TU4 length of revolving accounts threshold found!

[Curious_George2]

---

@Anonymous wrote:

---

@Curious_George2 wrote:

---

@Anonymous wrote:

Yeah, 1 point isn't much.  Proof though that a negative reason statement only needs to be worth a single Fico point to be present.  It does make me wonder how much the top negative reason statement is worth.  What's the top score possible on TU4 again?  I know it's under 850, but can't remember exactly... maybe 835 or so?  Whatever that top score is, has anyone ever reported hitting it, or is TU4 one of those models that's essentially impossible to get a perfect score on?

If it is possible to get a perfect TU4 score, I suppose whatever the top score is minus 820 could potentially be what my remaining negative reason statement is worth.

---

Congrats! And thanks for sharing.

I'm not sure I share the conclusion bolded above. Isn't it also possible you just went from, say, a 5 point penalty to a 4 point penalty and the rule is that factors are suppressed if they cost less than 5 points?

I'm pulling those specific numbers out of thin air, but can we really rule out the possibility that you are still getting a penalty that is one point lower than it was at AoORA<20?

---

@Curious_George2 i've never heard a theory before about a reason code only appearing if it's worth 5+ points. Is there any documentation for anything like that?

---

I was just using a made-up value -- 5 points -- as an illustration (hence my statement that I pulled the numbers out of thin air). Is there documentation of regarding minor reasons being suppressed? Dunno. Maybe not. To my knowledge, it's just a shot-in-the-dark theory I cooked up to account for observed facts. Sometimes there are fewer than four reasons shown with a score that is too low for those particular penalties to plausibly be the whole story. For example, Remy showed us her 761 EX2 with only two reasons displayed (Short account history and Accounts with balances). A few of us were discussing that about a month ago, and I floated this same suppression theory in the Spree Penalty thread, starting here. No one bought it then either.  That's fine, but I don't think anyone had an accepted explanation for what we saw.

To my mind, suppression of minor factors isn't a particularly radical concept, since we know FICO already does it in some contexts. Namely, when a classic score hits 800 or an industry score hits 850. Isn't the official explanation of that well known suppression essentially "these factors are about very small penalties, so we think it would confuse consumers if we showed them"? It seems like most folks here don't think FICO would apply that rationale by suppressing minor factors on lower scores. I don't see why they wouldn't.

[Curious_George2]

---

@Anonymous wrote:

So I've had the same 2 negative reason statements on my TU4 score for what feels like years...

1 - Length of time accounts have been established

2 - Length of time revolving accounts have been established

I just noticed through my mortgage lender that my TU4 score updated (on 3/9/21) and for the first time the second negative reason statement has dropped off, leaving only the first one I mentioned above.  In doing so, my score increased 1 point from 819 --> 820.

I feel very confident in saying that the factor that caused this change was my AoORA hitting 20 years on 3/1/21. 

---

Hold up. This just occured to me. How/where are you seeing reasons with a TU4 over 800? Don't those get hidden from view when your score is in the 800s? Mine are hidden on MF. Apparently you've got better sources than me!

[Anonymous]

---

@Curious_George2 wrote:

---

@Anonymous wrote:

So I've had the same 2 negative reason statements on my TU4 score for what feels like years...

1 - Length of time accounts have been established

2 - Length of time revolving accounts have been established

I just noticed through my mortgage lender that my TU4 score updated (on 3/9/21) and for the first time the second negative reason statement has dropped off, leaving only the first one I mentioned above.  In doing so, my score increased 1 point from 819 --> 820.

I feel very confident in saying that the factor that caused this change was my AoORA hitting 20 years on 3/1/21. 

---

Hold up. This just occured to me. How/where are you seeing reasons with a TU4 over 800? Don't those get hidden from view when your score is in the 800s? Mine are hidden on MF. Apparently you've got better sources than me!

---

As you quoted above, the score and reason codes come from my mortgage lender.  Different lenders/sources stop providing reason codes at different scores.  Many omit them at 800, where others go well above that.  Discover for example provides negative reason codes up to 849 and only omits them at 850.

[Anonymous]

My AAoRA is actually significantly higher than I had previously thought/calculated... perhaps because my oldest revolver (that just hit 20 years) appears 3 times on my CR, so I'm getting the benefit of that old account 3 times.

On 3/1/21 my AAoRA moved from 108 --> 109 months on TU, in years going from 9.0 to 9.08, so it wouldn't appear to me that a threshold was crossed there.

[Curious_George2]

---

@Anonymous wrote:

---

@Curious_George2 wrote:

---

@Anonymous wrote:

So I've had the same 2 negative reason statements on my TU4 score for what feels like years...

1 - Length of time accounts have been established

2 - Length of time revolving accounts have been established

I just noticed through my mortgage lender that my TU4 score updated (on 3/9/21) and for the first time the second negative reason statement has dropped off, leaving only the first one I mentioned above.  In doing so, my score increased 1 point from 819 --> 820.

I feel very confident in saying that the factor that caused this change was my AoORA hitting 20 years on 3/1/21. 

---

Hold up. This just occured to me. How/where are you seeing reasons with a TU4 over 800? Don't those get hidden from view when your score is in the 800s? Mine are hidden on MF. Apparently you've got better sources than me!

---

As you quoted above, the score and reason codes come from my mortgage lender.  Different lenders/sources stop providing reason codes at different scores.  Many omit them at 800, where others go well above that.  Discover for example provides negative reason codes up to 849 and only omits them at 850.

---

That's cool. 

I had assumed the 800 rule was a feature of the scoring software itself. Looks like I was wrong about that. Seems like this supports the claim that FICO always returns up to 4 (I think someone, maybe Birdman, said 5) factors and each lender or CMS front-end developer can make its own choice about which ones to display.

This makes it even more troubling to me that MF suppresses all factors at 50 points below max score. Starting to feel foolish paying them for info they are deliberately worsening in multiple ways. 

[Anonymous]

I can't say I disagree with you.  I've never had a MF membership and have never even considered one to be honest. 

Another fun fact is that Discover not only provides reason statements up to 849, but also provides more than 4 negative reason statements on their CLI denial letters if they're there.  Back when my scores were lower I usually had 5 statements on my weekly CLI request denial letters.

[Anonymous]

Yeah it’s up to five with inquiries. And the reason for not giving codes above whatever cut off is they say you’re already high enough basically. But that’s why you can go to the industry standard and typically still get codes. But it’s standard they cut them off at most places at 800 & 850.

But those are known features, now when it comes to version 8, codes can be screwy, but otherwise they should be showing if they’re below the cut off unless there’s a problem.

[Anonymous]

BBS, I’ve often wondered if there is a feature to exclude doubles of accounts. If there is, then your average age might be the initial calculation.

[Anonymous]

---

@Anonymous wrote:
BBS, I’ve often wondered if there is a feature to exclude doubles of accounts. If there is, then your average age might be the initial calculation.

---

I don't know, my gut says that's not the case.  The accounts aren't exact doubles, as they all contain slightly different information.  All of them have different account closure dates, last payment dates, etc.  The only constant across 3 is the date opened.  The accounts got duplicated when the card was "lost or stolen" so whatever the lost/stolen date is shows as the account closure date.