence of another major late payment on another account, having a 3-4 year old 60 day late payment removed from 1 of my accounts yielded zero point gain. Based on this I would say that if one possesses a major late payment on one account, removing a 60 day late from another account that's about half way to aging off has no impact on score. The question, then, is did those points already come back at some other threshold (perhaps the 2 year mark), OR, does a 60 day late in the presense of a major delinquency on another account simply not carry any weight?

Re: AAOA Scoring Method

cem13 — Wed, 18 Jan 2017 17:28:26 GMT

@Anonymous wrote:
. The question, then, is did those points already come back at some other threshold (perhaps the 2 year mark), OR, does a 60 day late in the presense of a major delinquency on another account simply not carry any weight?

In my case, the data is clear. After my BK7, I improved my score from high-500s to 700 in 23 months. I had several accounts with various degrees of negative marks that either aged off or GW off. None of these significantly impacted my scores.

As I mentioned, EQ has been my most consistant score hovering around 705 +/- 8 since DEC2013 (3 years). The only factor that triggers a myfico alert and scoring change is the CC utilization. Today my EQ is at 703 (F '08) and soon to move up as I pay my CC down.

AAOA and negative accounts dropping off did not change my score. I have two more cards with balances. I will pay them off individually to check the score improvement. If my scores go about my previous high of 714 (APR16), then I can say that something OTHER than CC utilization affect my scores.

However, if I get my CC balances back down to 1% on one card, and I max out at EQ (F '08) of 714; this will be basically the same file as I had before (SAVE AAOA and negative accounts aging off). Only the BK7 is reporting now. All accounts (both open and closed) are positive.

So my theory is based on my file: With a BK7 reporting, AAOA and Negative tradelines are NOT holding back my score since I did not see an increase as the AAOA improves and the baddies age or GW off. I suspect that I "maxed out" during my rebuild through the 2 year mark of the BK7.

Re: AAOA Scoring Method

Revelate — Wed, 18 Jan 2017 18:11:33 GMT

Well PR's > lates, sure that makes sense, though I don't know if a conclusion can necessarily be drawn from that.

Regarding AAOA, really have to post both before and after, current at 4 years, what'd you start with? Entirely possible 4 isn't a breakpoint either, just like 3. Re-calculate with closed accounts factoring and what's the starting and end point where your score has been basically fixed (like my Beacon 5.0, at 693 or 700 depending on inquiry count for 2 years now).

I may get the opportunity over the next year (if I don't get any EE) to see if the lates coming off affect my score, without the tax lien on there they clearly do, but with the tax lien I think directly correlates to your supposition.