@cem13 wrote:
@Revelate wrote:What I would personally be looking for is comparing the good score report with the lower score report:
1) Are the loan types reporting the same?
2) Are the original balances reporting the same?
3) Anything different in the responsibilty or term lines?
That sort of thing though I think that covers the majority of the potential differences.
It's sorta unlikely but just trying to rule out possibilities honestly, rebucketing is really really hard to positively prove and your situation is way way outside normal at least to my understanding of the algorithm
. Which is admittedly limited.
I reviewed my 3B report. Here are a few things that are different over the past few months.
1) CLI on SGV $5k -> $5,800 via Luv Button (balance $0)
2) CLI on Menards (Cap1) $5k -> $7,800 via Luv Button (balance $0)
3) I removed myself as AU from my wife's Barclay's Card in NOV15 but it reported in JAN16. It was opened APR12 so there was no hit to my AAOA. Her CL is only $5K so it does affect my Util %. It shows "terminated" on the report.
I did not see any changes to my installment accounts. They are updating monthly as I pay them down.
Any thoughts?
I'm kinda at a loss to add any more than I have or what TT suggested in going back to basics and ruling out maybe something obvious that got overlooked (though I doubt that's the case here, you're more methodical than most).
I know it's a cop out answer but I think waiting to see if there's a magic swing back in call it 4 months is appropriate and see if it was some other latent bucketing, but if you stay at the lower score range with your utilization relatively minimal and fixed, I think you may well be correct that you were rebucketed on the BK. It's certainly plausible anyway given we're a little in the dark on the negative score cards.
It is an interesting datapoint to be sure, I don't recall seeing similar but given the number of negative scorecards, age may well factor into that derogatory wise. I didn't see that with my tax liens though but EX/TU wouldn't have likely exhibited that as I had an ancient tax lien that fell out of SOL which was likely a bucket if there were, but I did not see that on EQ unless tax liens are on different timelines (which to be honest, they are in the rest of the algorithm) for something like bucketing and I was already over the breakpoint upon joining the forum. It's been a slow progression but the trend has always been upward, other than my getting whacked for my installment utilization bouncing around.
@cem13 wrote:@ Revelate
Here is one more thought on my file. In DEC15, I applied for Amazon store card and made a purchase. It reported within a week. Then in JAN16, it reported on the normal cycle (within 3 weeks of previous report). So Amazon reported twice for the same account within one billing cycle. The second report caught me off guard and I had $1600 / $5000 =32% UTIL. Then I PIF and in FEB16 & MAR16 it reported $0 balance.
So is it possible that the new account reporting twice, once at 20% util then the second at 32% util; would cause the drop? The issue is that the score should have rebound once the PIF hit in FEB16.
Is it possible that the new account did not get scored until JAN16 as opposed to the initial report of DEC15? In other words, was the initial reporting not scored or viewed differently than the normal monthly scoring?
I am thinking of canceling the card I like it but I dont NEED it. Would closing the account help?
If a tradeline gets reported in a halfway house state (sometimes you get an early report without the full details in it) I'm pretty sure FICO excludes it; making decisions based on obviously broken information would just be patently stupid so I'm fairly confident in that especially as all sorts of random trash likely winds up in the database over time so there'd have to be some sanity checking / error handling when looking at tradelines. It's possible it wasn't scored to 1/16 as a result but you'd likely have to go look at historical reports to see if there was a difference. If I were good about weekly refreshing my CK data could probably do that on my own file as an example, or CCT or if you got lucky FICO Essentials or what not.
If it only reported the single tradeline and it was reported accurately, as many reports in a month as get made doesn't matter: I routinely have Chase report twice a month now and never any flakiness there; revolving utilization has no memory anyway.
I don't know whether closing the account would help, we've never gotten good store card data. Once I get clear of the AAOA 2->3 year breakpoint which is probably around May or June (TU, one month behind as missing a tradeline heh) for me, I'm going to close my Walmart card which is my only store card and see if I can get a solid datapoint, up, down, or indifferent as it comes to retail tradelines. I'm guessing I'll be flat but hard to say, but it's easy enough to test just in process of cleaning up my revolving utilization now anyway so just have to maintain it for 3 months, not hard to get the quality datapoint currently is my thinking since I need to stay clean from here for possible school financing come September and I'm trying to catch the AAOA transition anyway.
@cem13 wrote:
@Revelate wrote:If a tradeline gets reported in a halfway house state (sometimes you get an early report without the full details in it) I'm pretty sure FICO excludes it; making decisions based on obviously broken information would just be patently stupid so I'm fairly confident in that especially as all sorts of random trash likely winds up in the database over time so there'd have to be some sanity checking / error handling when looking at tradelines. It's possible it wasn't scored to 1/16 as a result but you'd likely have to go look at historical reports to see if there was a difference. If I were good about weekly refreshing my CK data could probably do that on my own file as an example, or CCT or if you got lucky FICO Essentials or what not.
If it only reported the single tradeline and it was reported accurately, as many reports in a month as get made doesn't matter: I routinely have Chase report twice a month now and never any flakiness there; revolving utilization has no memory anyway.
I don't know whether closing the account would help, we've never gotten good store card data. Once I get clear of the AAOA 2->3 year breakpoint which is probably around May or June (TU, one month behind as missing a tradeline heh) for me, I'm going to close my Walmart card which is my only store card and see if I can get a solid datapoint, up, down, or indifferent as it comes to retail tradelines. I'm guessing I'll be flat but hard to say, but it's easy enough to test just in process of cleaning up my revolving utilization now anyway so just have to maintain it for 3 months, not hard to get the quality datapoint currently is my thinking since I need to stay clean from here for possible school financing come September and I'm trying to catch the AAOA transition anyway.
Thanks for the quick reply. It is greatly appreciated. I am now suspecting you are correct. The first report of Amazon in DEC15 did not get scored fully. Then in JAN16 when it reported, I took a big hit (34 pts). I PIF in FEB16 but the score did not change.
So there must be a trigger for the amazon store card. My AAOA is 5 years+ with the hit so that is not the cause. Maybe it is because a store card is only a few months old? I do not need the Amazon card anyway so I will just close it and see what happens.
My summer home mortgage company uses TU 8 only. That happens to be my losest at 679 and 680 is the minimum. My finances are great with a DTI under 20% so the LO is encouraging me to apply and "lets see your entire file". All I need is ONE-FRIGGING-POINT to be in the clear.
Also who uses FICO 8 scoring for a mortgage anyway? Then to boot, they only pull TU. So I do not get the advantage of taking the median score. Oh well, I guess I learned a little more about FICO scoring.
Well if you decide to close the account I'd certainly be interested in the data point.
There's two possibilities neither of which have been rigorously confirmed:
- Store cards are counted differently, this was the case under FICO NextGen but I haven't been able to find evidence of it for FICO 04 and later... that doesn't mean it doesn't count, it's just not in the literature or reason codes that I could find when I went looking. There was one post from a former FICO employee (Barry) who suggested there was indeed different qualities of tradelines, I suppose similar to the Michelin star ratings.
- Tradeline seasoning boundary at 6 months (or some greater time period potentially now); this information came from off the forum from a buddy of mine who'd worked in close proximity with someone implementing the algorithm 20+ years ago. Quality of data therefore horrifically suspect even if I know the exchange did take place, it was a while ago and the algorithm has changed many times over the past 20 years.
I've never seen a drop for the second theory, every drop I've taken as a result of new tradelines reporting has been explicitly tied to AAOA (FICO 04 and 8) or installment ratios (FICO 8), but I've never been clean for 2+ years of my life either from a new accounts perspective which has been pontificated by some of the people chasing 850, though I have been for more than 1 year so the boundaries may just be different.
Holy something snacks re: mortgage company! Which one is it? A portfolio lender can do whatever they want since they're keeping the loan on the books, but we've only had one report of anyone not using the FICO 04/98 industry standard... ever as near as I can tell on the mortgage forum, and even then they pulled all three FICO 8's and took the mid score just like the traditional model.
Admittedly you could put a blindfold on after drinking 2 fifths of Scotch and still be able to hit a mortgage lender using the standard trimerge with a dart, so if TU 8 is to your disadvantage you could find a different lender.
@cem13 wrote:I closed the account. It reports on the 14th so I will know Tuesday or Wednesday. I will keep you informed.
I did call a different Mortgage company and they use the traditional Mortgage scores and use the median score. This is good because my median mortgage score is 727.
Right on, if you would PM me the name of the portfolio lender, interested in looking them up even if I'd be an idiot to use them personally with my TU file heh.