optimum utilization for highest score

[drmceo]

So before reading the forums I would leave all my accounts at exactly 1% utiliaztion in order to have the highest score possible. But since reading the forums and reading it is best to have all my cards but 1 zeroed out and have a low balance on that one. I assume since that is the consensus as to the best way to keep a high score that it has been tested to verify that theory. I was wondering if within that testing that was done to get to that conclusion if it was also figured out what percentage yielded the highest score out of the 1-9% that you leave that 1 card balance at.

[Thomas_Thumb]

There are 12 scorecards for Fico and each scorecard is tweaked to weigh factors a little differently. Also,within each scorecards profiles don't react exactly the same relative to scoring implications associated with # of cards reporting balances or to perturbations in utilization.

Thus, there is no one size fits all regarding an answer to your question. General advise is best and avoids imposing undue constraints that may be non value add. As a general rule aggregate utilization should be below 9% and above 0% for best results. What translates to above 0%? - some say $2, some say $3, some say $5. Being conservative I'd go with a low end of $5 or 0.1% of CL, whichever is larger.

If optimization is your thing, test your own profile within the boundries of generally accepted guidelines to see what "optimizes" your score. Without a doubt the "window of optimization" for your profile will be narrower than some and broader than other profiles. Center points may differ as well.

[Anonymous]

Hi drmCEO.  You mention this rule:

All cards zero exceept one, and keep that one card at a smallish balance

I'll call that rule AZE1. 

The reason you hear that optimization rule a lot is because it is easy to state and because it is pretty much guaranteed to work no matter what your profile is.  With some profiles there's a lot of things that might result in the same maximally optimal result -- with others only AZE1 will optimize. 

For example, suppose Bob has 13 open cards and a total credit limit of $101,000.  He could easily have three cards showing a positive balance and a total amount owed of 5k and still have his CCs optimized for scoring.  Or he could have two cards.  Or one card.  And the balance could .also be a lot LOT smaller.  So he could use AZE1 or also 3 cards at a total of 5k.  (perhaps even higher, like 9k)  But the point is that AZE1 would also give Bob optimal results too.

Now consider Fred who has exactly two CCs with credit limits of $300 each.  For Fred, only AZE1 will work.  He can't have both of his cards reporting positive results, and while he has some range for his total utilization, it's not much.  (Needs to be under $54.)

So if a person uses AZE1 before a big credit pull, he's good to go -- regardless of the exact details of his profile.

As TT observes, the only hard evidence of a problem with way way WAY tiny balances is when a particular card has an absurdly small balance.  Like $1.50.  As TT observes, exactly where that cutoff is appears to be unclear.  $2.01?  $3.01?  And that weird behavior probably varies according to CC issuer.   But as a practical matter it can't be a big deal.  I haven't seen anyone claim that this has happened to them with a balance of $4, though there have been many reports of it at (say) forty cents.  So make sure you have a balance of $5 and you are almost certain to be fine.

[Anonymous]

Note that AZE1 has been pretty well tested for FICO.  But I believe that TT and a few others have found that the Vantage 3.0 model might require a but higher utilization for optimal scoring.

Thus, with Vantage, if you had a total credit limit of 10k and had a total debt of $90 that would be less than 1%.  I think TT found that there might be a slighly higher edge at something like 3% for Vantage.

You can ask TT if I am remembering him right.

[Thomas_Thumb]

---

@Anonymous wrote:

Hi drmCEO.  You mention this rule:

All cards zero exceept one, and keep that one card at a smallish balance

I'll call that rule AZE1. 

The reason you hear that optimization rule a lot is because it is easy to state and because it is pretty much guaranteed to work no matter what your profile is.  With some profiles there's a lot of things that might result in the same maximally optimal result -- with others only AZE1 will optimize. 

For example, suppose Bob has 13 open cards and a total credit limit of $101,000.  He could easily have three cards showing a positive balance and a total amount owed of 5k and still have his CCs optimized for scoring.  Or he could have two cards.  Or one card.  And the balance could .also be a lot LOT smaller.  So he could use AZE1 or also 3 cards at a total of 5k.  (perhaps even higher, like 9k)  But the point is that AZE1 would also give Bob optimal results too.

Now consider Fred who has exactly two CCs with credit limits of $300 each.  For Fred, only AZE1 will work.  He can't have both of his cards reporting positive results, and while he has some range for his total utilization, it's not much.  (Needs to be under $54.)

So if a person uses AZE1 before a big credit pull, he's good to go -- regardless of the exact details of his profile.

As TT observes, the only hard evidence of a problem with way way WAY tiny balances is when a particular card has an absurdly small balance.  Like $1.50.  As TT observes, exactly where that cutoff is appears to be unclear.  $2.01?  $3.01?  And that weird behavior probably varies according to CC issuer.   But as a practical matter it can't be a big deal.  I haven't seen anyone claim that this has happened to them with a balance of $4, though there have been many reports of it at (say) forty cents.  So make sure you have a balance of $5 and you are almost certain to be fine.

---

CGID

There are a few posters on this forum who reported higher (not the same) score with 2 or 3 cards reporting a small balance with Fico 08 as opposed to just one card. I have not seen any rigorous side by side data on that to rule out other influences. I would invite the OP to test 3 vs 2 vs 1 and report back since he has been accustom to reporting small balances on many cards. For true rigor aggregate balance across all cards reporting should be kept close to constant.

With regards to Fico mortgage scores (Fico 04 and Fico 98) I have not read of anyone realizing higher scores with more than one card reporting - but some have not experienced a score drop.

From a utilization % perspective, low is optimal for all Fico models - as long as it does not compute to 0%. Inverse tested Fico 08 when his scorecard was still dirty and gained 3 to 5 points dropping from about 8% Ag UT to between 4% and 5% Ag UT. No additional points gained dropping UT to the 1% - 2% range.

Regarding VS3, a 4% Ag UT gave me 1 or 2 more points (depending on CRA) relative to 1% or 6% in real world results. At the time, the CK simulator predicted similar results. Currently I am at 0.6% due to the AU card (Note VS3 does factor in AU card in utilization but not AMEX charge card). Here's what I get today using the CK VS3 simulator

Ag UT%	VS3 Score
0.6%	832
3.1%	829
3.7%	828
4.6%	833
6.5%	831
8.1%	829
10.1%	826