topic Re: TU98 vs. TU04 and FICO/FAKO in Understanding FICO® Scoring

TU98 vs. TU04 and FICO/FAKO

Walt_K — Sun, 24 Jul 2011 23:33:33 GMT

Anyone have any data on the correlation between TU98 and TU04? Also, any idea on what percentage of lenders are using TU04 vs. TU98. I feel like there is a disservice done by blasting FAKO scores because they are not what a lender sees when the only TU score you can buy on this site is an outdated scoring model. Now if it's true that TU98 is still widely used by lenders, or if it's true that TU98 is within say +-10 points of TU04 for most people, I suppose it isn't quite comparable to a FAKO score.

Re: TU98 vs. TU04 and FICO/FAKO

Booner72 — Sun, 24 Jul 2011 23:44:01 GMT

HI walt - did you read my post about the example of my scores TU04 TU98? My TU04 was 13 points higher. Pulled only 1 day apart and all info the same except 1 day of aaoa.

Re: TU98 vs. TU04 and FICO/FAKO

Booner72 — Sun, 24 Jul 2011 23:46:35 GMT

Wait - my TU04 was 13 points LOWER than TU98. Sorry. And I know that this isn't proof that the scores will be 13 points difference for everyone, or anyone but me. But I guess it's kind of close, right? Credit Karma was only 4 points different. Who the heck knows? I agree that it is not cool to advertise the TU98 here as a "real score" even though technically it's a real FICO, but it's not used for mortgages much. I read Lending Tree used it?

Re: TU98 vs. TU04 and FICO/FAKO

Walt_K — Sun, 24 Jul 2011 23:58:03 GMT

I just found and read your thread. Hoping for more info like that. I'm not concerned with the score from Vantage or TransRisk or whatever being the same number as a FICO score. As has been said many times, they are different scoring models and aren't even on the same scales. Making something out of the fact that they are close to the same number every once in a while is a little like finding it interesting when it is the same temperature Celsius as it is Fahrenheit (realizing I haven't written that word since high school and that I probably spelled it incorrectly).

TU98 and TU04 are at least on the same scale. So you can make more meaningful comparisons between the absolute numbers. But I wonder how wide a disparity is seen. I don't know personally because the only TU score I have access to is the one I bought here.

If it's a much narrower window for most people, then it isn't as big of a deal. And the more we can figure out about predictable differences in the weighting of the factors, the more useful TU98 would be for people. But if it's common to see 30 point differences between the two TU scoring models, then we might as well call TU98 a FAKO.

Re: TU98 vs. TU04 and FICO/FAKO

Walt_K — Mon, 25 Jul 2011 00:02:12 GMT

Also, I've read here a couple times that it is TU that is responsible for us not being able to pull TU04. But the scoring model is the property of FICO. Is the allegation that TU sells reports to Fair Isaac on the condition that it not apply TU04 to the data in those reports?

Re: TU98 vs. TU04 and FICO/FAKO

Booner72 — Mon, 25 Jul 2011 00:08:18 GMT

It would be interesting to see if they continue to be 13 points apart. I believe my LO is going to pull credit again before closing. I'll come here and pull another TU98 and see if it stays 13 points apart. My guess, it won't. It will be more or less.

It would be good if more people could pull their TU98 after they get scores from a lender and make the comparison. But it is a spendy thing to do. Nobody has 20 bucks to just throw around for the sake of figuring out a scoring difference when they are in the middle of a big purchase like a house or a car.

Re: TU98 vs. TU04 and FICO/FAKO

Walt_K — Mon, 25 Jul 2011 03:09:14 GMT

Booner72 wrote:
It would be interesting to see if they continue to be 13 points apart. I believe my LO is going to pull credit again before closing. I'll come here and pull another TU98 and see if it stays 13 points apart. My guess, it won't. It will be more or less.

It would be good if more people could pull their TU98 after they get scores from a lender and make the comparison. But it is a spendy thing to do. Nobody has 20 bucks to just throw around for the sake of figuring out a scoring difference when they are in the middle of a big purchase like a house or a car.

Agreed. It would be nice if we had some more data points for people that happen to know both. For instance, people who pulled TU and EQ here before going into a mortgage app to see where they stood and then pulled soon thereafter when they applied for the mortgage. But I wouldn't recommend anyone spending the extra $20 to satisfy my curiosity or engage in this very unscientific attempt at a correlation.

My main concern is that people bash on FAKO all the time because it's not what lender's see. We tell people that your FICO score is what really matters. And while the information is certainly out there that TU98 isn't the most up to date version of TU, that often gets glossed over. So a person who comes over here, finally thinks they've figured out to stop wasting energy worrying about FAKOs, goes and spends money to see their TU score, only to find out later that it may be unreliable as well.

Re: TU98 vs. TU04 and FICO/FAKO

Booner72 — Mon, 25 Jul 2011 03:23:21 GMT

+100 - I totally agree.

Re: TU98 vs. TU04 and FICO/FAKO

tombking — Mon, 25 Jul 2011 16:06:28 GMT

Has anyone had TU 04 be higher than TU 98? I'm wondering if we can figure out any patterns.

Re: TU98 vs. TU04 and FICO/FAKO

beamMEup — Mon, 25 Jul 2011 16:29:34 GMT

tombking wrote:
Has anyone had TU 04 be higher than TU 98? I'm wondering if we can figure out any patterns.

As far as figuring out patterns, llecs posted some helpful info on this thread:

I don't think anyone can accurately describe the key differences between both versions. Info is vague from FICO and we don't have access to TU04 to do a daily, weekly, or monthly comparison. It's my understanding they added bucket(s), changed how inquiries are duplicated (auto & mortgage), they treat lates a bit differently and scores on trending a bit more, and treats NPSL and HELOCs a bit differently per util. All this info is available on FICO's website via webinars, docs, etc.

That being said, my TU 04 and TU 98 were off by 11 points, as were DH's. (I don't remember which was higher for us, though - sorry). However, it's important to note that folks have widely varying spreads. There are so many variables and constantly moving targets in FICO scores that it's very difficult to decipher and untangle not just a single pattern, but the myriad factors comingling within a single FICO score. (Sort of like me trying to decipher the thinking in my teenage son's brain cells :smileyindifferent:)

Re: TU98 vs. TU04 and FICO/FAKO

Walt_K — Tue, 26 Jul 2011 00:35:37 GMT

Understood. I highly doubt you would find any kind of correlation that said your TU98 would always be 12 points lower than your TU04 or vice versa. If so, it would be an awfully stupid endeavor to create a new formula.

But it would be interesting to know if the two were always within 15 points of each other one way or the other for 95% of people or something like that. Or if you could say that for profiles that do not have AU accounts, they are closer, or for profiles that do not have charge cards, they are closer.

But I say if the spread is in the 30 point range, then we might was well start calling TU98 a FAKO. Unless, of course, there are a significant number of lenders still using this model.

Re: TU98 vs. TU04 and FICO/FAKO

dookdig — Tue, 26 Jul 2011 01:28:39 GMT

FYI, My TU04 from discover denial letter is 13 points higher than TU98 (747/734)

I believe new version favor more on very low UTI%( less than 7-9%). Utilizing more than 9% will likely result in lower TU04 than TU98.

Re: TU98 vs. TU04 and FICO/FAKO

AndySoCal — Tue, 26 Jul 2011 02:14:24 GMT

As been noted before the dedup time frame in FICO 04 was increased. I read another post on this forum where the poster had posted the TU socre model information from TU. There is not that much difference, the number of scorecards are the same. Some will benefit from the update some will not this can be said about every score model that comes out.

Re: TU98 vs. TU04 and FICO/FAKO

bigtim — Fri, 29 Jul 2011 11:26:10 GMT

Well, I just received my free TU FICO from my Walmart card and it is 28 points higher (730) than the one I paid for on 6/22 from here. I know myFICO still uses the older version, so given all the circumstances I feel I got a bargain AND a more "accurate" score in terms of a more up-to-date model.

Re: TU98 vs. TU04 and FICO/FAKO

visorboy1974 — Fri, 29 Jul 2011 17:35:29 GMT

bigtim wrote:
Well, I just received my free TU FICO from my Walmart card and it is 28 points higher (730) than the one I paid for on 6/22 from here. I know myFICO still uses the older version, so given all the circumstances I feel I got a bargain AND a more "accurate" score in terms of a more up-to-date model.

Does anybody know for sure what version Walmart uses?

Re: TU98 vs. TU04 and FICO/FAKO

MarineVietVet — Fri, 29 Jul 2011 19:37:52 GMT

visorboy1974 wrote:

Does anybody know for sure what version Walmart uses?

That is still being looked into.

From a BK years ago to:
EX - 3/11 pulled by lender- 835, EQ - 2/11-816, TU - 2/11-782

"Some people spend an entire lifetime wondering if they've made a difference. The Marines don't have that problem".

Re: TU98 vs. TU04 and FICO/FAKO

determinedtoown — Mon, 19 Nov 2012 07:20:04 GMT

I just got a denial from us bank. TU04 621 Dakota here is 565. BIG DEFERENCE! I want to buy a house so that is a significant gap!

Re: TU98 vs. TU04 and FICO/FAKO

determinedtoown — Mon, 19 Nov 2012 07:21:23 GMT

*fako ...spell check turned it into Dakota :)

Re: TU98 vs. TU04 and FICO/FAKO

TrishN — Mon, 19 Nov 2012 16:21:05 GMT

My fico TU 650 and Walmart TU 619 . I have no idea what was wrong lol. I paid for my fico on 11/15 and received Walmart store card afternoon . Checked same day but 31pt away. God know this

Re: TU98 vs. TU04 and FICO/FAKO

HiLine — Wed, 21 Nov 2012 17:12:15 GMT

determinedtoown wrote:
I just got a denial from us bank. TU04 621 Dakota here is 565. BIG DEFERENCE! I want to buy a house so that is a significant gap!

If you're looking for mortage, MyFICO's Score Watch service provides scores based on a model used by most mortgage lenders.