Scoring a perfect 300

[jamie123]

I hit 350 at one really low point in my life!

[Thomas_Thumb]

---

@Anonymous wrote:

Hey TT!  Thanks. 

Jello77's post "The Many Flavors of FICO" is invaluable and it it was one of the first things I ever read on the MF forum.  (As you can see it is a summary of a detailed discussion begun in 2012.  I even contributed to it, far down the thread and a while back.)

Thus, for example, J77 gives 334-818 as the real world range of FICO 04 as implemented by Equifax.  That doesn't mean that people could have scored higher than 818, but for some enigmatic reason the highest some guy ever scored was 818.   It means that, as Equifax implemented it in reality, the maximum score that is theoretically possible with that model is 818.

The Experian link supports this, because there is a gigantic jump from the lowest score theoretically possible (316) to the next number (465).  Contrast that with the top of the scale, where we see much more of a finely graded continuum as we move down.  We know that real people have really scored 850, and 849, and 848, etc.  In the case of the bottom up, we do not see a similar gradation, but rather a gigantic leap from the lowest the model could score to the just under 1%.

The Experian chart does answer one important question, which is that there are definitely people scoring less than 465.  (At the time that chart was created, about 1% of consumers were scoring less than 465.)  What it doesn't tells us is the lowest score ever recorded, which might be (say) 380, or even higher, like 430.

---

The high and low scores listed below the Experian Fico 8 table look like a statement of actual scores seen in the Experian database. Otherwise, there is no need to call out those scores separately. The 316 appears to be a real score from how I read it.

Also 818 has been achieved per the below posting:

Android01

05-20-2014 10:03 PM

I've had 850 on all three CRA's that use the FICO 8 version for over a year, and though I understand that anything above 760 is "bragging rights" as far as getting the best rates are concerned, I'm bragging!  

EQ FICO - EQ08 - 850 - 01/01/15

TU FICO - TU08 -  850 - 01/01/15

EX FICO - EX08 -  850 - 01/01/15

EX FICO - V2 -      840 - 01/01/15

EQ FICO - EQ04 - 818 - 01/01/15

[Anonymous]

Hey TT.  Well, it sounds like there are a couple ways to understand the ranges provided by Jello77 (which he is getting from various sources, e.g. the EX doc).

(1)  They are first and foremost reports of what has actually been achieved.  Secondarily some inferences from them are possible about what is possible in theory.

(2)  They are first and foremost descriptions of what is theoretically possible.  Secondarily some inferences from them are possible about what has actually been achieved.

It sounds like you understand them as #1.  I understand them as #2.  I also think this is how a number of other people see them.

Take the example of Jello77's range for EQ FICO 04 (334-818).  With perspective #1, it means that the highest anybody has ever managed to score is 818.  But perspective #1 leaves open the possibility that somebody could score higher next week, much like the idea of a sprinter setting a new record for the 100 meter dash.  The world record might be 9.58 seconds, but maybe someone else will come along next week and do it in 9.47 seconds.  Here one assumes that FICO's theoretical range as a given: it really is 300-850 and the"real world range" is just a description of what individuals have done.

With the perspective of #2, however, we see that as first and foremost a description of the model itself.  818 is the very highest it is possible to score.  Now additionally, if we add the observed fact that people do indeed get perfect credit scores, we can infer that people must be getting 818 -- since 818 is perfect in that model for that CRA.  But the range itself (in perspective #2) was fundamentally first a description of the model. 

So one difficulty that a proponent of perspective #1 would have as in understanding the ceiling of 818 for FICO 04 Equifax.  That's the reason I gave that to you initially.  Why is it nobody is ever able to break this sound barrier?  It becomes a kind of grand coincidence -- apparently none of us have good enough data at EQ.  But with #2, you can explain the range as a description of the model itself.  Furthermore, and this is key, once you see the 818 as a description of theoretical maximum then the other end is the theoretical minimum (not ncessarily an actual score someone achieved).

Another huge difficulty for #1, which I also mentioned, is looking at the actual data (such as we have it) as you get close to the low end of the spectrum.  Look carefully at the Experian chart we have been talking about (for FICO 8). 

http://www.experian.com/assets/consumer-information/brochures/risk-based-pricing-rule/experian-fico-...

You'll see that at the high end, the scores are finely gradated.  The people in the very top 1% are scoring in a tight cluster of three points: 848-850.  In the next lower percentile, people are clustering in another tight range of 7 points: 841-847.  And so on.  Each cluster is a very small range and much like the one before it.  Gradually the ranges for each cluster become larger as you move toward the middle scorers.  Then the clusters start to become smaller as you move toward the low end.  Then suddenly (!) there is a gigantic gap.  We go from a cluster of 10 points (465-475) for percentile 2 to an utter non cluster of 149 points (316-464) for percentile 1.

#1 can only view this as a bizarre fluke.  Why should that last percentile suddenly explode into such an enormous range?  I dunno, says #1.  But with perspective #2 it makes sense.  316 is not a description of what some specific person has scored.  It is the absolute lowest value that is theoretically possible.  The chart says that the lowest 1% of consumers score in that huge range: 316-464.  That will be true if a real person really scored 316, but would also true if the lowest score that has happened thus far is (say) 417. 

Here is an analogy that may help, back from when (most of us) were in high school.  We all remember taking tests.  And we can remember that the teacher would give us a score, which was the percent we got right.  Very often he'd group those as follows:

    A = 90-100

    B = 80-89

    C = 70-79

    D = 60-69

    F = 0-59

If someone asked him what his ranges were, he'd give them those numbers.  Note that each of the top four letter grades are reasonably tightly clustered.  A little range of ten points.  Suddenly the lowest category explodes in a huge range (0-59).  Does that mean that somebody actually answered every single question wrong (and therefore got 0% right)?  No.  The 0-59 describes what is theoretically possible.  It does not say that somebody actually did the absolute worst possible that one can imagine.  As a matter of fact, it's immensely unlikely that any of your classmates answered every question wrong, though it is quite possible that several of them got every question right.

Now one thing that the chart does tell us is that someone really did score 464 (or lower) -- much in the same way that, if the teacher says that somebody failed, we know that somebody must have gotten a 59 or lower.  But how much lower did the lowest real person actually score?  The chart (as given) can't tell us.

Hope that helps!  Best wishes...

[Whitebird]

Interesting thread.

I will say I know someone who has/had a FICO of 330. It was awhile ago. Don't know which version.

I'd like to comment on the older FICO 04 score range. I did check my Equifax occasionally. The highest I got was 817, with a usual range of 814-817. Yes, I also heard that 818 was the highest attainable.

For the FICO 08 version, my scores are no higher than 820-822. Difference: When the FICO 04 version was predominantly used, I had a mortgage reporting. That has since been paid off and it dropped off my report altogether (not showing in closed accounts) early by mistake (too involved to discuss). My current accounts are all CC. I have never had an auto loan. 

Now, my FICO reason codes hit me with "No installment accounts; no mortgage accounts; no auto accounts.  I have heard it said that installment, mortgage, auto accounts do not count for much with FICO, but I disagree. I believe it totally depends on your scorecard. For those in the 600s or so I feel they would do better to clean up their reports, lower utilization, all the usual advice, etc. For those people, installments indeed may not count for much until they reach the higher scorecards.  In the very high FICO scores, one is "punished" much more harshly for various FICO transgressions--in my case not having enough of a credit "mix."

[taxi818]

I actually think it is impossible. A good friend of mine asked me to look at his credit report so he could build. When he pulled the report i was in Total Disbelief. As we got his fico scores as well.

he had 2 Repos. 1 forclusure. and 36 collections. 3 charge offs. and well as over 200 lates. i thought for sure he would be in the 300s. When i actually looked at his score i was in shock. He was right around 451. so after all i mentioned if anyone needed a 300 it would be this guy. He has never paid a bill on time. also he had and eviction from his apartment. hehehe. But somehow. he was at 451 fico. unreal. so it is not possible. not matter how hard you try.

[Whitebird]

Sorry, you are forgetting bankruptcies and judgments. How late? Months? Years? How much owed? How long was the AAoA? Too many variables here. Looking at one report you believe is "really bad" is not exactly a sufficient sample. How many really, really bad reports have you seen?  I confess I have only seen one, the one report I mentioned, and yes it was a 330. I saw this on paper but I don't recall the CRA.  Worse than you have described. No car repos though. CCs, bank lines of credit, had foreclosures (2), judgments. Tax liens.

She had thus far worked the system. What's a bit different here is that she had a relatively high income and low expenses. I personally believe she had/has a credit addiction. Bought all kinds of things she never used.  The person in question was in hysterics because she had thus far actually kept her (new) husband from knowing the situation and she wanted to know how to face him as he now knew. I was dumbfounded when she called me. She was in tears; I had no idea. She had declared BK 5 years previously, so a new one was out of the question. I will say she pulled herself together and paid off some of her debt, but mostly she cut up all cards and to date has never accepted another one. The marriage did not survive.  I'm assuming much of her bad history has not dropped off, but I don't know.  I'm no longer in touch with this person. 

[cashnocredit]

The CFPB looked at 200,000 scores from CRAs including FICOs, VantageScore, and "Educational Scores."

Getting a FICO score at or below 480 will make you part of the "one percenters."   Below 440 is so infrequent it doesn't show up on the charts and likely less than 1 in 500 people can claim that rarified score territory..

See page 9

http://files.consumerfinance.gov/f/201209_Analysis_Differences_Consumer_Credit.pdf

It's a really interesting study about credit score variations. Most of the analysis comparing different scores is after normalizing them so the different ranges and distributions don't affect the analysis.

As much as I like experimenting, trying to hit bottom is one I don't plan on doing.

[jamie123]

I was just recently divorced...perhaps 3 months. I thought that it was an equitable divorce but little did I know...

My daughter needed a clarinet for music class. The school sent home papers saying that we could rent one at a local music store. I went down to the local music store to rent the clarinet. The salesperson asked for my Driver's license and said she just needed to run my credit and walked into the back room. About 5 minutes later the owner of the store walked out and said, "I just had to come out and see what you looked like for myself. I've never seen someone with a credit score as low as 350 before!" He was surprised because standing before him was a middle aged guy in khakis and a white buttoned down shirt! Needless to say, he refused to rent me the clarinet so I had to buy it. One of the most embarrassing moments of my life.

I was totally shocked because the last I heard, several months before, when I sold the McMansion in the burbs and bought a smaller house for my wife and daughter and a Condo for myself my scores were in the 800s.

This was back in the late 1990s when your credit reports and scores were virtually impossible to get but after much leg work I was able to get copies of my reports.

To make a long story short...

My ex-wife, shortly before our divorce, had fraudulently opened 5 credit cards in my name using my social security number and other information. Because of my excellent credit the CLs were high! About a month before the divorce she maxxed out all the cards and never made a payment!

She also filed for bankruptcy and let her house go into foreclosure and yes, my name was on that mortgage and some of her accounts listed in the bankruptcy so they all showed up on my reports! She was trying to shift all of our equitably divided debts onto me!

So yes...

Five recent credit cards in collections.

A recent bankruptcy.

A recent foreclosure.

And you too can earn a 350 but you really have to work at it!

[Anonymous]

All fascinating and delightful additions to the thread.  Many thanks to everyone so far.  This sentence will keep me cheerful for at least a month: "And you too can earn a 350 but you really have to work at it."

[sphinx313]

The bottom of my life is in my sig. I was below 400.