This past month for the second time I saw DWs scores change 15-20 points when one of my cards reported a balance then $0 reported.
The weird thing is that if the AU card has a balance the score is 15-20 points higher than when the AU reports $0.
This is what we are going to try this month:
- let a balance report
- pull her EX FICO score
- pay the balance to $0 that day
- wait a week
- pull her EX FICO again
All cards that were $0 last month will report $0. All cards with balance will report less balance (BT or 0% promo). Auto loan will report 1 more payment.
What I expect:
- increase of 15-20 points when the AU reports a balance
- decrease of 15-20 points when the AU reports $0
This is a Chase card, so the change will be reported in less than a week. We plan to do this in the next 2 weeks. Paying down the cards that will close in the next few days.
Maybe I missed something and other thing is at play.
Some notes:
- I don't see this happening with other cards (maybe because other cards will not report again when the balance is $0)
- getting scores more than once a month let us find this change
- she has a good score and not apping for anything
- this is something you can do when you are in the garden 
Will this test work? Anything I need to change?
The test seems legit at a second read, but the theory that the AU is counting and yet counting differently than a regular revolving tradeline is frankly a little surprising.
Are all other balances going to be held fixed (or close to) during this test? Would be interested to know how many revolvers and ones with balances though it still doesn't make much sense even if EX in particular has always been wonky on that compared to both EQ and TU historically.
Also seems reasonable heh.
I'd probably hedge bets and pull on the 13th and then pay on option a, as at least Amex has skipped a month for me and I hate leaving things to chance; not sure there's really any data improvement for this test by taking option b but has some additional risk of a failed test.
What are the other cards reporting balances? Another poster in a different thread suggested EX penalized him with just store cards reporting for example which might explain the variation here too... really struggling to come up with better rational other than there's something wrong with your other tradelines (unclosed dispute?) as AU's shouldn't behave like that.
@newhis wrote:9 open cards for her plus 4 AUs on EX. 1 auto loan 67% left.
7 cards at $0 including the only 1 store card she has.
So 6 cards with balance:
- less than 1% BBR
- 1% AU
- 11% BT will be $0 in August
- 12% (was 9%, I guess that is why she lost an extra 4 points between 4/19 and 4/26) PIF each month after statement date
- 14% 0% promo
- 38% BT was 41% last month
No disputes.
In a few days the AU will be $0 and everything else will stay the same. I'll report back if the score drops to 754 or stay the same.
IF you would, test if which AU it does matters, and also what's the aggregate utilization with and without the AU? Wonder if at $0 it's being excluded from revolving utilization metrics? Pure SWAG based on what you've reported TBH.