@Thomas_Thumb wrote:The credit report summaries do specifically state in their positive/negative factor summaries that mix of credit is a factor in scoring (revolving credit/installment loans/ charge cards). Having an installment loan either through a mortgage, car loan or consolidation certainly helps enhance the mix factor.
Aggregate credit card utilization rate is also listed as an important factor in the positive/negative summaries. Transferring debt to an installment loan reduces revolving credit card debt and therefore lowers utilization ratio. No surprise that this strategy can boost a score.
Having balances on too many credit cards is listed as a potential negative (Fico 4 model) and this supports consolidating balances onto fewer cards. So again, not a bad strategy.
However, the necessity of having all but on or two cards report a zero balance every month is an extrapolation of data, particularly for the Fico 8 model when balances are paid in full every month.
Necessity, no; however, when we've seen over and over again that files sub 850 get point boosts for doing so, why not maximize your score for a given application? Also to be fair, your comments were an extrapolation of what? Anecdotal data is the best we have.
It's not like we're talking isolated events, I haven't seen a single report different in the several years I've been looking; a difference may lie in how many revolving trades you have: the more you have, the less any one individual card with a balance on it hurts... step function similar to how inquiries appear to work for many people.
May also change with what bucket someone sits in, my score is hard to move as I'm in a dirty bucket, clean buckets (JagerBomb's pre-foreclosure fun as an example or yours as another) may get hit harder with comparitively small issues like numbers of revolving tradelines with balance.
What I was getting at is that it's nearly impossible to make blithe assertions about the model with any sort of accuracy as it is a complicated beastie, and the talking heads who do actually know, if they stated it, would be sued into the stone age; most, frankly, just don't know. As such there's very little concrete data that's not sourced from here or 1-2 other places in my experience.
It's not an exclusionary place, anyone who wants to contribute data to support their assertions come on down; however, it needs more rigor than a simple statement when it flies in the face of all accepted data previously. That, much like anywhere (Reddit, CNN, others) else, will draw fire for being sloppy.
@Thomas_Thumb wrote:I appreciate your last response.
Here is what I am looking at when I draw my conclusions regarding Fico 8 vs Fico 4:
1) I have 16 consecutive data points (months) showing Fico 8 scores of 850 - from 2/2014 to present (no Fico 8 from before that time)
2) I have Fico 4 scores from all three CRAs (EQ, TU and EX) during the same timeframe - these scores have varied over time
* My EQ score has always been lower than my TU and EX Fico 4 scores by 20 or more points during this time.
* My EQ 4 score dropped from over 800 to 765 and has since rebounded to 796.(EQ reports one hard inquiry from BB 6/2014, the others do not)
* The EQ 4 summary lists a negative factor on the MyFico 3B (3/2015) report: "too may credit cards showing a balance'.
* The TU 4, EX 98, and EX 4 scores were 823, 839 and 830, respectively (same 3/2015 report). None of these show the inquiry and none list a negative factor
3) None of the Fico 8 scores from EQ (classic or industry specific) list anything as a negative factor 3/3015 report. All of the Fico 4 scores fro EQ mention "too many credit cards showing a balance" as a negative factor
4) During the above timeframe I used whatever credit card I felt was most appropriate for the situation. Until the middle of last year I had no concern with how much I charged on a specific card as long as I kept it under the limit. My only concern was maintaining aggregate utilization on all cards combined under 10% at all times. In early 2014 I had posted 55% utilization on a store card statement that was subsequently paid in full. Next month it posted a zero balance. The month after that I again made "a large" purchase which resulted in a 80% utilization - again paid in full. Typically 3 or 4 of my cards have post a non zero balance in a given month. Since mid 2014 I have maintained individual card utilizations under 20%; under 6% in aggregate. Results on Foci scores during this time as follows:
* Fico 8 score remained at 850 every month - as reported through Discover card
* EQ Fico 4 score dropped from above 800 to 765 and then increase in steps to 777 and then 796 (EX and TU scores have been higher) - note I spot check Fico 4 scores - not monitored monthly.
5) I have read VantageScore 2.0 looks at balance vs payment and the TU credit report show this granularity in its credit report summary for each card. When my EQ Fico score dropped to 765, my VantageScore 2.0 was reported at 990 (range 501 to 990). I checked it again last week, still 990..
6) Monthly mortgage installment loan under 50% balance remaining
The above data is homogeneous and can thus be blocked in various ways to draw some profile dependent conclusions:
1) Fico 8 score model appears to look at monthly balance/payment ratio as a primary consideration (no impact on score when paid regardless of # cards reporting a balance and individual card utilization - as long as aggregate utilization under 5%)
2) Fico 4 score model is impacted by # cards reporting a balance and individual card utilization regardless of aggregate utilization or balances being paid in full. This is an older model and it likely does not have the ability to evaluate balance vs payment ratio (e.g. paid in full except by seeing a zero balance)
3) A single hard credit inquiry impacts Fico 4 model score but may not impact Fico 8 model score.
The above conclusions are within the boundries of the data set evaluated. They may not be applicable to profiles without a good credit mix (installment loan) and are predicated on payments in full very month. It certainly appears that the Fico 8 model does look at monthly balance to payment ratio and if so, it would be applied across all profiles although impact likely varies.
Well if you're game for doing some testing 
.
Amounts on reported tradelines matter for one portion of the revolving utilization calculation, and the number of revolving trades reporting a balance is a seperate calculation. There's also the aggregate revolving utilization which virtually everyone is familar with. FWIW I had a change from $3 to 73% balance report on a tradeline and my score didn't move at all, so I'm not surprised at your own findings on that. Others have suggested differently and my plan is to go get a silly $250 limit card and see if I can try to pin down if there are individual card utilization lines while keeping everything else fixed.
Amount of payment has been demonstrated not to factor into FICO; VantageScore does use a bunch of things that FICO doesn't, can't really pull any data from that other than to state they are different. All of the revolving utilization scorecard is instant-in-time as far as FICO goes. I've seen precisely one report of being bucketed where there was a 4 month lag between payment and score increase, but that's incredibly rare if it really existed at all. Everyone else's, whether they pay over time or pay lump sum, they get to the same point anecdotally.
Anyway for you if you're game: since it sounds like you have the 3B monitoring, just let $2 (reportedly some lenders zero out balances of $1, not tried that specifically but our purposes $1 vs $2 is irrelevant) report on every single revolving tradeline (cards and LOC's) on your report next month and watch your score change - you can fix it the very next month, but you'll get slapped likely to the tune of ~80-100 points on a FICO 8 score just for having a balance on all of your cards. I promise you it's a real thing in the model, and it would be interesting to see (Since we're starting to collect data points on it) how much you gain and lose as the cards report.
Closest I've seen to an 850 trying that was a ~798 at the time IIRC, went down to 720ish just from that experiment before fixing it next month for the expected complete recovery.
Fair enough, but if you let $2 report on all cards, you'll see the full effects within 2 months based on Discover, and probably for the purposes of this discussion depending when they cut the score you may well likely see a difference at 1 month. I don't think it'll take more than half of your cards reporting a balance to temporarily knock you off the 850 number, and I assure you all of our testing has demonstrated all cards reporting even a trivial balance is a negative.
Trivial aggregate, non-trivial number of balances calculation to demonstrate it does matter in the algorithm.
@Thomas_Thumb wrote:I actually don't participate in monthly credit monitoring. The only regular monthly score I get is the Fico 8 from TU on my monthly Discover card statement.
I purchased the one time MyFico 3B report for $59 in late March and plan to purchase it again in August. I was surprised to see 19 different scores. It makes for an interesting, robust, comparison since the scores come from the same data sets.
I do obtain true Fico scores with associated credit ireports from each CRAs at periodic time interval. These scores are point in time snap shots.
Short term, my plan is to let the one inquiry age off the EQ report, reduce spending to two cards and then see how all 19 scores are affected - assuming the report format does not change between now and late August.
After that I likely will go back to reporting a small balance on 4 cards - with payment in full as always. This will allow me to look at the impact of: one inquiry with four cards reporting against no inquiry with 2 cards reporting and then against no inquiry with four cards reporting. This isolates factors.
The expected outcome is no shift in Fico 8 score - maybe a couple points in an industry specific one (likely the EQ one), but a significant shift in Fico 4 (particularly EQ).
I look at month to month results as they relate to monthly summaries for installment loans and credit card statements. My credit card statement amounts match my credit report summaries as I pay each in full after I receive it. I do not make make partial or multiple payments per month..Keeps thinks simple that way. Makes for straiight forward analysis as well - no noise factors just the big picture.
I am unaware if any of my scores fluctuate daily and really don't need or want to know. So the best I can offer is periodic snapshots.
Congratulations on the 850. Thanks for the details behind your experiences the last year.
The common mantra here is "only let one card report a balance" and that seems to be based on the score boost commonly seen from making that change. As you note, however, a stable file with more cards reporting can also be at 850, and that is one of my main arguments against the "one card" advice: If the file is stable, it will be at a certain score. If the file is without negatives, and has a long history, it will be "optimized" with lower utilization. When the file has negatives, people look for ways to window dress the file by a few points. I also think "one card" is not good advice for building actual credit availability. Too little actual experience with showing small borrowings and payments can make lenders jumpy when suddenly balances appear.
Your 50% Mortgage balance is probably a major satabilizing factor in your file; with that much debt on a lowered utilization, it is a huge positive.
However, I would suggest you sign up for the monthly score monitoring here, to give your arguments more weight. In the debates here, the data source needs to be consistent. Even the CRA FICO scores are suspect.
Good luck!
