cancel
Showing results for 
Search instead for 
Did you mean: 

Anyone else think the algorithm is wrong???

tag
Brian_Earl_Spilner
Credit Mentor

Re: Anyone else think the algorithm is wrong???


@Thomas_Thumb wrote:

@Brian_Earl_Spilner wrote:
Lenders will always use their own criteria for approvals.  I've always been an outlier and able to get approved for things my scores say I shouldn't. It also goes the other way. Cap1 is not friendly to people with 800+ scores. 

Doom!?

 

Will my Discover card CL be slashed when Cap One takes over? Will my card be cancelled?


 


The sad thing is yes, your limits could be slashed. It wasn't that long ago when cap1 reviewed accounts and started slashing them based on usage. 

    
Message 31 of 53
Brian_Earl_Spilner
Credit Mentor

Re: Anyone else think the algorithm is wrong???


@ThomasJNewton wrote:

@CorpCrMgr1 wrote:

 

Being married to a  programmer, it is clear to me that the algorithms are formed from an opinion. Actually, a lot of opinions. Somewhere it was decided what would count as good vs. what would count as not good. Then with 0's and 1's this was programmed into a computer. Of course a lot of input and tweaking must of taken place.

 

Giant datasets are analysed for trends. Profiles that are otherwise identical are compared for particular actions as to how likely those accounts go 30/60/90 days late. Those factors that have the highest correlation get built into their models. This is not an opinion. No one in the process says: "I think being 30 days late is bad and so we should punish anyone who does that." (an opinion), but instead, after analysis of millions of profiles it is observed that: "those accounts with a 30-day late payment are x% more likely to have accounts hit 90-days late, so points should be removed from their scores." (an analysis). These decisions are not being made by programmers, but by statisticians and data scientists.

 

Since there is no way for credit bureaus to know why any individual did anything, they cannot adjust for problems like a checking account number changed  and the auto-pay information was not updated correctly to one just did not have enough money to make the payments. One's credit issuer can take those things into account and decide to remove derogatory information they supplied, but the CRAs cannot.

 

 


I'm going to play devil's advocate here for a moment. @CorpCrMgr1 believes there was outside influence by people on how credit should be reported and interpreted. They're not wrong. The reason why reporting of delinquencies is 7 years is due to congress. Some representatives believed it shouldn't report past 2 years because it could unfairly penalize people if reported longer. Others felt it should last 10-13 years so lenders would have more data to use in their decisions. 7 years was the compromise. How much different would people's experiences be if the time limit was based on one of those extremes? It's been reported that 43% of Americans have missed a payment in the last 5 years. So (we'll say single derogatory for argument's sake), their ability to obtain credit and what rate they would receive swings based on how long a single late is reported and not on their actual creditworthiness.

 

But, I believe that's where the influence stops. Algorithms are created using data, statistics, and correlation. 

    
Message 32 of 53
FicoMike0
Valued Contributor

Re: Anyone else think the algorithm is wrong???

Wow, this thread has wandered all over the place. What I've concluded is that it's data driven, within legal constraints. They aren't allowed to consider baddies more than 7 years old, or your income, change of income, race, marital status, etc. otherwise, it's a a big data mining exercise.

Is it fair? That's not it's goal, even if we cout agree on a definition. I believe it's goal is to predict being 90 days late in two years. So, I can see being late at all lowering my score.

 

Message 33 of 53
CorpCrMgr1
Valued Contributor

Re: Anyone else think the algorithm is wrong???

@Brian_Earl_Spilner  is it "sign in" or "log in" ? When programmers cannot be consistent on how one enters a site indicates the different thought process of programmers. Imagine programmers putting together a credit grading process. The sign in criticism originated from a programmer sitting at Ford HQ on 12-31-99 at midnight to save the system from crashing. I sat next to her and played now primitive computer games. FWIW, the system was good to go!

Message 34 of 53
Horseshoez
Senior Contributor

Re: Anyone else think the algorithm is wrong???


@CorpCrMgr1 wrote:

@Brian_Earl_Spilner  is it "sign in" or "log in" ? When programmers cannot be consistent on how one enters a site indicates the different thought process of programmers. Imagine programmers putting together a credit grading process. The sign in criticism originated from a programmer sitting at Ford HQ on 12-31-99 at midnight to save the system from crashing. I sat next to her and played now primitive computer games. FWIW, the system was good to go!


The point is, programmers have exactly zero to do with the credit grading process.

Chapter 13:

  • Burned: AMEX, Chase, Citi, Wells Fargo, and South County Bank (now Bank of Southern California)
  • Filed: 26-Feb-2015
  • MoC: 01-Mar-2015
  • 1st Payment (posted): 23-Mar-2015
  • Last Payment (posted): 07-Feb-2020
  • Discharged: 04-Mar-2020
  • Closed: 23-Jun-2020

 

I categorically refuse to do AZEO!

In the proverbial sock drawer:
Message 35 of 53
CorpCrMgr1
Valued Contributor

Re: Anyone else think the algorithm is wrong???

@Horseshoez  IT professionals enter the programs. The parameters are drawn up somewhere. I posted this several weeks ago. MyFICO Eq score dropped four points because of a reported $190 charge on my Chase CL CC of $16k. Then the charge was paid the same day and my score went back up four points. I owed $230. The total on my other cards totaled less than $1K, under 1%. My TCL is over $225k. I did not pay $190. I had no $190 balance. Somehow Equifax determined I owed money that I didn't. Somehow Equifax determined I made a payment that I didn't. A decision was made to drop my score. A decision was made to restore my score.

Message 36 of 53
Blender
Established Contributor

Re: Anyone else think the algorithm is wrong???


@Brian_Earl_Spilner wrote:

@AlgorithmIsWrong wrote:

Prediction is pretty much an opinion in my eyes and peoples credit worthiness should not be  judged based on an opinion.  If one late payment hurts you by 20+ points when all your prior payments were on time, one would take into account all the positive history and slightly ding the person for a late payment. Key word there is "slightly."   Now, if the late payments continue, then sure, I could understand the huge decline.  This is why the system is flawed.  

 

Perhaps, delayed reporting from agencies to the bureaus is warranted.  Where a 3 to 6 month period is scored rather than individual months.


The score is a quick glance at likelihood to perform and is based on datapoints. As mentioned before, if they find a correlation between datapoints that is consistent across enough people, it can be assumed that it is a fairly accurate predictor of what can happen and factored into the algorithm.

 

Now, is it fair? I don't think so. I just finished a 7 year rebuild meaning I went 7 years since my last derogatory, but was penalized until the end. I also don't like the fact that paying my chargeoffs and collections still hurt me more than a bankruptcy. I fulfilled my obligations, people who filed bankruptcy did not. I didn't hit the 700s until about 5 years in, while some filing bankruptcy were able to do it in as early as 2 years. I was relegated to toy limits while people literally a week out of bankruptcy were able to get multi-thousand dollar limits from a lender they just burned. So I get what you're saying in regards to scoring us on the overall reporting versus a single incident. But that's not how it works. The score is based on a snapshot of your profile at that moment in time. The new trending scores will hopefully help in a situation like mine, showing I was paying everything down and off and scoring me appropriately for it.

 

Having said all that, the score is just a number and doesn't dictate likelihood of approval. Lenders will always use their own criteria for approvals.  I've always been an outlier and able to get approved for things my scores say I shouldn't. It also goes the other way. Cap1 is not friendly to people with 800+ scores.  I use it as just a guideline and to quickly verify nothing is wrong. If I see my score drop 30 points, I know there's something that needs my attention. Otherwise, I don't bother checking it before an app as the lender will either approve me, or they won't. 


Was in the same boat, and now Im helping friends in similar situations with their credit mistakes. I just don't understand this.

6/2024


Message 37 of 53
Horseshoez
Senior Contributor

Re: Anyone else think the algorithm is wrong???


@CorpCrMgr1 wrote:

@Horseshoez  IT professionals enter the programs. The parameters are drawn up somewhere. I posted this several weeks ago. MyFICO Eq score dropped four points because of a reported $190 charge on my Chase CL CC of $16k. Then the charge was paid the same day and my score went back up four points. I owed $230. The total on my other cards totaled less than $1K, under 1%. My TCL is over $225k. I did not pay $190. I had no $190 balance. Somehow Equifax determined I owed money that I didn't. Somehow Equifax determined I made a payment that I didn't. A decision was made to drop my score. A decision was made to restore my score.


While what happened to your score is not obviously logical, I can assure you nothing an IT professional did biased that score.  

Chapter 13:

  • Burned: AMEX, Chase, Citi, Wells Fargo, and South County Bank (now Bank of Southern California)
  • Filed: 26-Feb-2015
  • MoC: 01-Mar-2015
  • 1st Payment (posted): 23-Mar-2015
  • Last Payment (posted): 07-Feb-2020
  • Discharged: 04-Mar-2020
  • Closed: 23-Jun-2020

 

I categorically refuse to do AZEO!

In the proverbial sock drawer:
Message 38 of 53
CorpCrMgr1
Valued Contributor

Re: Anyone else think the algorithm is wrong???

@Horseshoez  somehow the erroneous information was entered into my CR to cause my score to drop. Even if the charge was correct, a four point drop when I'm at 1% utilization doesn't make sense with a FICO score. Then a payment appears. Trust me I checked all financial institutions, no payment was made. I checked other CC beside Chase, which is the card with the supposed charge. Nothing. 

Information is being entered into CR that are incorrect. Others claim mistakes. I've been the victim of cross listings with a family member. Addresses listed that I have no connection. The complete misspelling of my first name. 

Only Forum posters are stating their issues. I've helped dozens of people fix their CR. The mistakes are not a few. The first step I always take is to remove what is wrong.

Message 39 of 53
Horseshoez
Senior Contributor

Re: Anyone else think the algorithm is wrong???


@CorpCrMgr1 wrote:

@Horseshoez  somehow the erroneous information was entered into my CR to cause my score to drop. Even if the charge was correct, a four point drop when I'm at 1% utilization doesn't make sense with a FICO score. Then a payment appears. Trust me I checked all financial institutions, no payment was made. I checked other CC beside Chase, which is the card with the supposed charge. Nothing. 

Information is being entered into CR that are incorrect. Others claim mistakes. I've been the victim of cross listings with a family member. Addresses listed that I have no connection. The complete misspelling of my first name. 

Only Forum posters are stating their issues. I've helped dozens of people fix their CR. The mistakes are not a few. The first step I always take is to remove what is wrong.


Fair enough, but I have to ask, what does erroneous information have to do with programmers?

Chapter 13:

  • Burned: AMEX, Chase, Citi, Wells Fargo, and South County Bank (now Bank of Southern California)
  • Filed: 26-Feb-2015
  • MoC: 01-Mar-2015
  • 1st Payment (posted): 23-Mar-2015
  • Last Payment (posted): 07-Feb-2020
  • Discharged: 04-Mar-2020
  • Closed: 23-Jun-2020

 

I categorically refuse to do AZEO!

In the proverbial sock drawer:
Message 40 of 53
Advertiser Disclosure: The offers that appear on this site are from third party advertisers from whom FICO receives compensation.