EX Fico 8, -5 w/1 CC > 8.9%

[Anonymous]

I was at AZEO a few days ago, EX Fico 8 score at 850 as usual.  As many of you know, I've gone from AZEO to 8 of 8 revolvers with a balance and been able to maintain that 850 score.  That's with all cards having tiny balances reported of course.

Anyway, a few days ago my AZEO account was at 3% reported utilization with 1% aggregate utilization... 850 score.  Today with the AZEO card bumping up to 23% utilization (aggregate utilization shifts to 3%) I saw a 5 point ding.  It's also worth noting that the ding could actually be greater than 5 points due to potential top-end buffer.

I know many people report being able to take an individual card across the 8.9% threshold and see no scoring penalty until they cross 28.9%, but that's not the case on my profile.  I should also mention that the reported balance on the card in going from 3% utilization to 23% utilization moved from ~$750 to ~$6100, so it's possible that some raw dollar value being crossed may be a factor, as I know some have theorized dollars could be considered in part.  Curious to hear what you all think.

I plan on reporting a balance at about the mid-point of those two numbers above, say $3000-$3500 next cycle.  That will take utilization on that account down to probably 11%-13% or so... not enough to cross the 8.9% threshold, but definitely chopping the dollars reported about in half.  While not a perfectly clean test, if my score does rebound the 5 points it may give a little more credit to the theory that dollars do at times matter when it comes to utilization.

[Anonymous]

BrutalBodyShots ... don't go falling out of the High Achievers by falling on the sword

[Revelate]

Nice BBS.

Yeah I'm fairly confident that individual utilization metrics are profile dependent, there's been too many datapoint differences between my own data when I had that tax lien (when they counted) vs. their clean profiles that it almost has to be something weird.

Actually I am for once in a position to test this: I need to make an estimated tax payment before July 15th (of around 5k) which I'm planning to drop on the CFU: actually I'd probably be better to do this right now just so I can get the clean AAOA 5 year datapoint on EX again (not too worried about the additional grace period).

Should come in at around 29.2% revolving and like 3.5% aggregate assuming the HELOC is excluded from TCL.

Since the rest of my balances are easy to keep fixed I'll definitely catch EX and I'll try to catch TU as well (60D late derog scorecard), and heck might as well try for EQ on the 30D scorecard: maybe we can nail down profile dependence on it.  Will try to catch like a 10 or 9% datapoint on the way down, daily pulls on EX should allow me to distinguish between that.  Interest on a few days isn't that big of a deal and Chase isn't going to complain with the float for a month.

[Anonymous]

Good stuff Rev, I look forward to the result.

Do you think me crossing the 8.9% threshold on the individual card is the score-impacting factor at play, or do you think there's a chance that it's not the percentage but rather the dollars that could be considered?

[Revelate]

---

@Anonymous wrote:

Good stuff Rev, I look forward to the result.

Do you think me crossing the 8.9% threshold on the individual card is the score-impacting factor at play, or do you think there's a chance that it's not the percentage but rather the dollars that could be considered?

---

I think the only place I've seen balance itself potentially be a factor is EX FICO 2 / 3 (I think have to go double-check some December reports), I haven't seen or heard of it on FICO 8 and we've had some pretty gnarly balances over time.  

I think most of the newer algorithms including potentially TU / EQ FICO 04 are percentage based, EX FICO 3 might be an outlier on that reason code.   In any event my TU FICO 4 is sitting in a derog scorecard with only 3 reason codes so if it's even a small ding it should show up there if it exists in the bottom 4 scorecards at all.  Might not.

I honestly don't know why we don't have better data around individual tradeline utilization honestly =/

[jamie123]

---

@Anonymous wrote:

Good stuff Rev, I look forward to the result.

Do you think me crossing the 8.9% threshold on the individual card is the score-impacting factor at play, or do you think there's a chance that it's not the percentage but rather the dollars that could be considered?

---

I would higly doubt that dollars are included in any FICO calculation. The whole basis of FICO scores in the first place is to be able to generate a credit score that doesn't discriminate between rich and poor people.

That's why a poor guy like you, BrutalBodyShots, can generate a perfect score and a rich guy like Revelate has scores that suck! 

(Just joking!!! Lighten up!!!)

[Anonymous]

---

@jamie123 wrote:

---

---

I would higly doubt that dollars are included in any FICO calculation. The whole basis of FICO scores in the first place is to be able to generate a credit score that doesn't discriminate between rich and poor people.

---

I'm quite sure that dollars are included in some FICO algorithms... just possibly (likely) not FICO 8.  Rev above referenced a few that he believes may take into consideration dollar amounts.  I don't think dollars in this situation causes any discrimination between rich and poor people, as it's not income that's being considered but rather debt.  If there's data out there that suggests that groups of people that have X level of revolving debt in dollars are Y% more likely to default it wouldn't be out of the question that the algorithm could consider a ding based on dollars.  This would have nothing to do with income though, so no discrimination. 

[jamie123]

---

@Anonymous wrote:

---

@jamie123 wrote:

---

---

I would higly doubt that dollars are included in any FICO calculation. The whole basis of FICO scores in the first place is to be able to generate a credit score that doesn't discriminate between rich and poor people.

---

I'm quite sure that dollars are included in some FICO algorithms... just possibly (likely) not FICO 8.  Rev above referenced a few that he believes may take into consideration dollar amounts.  I don't think dollars in this situation causes any discrimination between rich and poor people, as it's not income that's being considered but rather debt.  If there's data out there that suggests that groups of people that have X level of revolving debt in dollars are Y% more likely to default it wouldn't be out of the question that the algorithm could consider a ding based on dollars.  This would have nothing to do with income though, so no discrimination. 

---

I just can't see getting a ding for a dollar amount of debt though. You were talking about what, $6100? I would consider that a trivial amount for a lot of consumers today and I just can't see them using dollars without salary information at that point. It would open them up to a discrimination lawsuit if it were ever discovered.

[Anonymous]

---

@jamie123 wrote:

---

---

I just can't see getting a ding for a dollar amount of debt though. You were talking about what, $6100? I would consider that a trivial amount for a lot of consumers today and I just can't see them using dollars without salary information at that point. It would open them up to a discrimination lawsuit if it were ever discovered.

---

I don't agree with you regarding the discrimination.

Dollars in some instances can be a better indicator of default risk outside of utilization percentage if you think about it.  Take these two otherwise identical profiles:

Cornelius has 5 credit cards, all with $50k limits for a total of $250k in revolving credit lines.  He's got $62,500 in revolving credit debt, or 25% overall utilization.

Rupert has 5 credit cards, all with $500 limits for a total of $2500 in revolving credit lines.  He's also got 25% overall utilization, but in dollars it's only $625.

Both of these individuals with otherwise equal profiles would possess the exact same FICO scores.  I don't think it's an absurd argument at all to suggest that if an analysis were done that if you took a large sample size of individuals with $62,500 in revolving debt and a large sample size of individuals with $625 in revolving debt that the first group would likely have more individuals that default on that more significant amount of debt.  Again, this has absolutely nothing to do with income levels and we're only talking about the probability of default on those two different amounts of debt.

[Anonymous]

Why are you always picking on Cornelius and Rupert?😉😁

I’ve hit 23% and took a ding myself.