@Thomas_Thumb wrote:Inverse, welcome to the board.
I suspect your tax liens are affecting your upward potential "glass ceiling". Once these drop off and your profile is clean, you should be in a better position to evaluate positive/negative response to various factors. Until then, responses may be muted if you are near the upper or lower boundry of a scorecard. Also, the data I have seen posted suggests responses to factors differ depending on whether a profile is clean or not clean.
I posted a hypothetical framework for a simplified Fico 8 model on another thread, Unfortunately, my proposed framework only relates to clean files. Nonetheless, it might provide some insights for you and constructive feedback is always welcome. I have not seen a structured Fico model presented publicly and thought it could be interesting to build one from the ground up based objective review of poster data sets where confounding of factors can be minimized.
A representative risk assessment model with publically open architecture would be a nice tool for planning strategy for people to build (or protect) credit worthiness.
Thanks for sharing your data.
Sigh, are you seriously suggesting that because people have dirty reports that their data is useless?
Over and over it has been demonstrated previously and currently that the number of tradelines reporting a balance matters and every single time you discount it with some blahblahblah statement.
(Edited)
@Revelate wrote:
@Thomas_Thumb wrote:Inverse, welcome to the board.
I suspect your tax liens are affecting your upward potential "glass ceiling". Once these drop off and your profile is clean, you should be in a better position to evaluate positive/negative response to various factors. Until then, responses may be muted if you are near the upper or lower boundry of a scorecard. Also, the data I have seen posted suggests responses to factors differ depending on whether a profile is clean or not clean.
I posted a hypothetical framework for a simplified Fico 8 model on another thread, Unfortunately, my proposed framework only relates to clean files. Nonetheless, it might provide some insights for you and constructive feedback is always welcome. I have not seen a structured Fico model presented publicly and thought it could be interesting to build one from the ground up based objective review of poster data sets where confounding of factors can be minimized.
A representative risk assessment model with publically open architecture would be a nice tool for planning strategy for people to build (or protect) credit worthiness.
Thanks for sharing your data.
Sigh, are you seriously suggesting that because people have dirty reports that their data is useless?
Over and over it has been demonstrated previously and currently that the number of tradelines reporting a balance matters and every single time you discount it with some blahblahblah statement.
A file with three negatives, which seems to top out at 740 (not bad, actually, for negatives being in the file) while the cardholder tries to improve the score, certainly sounds like a ceiling to me. Said another way, it would sound wierd to say someone with negatives on their file could expect to get to 800. Once this poster has the liens off the file, then the weight is lifted (as it will be for your file) and the score should skyrocket. Until then, with the weight on the file, I'd have to agree that the responses to various changes could be reacting strangely.
Having said that, I will agree that going from 1/3 to 3/3 cards (100% of cards reporting something) was well tested with equal utilization, which I applaud. There was obviously a significant score change there. I would expect a similar score change for someone close to 850, going from few cards to 100% of cards reporting, because the file is already at the top of the range, and there is a lot to give up for getting away from relative perfection. This poster, as well, is near the top of their current range, so there is a lot to give up by raising risk by showing 100% of cards reporting.
@jamie123 wrote:
FICO 08 scores are generally used by lenders for UNSECURED credit like credit cards and once your scores are above 800 you will probably get the best interest rates on a credit card that any lender offers. It really won't save you any money though because by having scores in the 800s you are probably not carrying balances month to month.
Consider yourself lucky. I guarantee that you haven't been divorced, had a medical emergency that laid you up for a few months during the last seven years and went through the last recession virtually unscathed. Sometimes bad things happen to good people that leave them with some dings on their credit. You will find a lot of them here on MyFICO trying their best to repair them.
Well, there is that, however I've got extensive history with carrying balances, and two BofA cards that give me 0% APR offers with a 4% fee, so there are other ways to finnagle a low APR on carried balances 
Indeed, "There but for the grace of God..." so to speak. Some of us have been fortunate, many have not, and we each have to make the best of our situation.
I ran multiple iterations with this score generator last year, documenting the variations carefully. It topped out at 830:
5+ cards, 20+ years both loans and cards, no apps in the last two years, thus no new cards, 0-4 cards reporting, less than $500 total amount outstanding, no lates of course, util 0% - 9%.
But my main purpose was to see the effect of changes in utilization, not getting new cards, number of cards less than 9 reporting (0 to 4 cards is just impossible for my ego) and from that perspective, I found it useful. I too have exceeded the top of my range for my current profile, so for absolute measure, I think the estimator is just a notch above useless. For getting an idea of directional impact for changes to the major measures in your file, however, I found that effort worthwhile. I think Time will tell.