FICO Score Increase

[BMW_328i]

Hello, My TU score increased by 19 points with no reason or description. I do have a CC that hit the 1 year mark- have a been "rebucketed"? Thanks in advance!

[dragontears]

---

@BMW_328i wrote:
Hello, My TU score increased by 19 points with no reason or description. I do have a CC that hit the 1 year mark- have a been "rebucketed"? Thanks in advance!

---

Is that your youngest account? 

[Slabenstein]

If you're on a clean card and don't have any revolvers that are younger, it sounds like new revolver-->no new revolver scorecard reassignment, yeah.

[BMW_328i]

I have a dirty scorecard and it's currently the oldest- the other CC is two months.

[Slabenstein]

---

@BMW_328i wrote:
I have a dirty scorecard and it's currently the oldest- the other CC is two months.

---

Ah, so it wouldn't have been no-new-revolver in any case.  We don't know what segments the dirty scorecards, tho one theory is PR/no PR and recency of delinquency.  If you want to post a more complete picture of your profile before and after the score change we can help you try to narrow down what the cause might have been.

[Anonymous]

I've had my score jump that many points or more for no apparent reason. Maybe it could be another one of those random events with a dash of good vibes from your improved credit history. Congrats 🎉   

[Anonymous]

For the dirty cards, I'm pretty sure PR/no PR is the first level Segmenter from the testers of the old models via reason code resolution, plus obscure references in fico documentation.

However, with version 8, two new dirty scorecards were introduced that we have less information on and here's where theory comes in as @Slabenstein alluded to. 

I have theorized Version 8 introduced a recency segmenter, breaking each dirty scorecard into 2, a recent and mature segment of itself. Fico documentation also obscurely references recency, but that is far from concrete. However, datapoints do support my theory that recency segments delinquency scorecards at 2 years.

Unfortunately, we are still collecting datapoints to support my theory as far as the proposed PR card split goes, but fico mentioned recency quite prominently in it's documentation and it seems to make sense, since it appears the theory works out for the delinquency card.

Would be nice to find a Segmentation Threshold for reassignment for the PR card though. Time will tell. 

@BMW_328i  as for your change, when was your most recent Delinquency? Do you have any public records? if so when was the most recent?

[Anonymous]

@BMW_328i  another possibility is Age of oldest Revolving Account which is a Scoring Factor. You said thats your oldest, but you said you were in a dirty card, so I don't know if you have any that are older that are maybe closed?

[BMW_328i]

My last derog was two months ago on an installment loan. It's current now. Never had PRs. I also have a collection.

---

@Anonymous wrote:

For the dirty cards, I'm pretty sure PR/no PR is the first level Segmenter from the testers of the old models via reason code resolution, plus obscure references in fico documentation.

However, with version 8, two new dirty scorecards were introduced that we have less information on and here's where theory comes in as @Slabenstein alluded to. 

I have theorized Version 8 introduced a recency segmenter, breaking each dirty scorecard into 2, a recent and mature segment of itself. Fico documentation also obscurely references recency, but that is far from concrete. However, datapoints do support my theory that recency segments delinquency scorecards at 2 years.

Unfortunately, we are still collecting datapoints to support my theory as far as the proposed PR card split goes, but fico mentioned recency quite prominently in it's documentation and it seems to make sense, since it appears the theory works out for the delinquency card.

Would be nice to find a Segmentation Threshold for reassignment for the PR card though. Time will tell. 

@BMW_328i  as for your change, when was your most recent Delinquency? Do you have any public records? if so when was the most recent?

---

[BMW_328i]

TU also gave me like 13 points or so when it first reported current. IDK. Looks like I have to be a Harvard grad to understand these algorithms.