I now have SW history on 3 good sized changes to my Amex Plat card correlated with score changes. The last change was an increase from 1k to 6k and FICO 8 dropped 5 points from 778 to 773. In comparing the sensitivity of Amex to my Chase Visa changes in the Visa balance have about twice the impact of Amex charge card balances. It could be a result of the difference between the high balance (60k) on Amex v CL of 17k on the Visa.
I am now convinced that charge card balances affect FICO 8 scores unlike FICO 4. 3 FICO score changes tracking balances changes with no other card balance change is fairly compelling. In almost 2 years of FICO 4 history there was never any FICO change from Amex balances.
I'm not convinced based on my own observations but I live in a dirty bucket and as such it may not track.
Talking Amex NPSL specifically: I do know my TU score based on Wally didn't move when I set a new high balance on my Zync of 11.3K.
I had a standard revolving tradeline change from 0 to 42% balance where I lost a point for number of tradelines reporting a balance, then to 66.4% balance, and SW didn't blink. Suggesting that a 1K to 6K balance on an NPSL tradeline which has typically been calculated on high balance, resulted in a non-trivial change in your score when we're talking 10% utilization on that tradeline and by your description, far less overall, seems like a stretch honestly.
I know that every small little blemish for clean sheets is horribly magnified whereas I can abuse my reports by leaving balances which most forumers would consider absurd and my score doesn't shift much if at all, but I can't help but think there isn't something going on here as it really doesn't make much sense when I compare it against my own data.
Until there's a lot of other people that can contribute on this analysis I can't really put much credence in it: that's not to say you're wrong, just one anecdotal report with limited data given how many vagaries exist in the algorithm doesn't a fact make, and people haven't been up in arms over this before with Wally and other FICO 8 monitoring for the past several years regarding Amex.
I haven't seen any impact on TU FICO 8 score either but the granularity isn't what EQ SW provides. Still, I've had balances ranging from 1k to 20k over the last year or so and haven't seen any evidence of TU FICO 8 impact.
I wonder if it is specific to EQ? Could explain the lack of data since the changeover is relatively recent.
As for revolving tradelines. I did a test on my low CL Wally card and let a 90% balance report. It made no discernable difference on SW either probably because it was so low a CL compared to the overall UTIL and CLs.
Both TU and EX FICO 8 still appear to be unaffected by charge card balances.
By happenchance, I had just pulled EX FICO 8 a week ago before Amex reported and, since only Amex has reported so far this month, I pulled another ($20 blown for science) There was exactly no change with Amex reporting from 1k->6k.
So it appears to be something I'm only seeing on EQ SW's FICO 8. Even so, the FICO score changes I see on SW are only half the size of similar changes from ups and downs on on my VISA and MC though, of course, that may be due to the high limit reporting.
So if it is something that only affects EQ FICO 8 and at a lower level then that might explain what's happening. We need more people chiming in that run large balance changes on a charge card and are using SW. Perhaps it is something weird in my file Amex entry though I don't see anything in the full reports..
That would be seriously wierd if it were EQ specific on Amex charge card balances affecting score: while the algorithms *are* different bureau to bureau (I've never truly understood that personally but I guess it's a competitive advantage thing), usually the core features of the algorithm (like ignoring collections under $100 in FICO 8) are consistent across all 3. Infinite universe and all, but I would assume this one is a core feature too.
To be fair there's a lot we don't know about FICO, and it's hard to test given the sophistication of the algorithm as my results won't be 1:1 similar with yours in all things given the nature of our respective files, but EQ changing the factor which excludes revolving tradeslines with a Term = 1 month, seems awkward at least.
Well speaking of awkard, I only see that Terms = 1 month on EX, both my EQ and TU reports have it listed as an "Open Account" rather than "Revolving Account" which I have on the rest of my credit cards.
If you have a EQ report laying around, wouldn't hurt to check to see if it is listed similarly. As they say, the devil is in the details: why can't the CRA's get on the same page with elements which are reported to all 3 bureaus? Sigh.
@AzaleaB wrote:
@Revelate wrote:That would be seriously wierd if it were EQ specific on Amex charge card balances affecting score: while the algorithms *are* different bureau to bureau (I've never truly understood that personally but I guess it's a competitive advantage thing), usually the core features of the algorithm (like ignoring collections under $100 in FICO 8) are consistent across all 3. Infinite universe and all, but I would assume this one is a core feature too.
To be fair there's a lot we don't know about FICO, and it's hard to test given the sophistication of the algorithm as my results won't be 1:1 similar with yours in all things given the nature of our respective files, but EQ changing the factor which excludes revolving tradeslines with a Term = 1 month, seems awkward at least.
Well speaking of awkard, I only see that Terms = 1 month on EX, both my EQ and TU reports have it listed as an "Open Account" rather than "Revolving Account" which I have on the rest of my credit cards.
If you have a EQ report laying around, wouldn't hurt to check to see if it is listed similarly. As they say, the devil is in the details: why can't the CRA's get on the same page with elements which are reported to all 3 bureaus? Sigh.
I just checked my credit reports and my Amex Green is reported the same way - on TU and EQ, it's an open account but on EX, it's a CC, Rev with Term of 1 Month.
I also just got an SW alert a couple of days ago with a 5pt point increase in my score (not updated in my sig) and the comment that my Amex Green balance had decreased to $0 - with no other changes. Still, it seems odd and I looked online and found a comment from someone from FICO stating that the high balance on an NPSL account is not factored into the Fico08 scores and that it was considered in Fico04 scores.
"According to Paperno, the newest versions of the FICO scoring model, including FICO 8, do not consider open accounts for utilization, which means that spending on those accounts won't affect this portion of your score." (Paterno is the Consumer Operations Manager at FICO.)
Read more: http://www.bankrate.com/finance/credit-cards/do-no-limit-cards-affect-score.aspx#ixzz3AUDyF0qM
Thank you for the research Azalea!
Pizzadude is correct, a charge card does count on the number of non-zero balances calculation, but it does not for the aggregate utilization calculation which is likely why you saw the increase if you stumbled over one of the step function boundaries on that. 5 sounds about right for that with your pretty report, I only get 2-3 down around the 700 mark on a dirty file.