Since a bunch of people have been playing with this over time, wanted to make a consolidated thread for datapoints to hopefully pull it altogether. Since different people have different numbers of cards, their own data is likely going to be different. Likely is some ratio / percentage of revolving trades which causes the drop as people with smaller counts have reported changes going from 1->2, and others have reported no change 2->3 or higher and the evident difference is different numbers of revolving tradelines.
This doesn't have much real FICO value since from all testing smaller numbers of tradelines reporting a balance (non-zero) is better and as such the "only 1 card with balance" is a workable and simple optimization mantra, but it's a curiosity point for me and apparently others so why not?
This main post will be kept updated semi-regularly as reports come in; do need source including bureau, and FICO model and the obvious . This is really designed for people with a monitoring solution that reports small changes or otherwise close-in-time pulls to try to isolate it, some data might get skewed but this one isn't too hard to test on a fairly stable report. Also ideally this is held to minimal utilization numbers for isolation and consistency purposes. Format will be work in progress.
Reserved
Reserved
Sample from a recent data point, probably doesn't need to be this fully detailed but cutting and pasting:
Source: pair of MF 3 report views - FICO 8, 04, other models, all 3 bureaus.
Total revolving trades: 9
Total revolving trades with balances: 1 vs 2
Balances:
5/24: 1/9 cards with balance. $43/1000 individual balance, $43/34600 aggregate balance
5/29: 2/9 cards with balance. $43/1000 and $86/8400 individual balances, $129/34600 aggregate balance
No FICO change on any 3B report model except TU Bankcard 04. TU Bankcard 04 lost 5 points, everything else flat at +/-0.
Source data for the curious:
Older data points, different format (note, my spreadsheets are NOT necessary was just how I kept my data):
Source: MF Equifax Scorewatch // FICO 8
Didn't keep the balance information, but aggregate utilization was minimal and so was individual tradeline utilization. The -3 for 2/9 -> 3/9 tradelines reporting a balance has held throughout all of my testing on EQ FICO 8 with my report.
| Historical | Score | Bal Reported | Total TLs | Agg. Util % | Change | |
| Equifax Beacon 9 | 693 | 2 | 9 | 0.4 | ||
| Equifax Beacon 9 | 690 | 3 | 9 | 1.3 | -3 | |
| Equifax Beacon 9 | 690 | 4 | 9 | 1.5 | 0 | |
| Equifax Beacon 9 | 686 | 5 | 9 | 1.6 | -4 | |
| Equifax Beacon 9 | 703 | 5 | 9 | 1.6 | N/A | Tax lien came off |
| Equifax Beacon 9 | 707 | 4 | 9 | 1.2 | 4 |
Edit: Whoopsie.
Previously going from 2->3 out of 9 cards had a FICO drop associated with it.
Now standing at 14 revolving tradelines, busy report but no change going from 1->2->3 revolvers with balances so far.
-3 points for going from 3->4 with balances; same as an old change before.
3/14 = 21%
4/14 = 28-29%
2/9 = 22%
3/9 = 33%
Certainly looks like 25% is the breakpoint.
@Thomas_Thumb wrote:Potential mortgage effect?
Perhaps not given Inverse has 2 mortgages but his scores changed with # cards (out of 3) reporting while he had a tax lien. I am curious to see if that changes now that he has no lien and an 850.
P.S. Wonder if your Fico 08 Bankcard scores are changing.
No effect apparently on this calculation; I had a mistake in my data, CK had the BOFA card, Equifax MF monitoring did not.
So 2 -> 3 cards, no change.
3 -> 4 cards, -3 points: same as what 2->3 used to be; identical score movement both pre and post mortgage reporting.
Looks like 25% is the breakpoint for taking a hit for number of tradelines with balances based on my data, and that holds I believe with the rest of the data reported here. Maybe someone with 8 or 16 cards can test that more concretely.
@Thomas_Thumb wrote:
Perhaps not given Inverse has 2 mortgages but his scores changed with # cards (out of 3) reporting while he had a tax lien. I am curious to see if that changes now that he has no lien and an 850.
Maybe I will luck out and EQ and TU will drop the final lien before November 12th.
My cards report 12th, 16th, 20th so I can provide decent utilization testing.
With the changes at CCT, I can't get 3 3B pulls anymore, but I may have found a hack to deal with them 
If so, I will pull 1,2,3 cards reporting and keep the utilization low just to test the effect of additional cards.
If not, I will at least get EX FICO 8 data on number of cards reporting.
Hrm, confused myself somewhere, time for spreadsheet.