Maximizing the final 25% of the pie to 850?

[Anonymous]

I pose this question to guys like TT, anyone with "perfect" 850 scores (now or in the past) as well as anyone else that has knowledge of maximizing the final 25% of your FICO score.  I say the final 25% because these final two categories I feel are a bit less concrete than the other 3 in terms of knowing the impact (point wise) of their changing data. 

The first 75% of the score to me is pretty simple:  Perfect payment history, 1% utilization, and a mix of accounts that includes say 3 revolvers and an installment loan paid down to < 9% utilization.  I'm assuming for this thread that those 3 slices of the FICO pie which make up 75% one's score combined are maximized.

So, the final 2 categories that make up the final 25% of your score are length of credit history and new credit.  These categories are mostly spoken about with respect to AAoA, AoOA, AoYA, "new" accounts and of course inquiries.

My question is how "good" do these final pieces need to be in order to achieve that perfect 850?  We also know that a buffer exists, meaning someone can have a score that's actually better/stronger than 850, but of course it's still represented as 850.  So can you be lacking in one or more of these smaller areas?  Can you have a few inquiries and still make it?  Can you have a few inquiries AND have an AAoA of < 7 years and 8 months?  Can you only be lacking in one of these categories?  I'd be very curious to hear some data points here from those that have gone down this path already or those that are in the know through previous research.  Thanks everyone!

[Anonymous]

A key limiting factor is what scorecard you are in.  FICO 8 uses three factors to assign clean profiles to one of eight scorecards:

(1)  Age of Oldest Account

(2)  Age of Youngest Account

(3)  Total number of accounts

In order to be able to score an 850, you have to be in one of the top scorecards (top two?  three?).  All other scorecards have a ceiling that prevents you from scoring higher than a certain number.

[Anonymous]

Right, so for this illustration I'm making the assumption that someone is in the top scorecard(s) making 850 attainable. 

I'm quite certain that I've read of someone with an 850 score adding a new account and not losing their 850, so if that AoYA drop to 0 and potential scorecard reassignment doesn't impact score (visibly, again, there's a buffer) that would be a data point of a piece of the puzzle not being "perfect" but a top score is still able to be reached.  I'm curious about this as well with inquiries, like how many scoreable inquiries one may have but still have an 850 score.

[Thomas_Thumb]

---

@Anonymous wrote:

Right, so for this illustration I'm making the assumption that someone is in the top scorecard(s) making 850 attainable. 

I'm quite certain that I've read of someone with an 850 score adding a new account and not losing their 850, so if that AoYA drop to 0 and potential scorecard reassignment doesn't impact score (visibly, again, there's a buffer) that would be a data point of a piece of the puzzle not being "perfect" but a top score is still able to be reached.  I'm curious about this as well with inquiries, like how many scoreable inquiries one may have but still have an 850 score.

---

My belief is at least two of the eight clean scorecards can reach 850.

As CGID mentioned the clean scorecard assignment factors are age of file (Age of oldest account - could be open or closed), file thickness (number of accounts - could be open+closed or just open - not sure on this but open is my bet) and age of youngest account. If the factors are given an H or L rating then I'd guestimate the following scorecards as having 850 potential:

1) H (oldest), H (thickness) and H (youngest) - large buffer

2) H (oldest), H (thickness) and L (youngest) - small buffer

* One must have an established file to get an 850. If a file is sufficiently thick and aged a new account would cause a change in scorecard but not necessarily drop score from 850.

What are the values for L and/or H? Who knows for sure. The amount of buffer potential likely depends on the scorecard.

a) For oldest account (file age) I'd put H at 10 years minimum although one could argue 12 or 15 years. However, if 850 is not achievable if age of file is not "H" per the above then the value for H must be less than or equal to the youngest file age posting an 850.

b) For thickness a thin file is routinely defined as three or less. Let's say H = 4 open accounts minimum although it could potentially be minimum 4 total with at least 3 open.

c) For youngest account I'd peg H = 2 years or greater.

Given the above definitions my profile's scorecard would rate H-H-H:

 - Oldest account age = 33 years (still open)

 - Youngest account age = 6 years (open)

 - Average age of Accounts = 17 years

 - Total Accounts = 9; with 2 caccounts being closed and 7 open accounts as follows:

   (4 credit cards, 1 AU credit card, 1 charge card and one 15 year mortgage opened 12 years ago) 

Note: A non H rating for youngest account can still produce an 850 score but scoring factors would need to be more highly optimized.

Is an open installment loan required (or even a closed one)? - Not according to Ubuntu's posts. If that is true then an H-H-H scorecard should have at least a 20 point buffer to offset lack of installment loan "bonus points"

An alternate sceme might be 3 of 8 scorecards able to reach 850 as follows:

1a) H (oldest), H (thickness) and H (youngest) - large buffer

2a) H (oldest), H (thickness) and L (youngest) - small buffer

3a) H (oldest), L (thickness) and H (youngest) - small buffer

In this case H for thickness would be substantially higher, say H = 12 accounts minimum. I have 9 total accounts (7 open) and even when I had 10 total accounts got reason statements that suggest lack of sufficient accounts is a factor holding down my Eq Fico 8 Bankcard score. Don't recall such statements back in 2015 when account total was 12 (2 mortgages & 10 cards). Thus, H = 12 or more speculation. In this scenerio you can reach 850 with an H-L-H scorecard but, file would need to be non thin (at least 4 accounts either all open or perhaps 3 open with 1 closed)

[Anonymous]

Thanks for your weigh-in on the subject, TT!

How would you quantify "small buffer" and "large buffer?"  Would 0-10 points be reasonable for "small" and 10-20 for "large" or do you think I'm off on that?

The buffer I assume would absorb the impact of scoreable inquiries, where maybe 1-2 wouldn't be felt with a small/large buffer in place, but perhaps 3-4 would.  Does that make sense?

[Thomas_Thumb]

Those are good ballparks ranges for buffers although I suspect the large buffer may go up to 25 points.

I have speculated that the buffer is not one big catch all but may be the sum of category specific buffers. As we know, categories are made up of multiple factors.  If there are category specific buffers then a small ding for a factor in a sub optimized category may drop score from 850 but a small ding in a factor for a maximized category may not.

Can category buffers exist for clean profiles below 850? Do category buffers exist for dirty scorecards?

[tacpoly]

Here are my stats

Oldest account age: 24 years
Average age of accounts: 10 years
Age of newest account: 3 months
Number of inquiries: 1
Number of accounts: 16 (open: 5cc, 1au, 1loc, 1loan)

[Anonymous]

Good points to consider above, TT.

tacpoly, what are those stats for?  Do you have an 850 score and that's your profile data that makes up that score?

[Thomas_Thumb]

---

@Anonymous wrote:

Good points to consider above, TT.

tacpoly, what are those stats for?  Do you have an 850 score and that's your profile data that makes up that score?

---

BBS - Tacpoly mentions having 850 scores in another thread.

Her data is in-line with what I have experienced regarding inquiries: One hard inquiry will not inhibit ability to achieve or maintain a top score.

I've had a couple HPs in the last three years (one for a CLI and the other unauthorized from PenFed) and neither impacted Fico 8 score.

Another poster "IV" had 850 scores a couple years back and then added three new accounts (one being a loan) which resulted in a score drop to the mid 820s in 2014. (shortcut to post below).However, IV may have crossed an AAoA threshold or opened a high B/L installment loan in the process. Of note, IV is back at 850 on two CRAs

http://ficoforums.myfico.com/t5/Understanding-FICO-Scoring/Utilization-Percentage-question/m-p/4073409#M95021

Others, in addition to Tacpoly mentioned having a "new" account (in one case I think it was two with corresponding HPs) while maintaining top scores.

For "maximizing" the last 10% of Fico 8 score (excluding utilization and accounts reporting balances) here are some thoughts:

1) AAoA 8 or more years for maximum potential

2) Hard inquiries 0 or 1 per CRA.

3) Age of file over 10 years, perhaps 15 for maximum potential

4) New accounts, less than 2 in the last 90 days. Fico does not define the timeframe for classifying an account as "new". It could be 1 year but, likely less.

5) Minimum 5 open accounts with one open being a loan.

[Anonymous]

Again, nice insight offered above.

TT, question for you regarding a "spree" which I would say is similar to what IV did.  However, in his/her case an installment loan was added (presumably at > 9% utilization) and it was also stated that an AAoA threshold may have been crossed.

Do you think that a spree of who knows, 3-5 accounts, within a year wouldn't adversely impact score?  Certainly the inquiries would become unscoreable and AoYA would reach 1+ year (no "new" accounts for sure).  This illustration above is assuming that an AAoA threshold isn't being crossed (8+ years both pre/post spree) and AoOA is 15+ years.  Do you think that difference between AoYA being 1 year verses 2 years would keep such a person from grabbing an 850?  I suppose in this illustration the "spree" could be even larger for someone with an extremely thick file where their AAoA wouldn't drop under 8 years with the new accounts.  What are your thoughts here?