EX 8 decrease of 20 points @ 36 months AoOA!

[SouthJamaica]

It's my understanding that both Age of Oldest Account and Age of Youngest Account can affect one's scorecard assignment.

I have found that when I have an upward scorecard reassignment due to improved AoYA, my EX FICO 8 scores increase; they do not drop.

My most recent AoYA reassignment gained 26 points, each of which I promptly lost -- point for point -- when I got myself reassigned downwards.

Needless to say, this is all guesswork, because FICO never tells me I've been reassigned

And I really don't know about AoOA: my oldest account is so old that the difference between one year and the next is meaningless in the greater scheme of things. Probably my scorecard would never change in that department.

[Thomas_Thumb]

Of course increasing AoYA above 12 months increases score. However, if someone has a low AoYA perhaps score is not penalized if they are new to credit. Then their profile hits the AoOA threshold and boom! They get slammed with a 20 point penalty for their 1 month old account.

With AoOA you are either above or below a scorecard reassignment cutoff. Most of us cross above the AoOA only once. From then on we have established credit.

My AoOA is 40 years and my AoYA is 10 years. Assuming the AoOA is 3 years I could not get there without all accounts falling off my file!

Would I experience the new account penalty if my AoOA were under 3 years and I opened a new account? Not sure as I could never test it! Only someone relatively new to credit could test. I'd need to look at their data.

Impact of some clean scorecard reassignments are a one shot deal. What happens scorewise likely depends on the level of scoring factors when the reassignment happens.

Not sure the OPs score changes are the result of a scorecard change. However, unexplained score shifts at 3 years were mentioned multiple times in Birdman's thread on credit. He and CassieCard concluded 3 years was the likely AoOA cutoff.

[SouthJamaica]

Very good point. Most of us experience the AoOA upward reassignments rarely.

In my case, I experienced them long before I was conscious of FICO scoring and all that that entails.

So if I read you right, you're saying that there might be something already existing in OP's profile which didn't cause a problem in the lower scorecards, but does cause a problem in the higher FICO 8 and 9 scorecards.

[Thomas_Thumb]

Yes, that is the hypothesis. I only used the new credit account as an example because the OP has one.

Scorecard reassignment could still be an innocent scapegoat and not cause. The only thing we can say is there is only one AoOA cutoff for Fico 8 on scorecard reassignment. If it is truely 3 years then it can't also be 4 years or 5 years.

[FICO-Quest]

Thanks - A good discussion that prompted a re-read of the Credit Scoring primer thread from few years ago, and a review of my detailed scores I've been tracking since Sept '21.  One thing abundantly clear is that I have not been seeing a new-account penalty imposed on my scores prior to crossing the 36 month AoOA threshold, only for any incremental inquiries (>2, >4, etc).  Now that I've crossed into the new scorecard (> 36 months AoOA) I have 2 revolvers opened 7-8 months ago, and 1 revolver opened 80 days ago (< 3 mo's) that the scoring algorithm might not like as much.  I know it's not particularly reliable, but FWIW the Fico Score Simulator predicts point gains of 5-10 points directly corresponding to the aging of those 3 revolvers at their respective 6 & 12 months of age milestones.

[SouthJamaica]

---

@Thomas_Thumb wrote:

Yes, that is the hypothesis. I only used the new credit account as an example because the OP has one.

Scorecard reassignment could still be an innocent scapegoat and not cause. The only thing we can say is there is only one AoOA cutoff for Fico 8 on scorecard reassignment. If it is truely 3 years then it can't also be 4 years or 5 years.

---

Just curious: how do we know there's only one?

[Thomas_Thumb]

Because the segmentation factors are binary. Either A or B, either 1 or 2, either on or off. Or put another way the scorecard segmentation is 2 level high/low differentiation.

Older Fico models have a different # of scorecards than newer models. The numerical values used to segment those scorecards are binary as well. However, since the model is different, it could assign a different value to a segmentation factor. Key point is that within a Fico model the value of a segmentation factor is set. For a binary H/L segmentation "there can be only one" value.

[SouthJamaica]

---

@Thomas_Thumb wrote:

Because the segmentation factors are binary. Either A or B, either 1 or 2, either on or off. Or put another way the scorecard segmentation is 2 level high/low differentiation.

Older Fico models have a different # of scorecards than newer models. The numerical values used to segment those scorecards are binary as well. However, since the model is different, it could assign a different value to a segmentation factor. Key point is that within a Fico model the value of a segmentation factor is set. For a binary H/L segmentation "there can be only one" value.

---

Thanks. Well @FICO-Quest we've both got an education from @Thomas_Thumb 

[BuckyB]

Keep in mind too that FICO scoring models are purposely "detuned" to keep them "palatable" to the government and society (see this paper: FICO scores are only 98% as accurate as they could be) so trying to make sense of every few points gained or lost is an exercise in futility.  Scores are not 100% based on logic and math and so data points will never give full insight into what's inside the FICO algorithm black box.

[SouthJamaica]

---

@BuckyB wrote:

Keep in mind too that FICO scoring models are purposely "detuned" to keep them "palatable" to the government and society (see this paper: FICO scores are only 98% as accurate as they could be) so trying to make sense of every few points gained or lost is an exercise in futility.  Scores are not 100% based on logic and math and so data points will never give full insight into what's inside the FICO algorithm black box.

---

IMHO trying to understand FICO scoring can be quite useful.