Reason code/statement concordance chart

[Curious_George2]

---

@iv wrote:

UPDATE: version 0.5.2

Source here: https://drive.google.com/file/d/1KawXdpMM1dUJ2Qlaez_qcFHMhYjzBBUB/view?usp=sharing

New features!

Resilience Score Reason Code numbers are now displayed on FRI screen:

Numeric 0-5 Ratings are now displayed on FICO 8 Score Ingredients screen:

Scores without Reason Codes (within 50 points of model max) are now also included in CSV export

Resilience Scores are now included in CSV export (including all prior FRI scores, not just current)

Additional columns in CSV export for numeric Score Ingredients

Available for Firefox in AMO, and for Chrome in CWS. Source available in post #3.

---

I'm having a blast exploring the spreadsheet that your CSV export produces. Thank you so much for building this incredible tool! 

I have a few questions about what's in the spreadsheet. I don't expect you to have the answers (though at this point, it wouldn't surprise me!), I'm posing them to anyone who cares to explore or answer.

It seems like there should be something called a Classic 9, but there's not. It's labeled as a Mortgage score in the spreadsheet. Is that meaningful?

Those 0-5 Ratings of the Score Ingredients seem really interesting. I wonder what we could learn from them. As just one small example, I have 5 at all points for Credit Mix, so it seems like I have maxed out that portion of the score. But all I have are bankcards, an open-ended card, and non-mortgage installment loans. That seems like pretty good evidence that a mortgage is not needed to max out credit mix. It's cool that these five numeric ratings continue to show up for scores that are too high to show Reason Codes, and that you have discovered that they're available from the past, so we can see which actions did and didn't produce changes in our particular ingredient ratings. 

[Anonymous]

---

@Curious_George2 wrote:

---

@iv wrote:

---

(Also, just a note - this is only tested and setup for 3B reports. I don't have any 1B reports to test with, so I haven't added support for those.)

---

I'd say don't sweat it and don't bother trying. The 1B reports are in a completely different format. I think it might be the built on the platform MF was using in 2012. I'm guessing little or none of your work up to this point would be applicable to the 1B reports, and it seems like relatively few people purchase them, so the benefit would be small.

---

I initially tried it with the 1B reports first, well because the formats are way better (thorough/informative/correct order of negative reason codes), but alas, I got no codes along with each negative reason.

[iv]

---

@Curious_George2 wrote:

Here's all I can currently contribute to filling in the gaps on the chart in post #1.

---

Excellent! Slowly getting there!

---

@Curious_George2 wrote:

For Code 02, the MF Short text is "Missed payments". The chart shows this as (All), but I have only seen it on the following: EQ5, EQ AU5, EQ BC5, and EX BC3. There is zero overlap between those bureau-versions and the ones where I have seen Code 18, which has essentially the same meaning. This make me question the (All) designation for 02. I propose a hypothesis that 02 and 18 are mutually exclusive within a given bureau-version.

---

Entirely possible - we know that the "base" reasons nominally exist everywhere... but have seen in practice that it appears to not always be true. It does seem to be especially focused on the "early/older(?)" low-numbered reason codes - you are seeing that here with 02/18, and there's also the oddity noted earlier in this thread with 03/33 and 04/32.

---

@Curious_George2 wrote:

There is a positive reason Code 513. The MF Short text is "Bills paid on time recently". The long text is "You've recently been paying your bills on time".

---

Thanks!

The positive reason codes can also provide partial info on the matching negatives - we have 516 but not (yet) 16, 519 but not 19, and 540 but not 40.

[iv]

---

@Curious_George2 wrote:

I'm having a blast exploring the spreadsheet that your CSV export produces. Thank you so much for building this incredible tool! 

---

I'm glad you like it! It's just a few hundred lines of javascript, surfacing data that's already there, just not directly visible...

...currently, at least.

---

@Curious_George2 wrote:

It seems like there should be something called a Classic 9, but there's not. It's labeled as a Mortgage score in the spreadsheet. Is that meaningful?

---

Yeah, I thought that was interesting, too. I put a comment in the source about it, but hadn't mentioned it here yet.

Yes, in the back-end JSON data, FICO 9 Classic is marked as "score_type: MORTGAGE"!

Obviously, in the normal UI, it's not in the mortgage section, and is just marked FICO 9.

Now, it's not clear if that was just a kludge, used when FICO 9 was first added to the reports, and there was no CLASSIC catergory to use... just like EX FICO 3 is marked BANKCARD, even though it's actually a Classic model. (FICO 8 Classic scores are stored differently from all the other types in the JSON.)

...Or! It could also have been FICO jumping the gun (incorrectly) on assuming that Freddie/Fannie would modernize their score requirements from 5/4/2, and were settling on FICO 9 Classic. Clearly, that didn't happen (and probably won't - either 8 Classic or 10T seem more likely choices today), but maybe that was the expectation at the time? Total speculation, either way. But interesting.

(I did consider special-casing FICO 9 for the CSV export, to override the MORTGAGE tag with CLASSIC, but decided not to, at least for the moment.)

---

@Curious_George2 wrote:

Those 0-5 Ratings of the Score Ingredients seem really interesting. I wonder what we could learn from them. As just one small example, I have 5 at all points for Credit Mix, so it seems like I have maxed out that portion of the score. But all I have are bankcards, an open-ended card, and non-mortgage installment loans. That seems like pretty good evidence that a mortgage is not needed to max out credit mix. It's cool that these five numeric ratings continue to show up for scores that are too high to show Reason Codes, and that you have discovered that they're available from the past, so we can see which actions did and didn't produce changes in our particular ingredient ratings. 

---

Yup! I'm thinking about other ways to analyze that data, too... matching up changes in Reason Codes, Ingredients, and Scores, to narrow down changes. Not sure yet what the best presentation method for the data would look like.

It's not (yet?) clear what the exact relationship between score criteria/reason codes and the ingredient numbers are - but with more people looking at it, we may be able to identify some patterns...

[iv]

---

@Anonymous wrote:

---

@Curious_George2 wrote:

---

@iv wrote:

---

(Also, just a note - this is only tested and setup for 3B reports. I don't have any 1B reports to test with, so I haven't added support for those.)

---

I'd say don't sweat it and don't bother trying. The 1B reports are in a completely different format. I think it might be the built on the platform MF was using in 2012. I'm guessing little or none of your work up to this point would be applicable to the 1B reports, and it seems like relatively few people purchase them, so the benefit would be small.

---

I initially tried it with the 1B reports first, well because the formats are way better (thorough/informative/correct order of negative reason codes), but alas, I got no codes along with each negative reason.

---

Yeah, I'm only processing the 3B reports right now, not the 1B (or the identity reports, or the alerts).

(Transunion alerts do actually have short-text reasons attached for score changes, all the way up to 849, and I'll probably add that at some point.)

It would probably be quite easy to add 1B support, I'd just have to pay for one of each CRA's version of the 1B to check the data formats... If there's requests for it (like if you have older 1B reports with data on them that is not longer on current 3B reports, and could be useful...), I could spend the $50 and an hour or so of time to add 1B support.

[LP007]

---

@Anonymous wrote:

---

@Anonymous wrote:

---

@iv wrote:

---

(Also, just a note - this is only tested and setup for 3B reports. I don't have any 1B reports to test with, so I haven't added support for those.)

---

@iv wow that's awesome. @Anonymous May be able to test that. 

I lost all the 1Bs I had when I canceled my subscription back in December. And all my 1B/3B PDFs don't have the IDs in them, so that's 2 years of hidden codes lost to the Great Null & Void.

---

Off-topic..................

Who locked the Resilience Squad Leaderboard posts? (

It's pad-locked with no reply buttons.

[SouthJamaica]

---

@LP007 wrote:

---

@Anonymous wrote:

---

@Anonymous wrote:

---

@iv wrote:

---

(Also, just a note - this is only tested and setup for 3B reports. I don't have any 1B reports to test with, so I haven't added support for those.)

---

@iv wow that's awesome. @Anonymous May be able to test that. 

I lost all the 1Bs I had when I canceled my subscription back in December. And all my 1B/3B PDFs don't have the IDs in them, so that's 2 years of hidden codes lost to the Great Null & Void.

---

Off-topic..................

Who locked the Resilience Squad Leaderboard posts? (

It's pad-locked with no reply buttons.

---

The moderators do that

[sarge12]

---

@SouthJamaica wrote:

---

@LP007 wrote:

---

@Anonymous wrote:

---

@Anonymous wrote:

---

@iv wrote:

---

(Also, just a note - this is only tested and setup for 3B reports. I don't have any 1B reports to test with, so I haven't added support for those.)

---

@iv wow that's awesome. @Anonymous May be able to test that. 

I lost all the 1Bs I had when I canceled my subscription back in December. And all my 1B/3B PDFs don't have the IDs in them, so that's 2 years of hidden codes lost to the Great Null & Void.

---

Off-topic..................

Who locked the Resilience Squad Leaderboard posts? (

It's pad-locked with no reply buttons.

---

The moderators do that

---

Any idea of why they locked out that thread to further comments? I have never seen much contentious posts there. They recently locked the Fico high acheivers thread too.

[iv]

---

@sarge12 wrote:

---

@SouthJamaica wrote:

---

@LP007 wrote:

Off-topic..................

Who locked the Resilience Squad Leaderboard posts? (

It's pad-locked with no reply buttons.

---

The moderators do that

---

Any idea of why they locked out that thread to further comments? I have never seen much contentious posts there. They recently locked the Fico high acheivers thread too.

---

No idea.

Could just be getting too long - there have been other over-length and older (in terms of initial post date) threads that have been closed to encourage starting a "fresh" version... but without a comment, who knows?

And given the last bullet point in...

https://ficoforums.myfico.com/t5/User-Guidelines-General/myFICO-Forums-Terms-of-Service/m-p/4

... I'm not going to guess or question further.

[iv]

---

@iv wrote:

---

It would probably be quite easy to add 1B support, I'd just have to pay for one of each CRA's version of the 1B to check the data formats... If there's requests for it (like if you have older 1B reports with data on them that is not longer on current 3B reports, and could be useful...), I could spend the $50 and an hour or so of time to add 1B support.

---

@Anonymous @Curious_George2 

Ok... I added a set of 1B reports to try this out.

Not only do they LOOK like the older site layout - they literally ARE the older site, embedded in the new one as an IFRAME.

Much more limited data available than the current 3B reports - no Ingredient numeric scores, just the "ratings", and no numeric Reason Codes... but since we have a fairly large list of the mappings now, I embedded a lookup table for the ones we already have.

IF your 1B codes are already on the list here, then they will be displayed on the 1B Reports as of version 0.6.1.

(Of course, that means that if you see a number there, we already have it... and if the number is missing, you have a code we don't have... but the 1B Report can't be used to add to the list.)

The data that's actually available is also added to the end of the CSV export - but with "null" for the numeric Reason Codes (it does have both long and short text), and text-only (vs numeric) for the Ingredients columns.