Jump to content

Sign in to follow this  
donkeyshark

How Your Credit Score is Calculated

Recommended Posts

I've been reading through these forums now for a couple months and there seems to be a common theme woven into several questions that I think I can help shed some light on. I have read numerous posts that read something along these lines: "How much will paying off TL X or Y affect my credit score?" or "What will happen to my scores if my reports split?" These questions (and often their accompanying responses) are framed upon a misunderstanding as to how credit scores are actually calculated. As a statistician that works directly with these same types of algorithms (though not for FICO), I thought I'd try to clarify how these scores are calculated mathematically in an attempt to help you guys gain a better understanding as to how a company like FICO computes your scores. First I will explain the mathematics of machine learning from a 50,000 foot view for those of you who care to learn more, then I will provide cliff notes below for the mathematically challenged (or at least mathematically uninterested).

 

If you were to ask most people to guess how their scores are calculated, you will likely find that the most common guess would be that FICO assigns a certain weight to each TL and a certain penalty to 120-day lates, a certain penalty for BKs, a penalty for COs, etc. and that these penalties would decrease in weight as time progresses. However, this is not true, this is not at all how scores are calculated.

 

Math Nerd Explanation:

The science of credit scoring actually stems from a branch of artificial intelligence known as "Machine Learning" or "Predictive Analytics" where computers are trained how to think. There are many different modeling methods that we as statisticians use to make predictions (after-all, this is precisely what a credit score is - a numerical prediction of how likely you are to default on a loan). One of these modeling methods is called "Linear Regression". This is the type of linear mathematical model that most people think of when they are asked how something is scored - ie each feature or variable is assigned a weight that is used in calculating a particular score (e.g. - a certain penalty for BKs, a certain positive weight for good TLs etc.) You see this scoring method every time you watch a football game. A touchdown is assigned the weight of 6 points, a field goal is assigned the weight of 3 points, etc. A linear regression model would look something like this: y = b0 + b1x1 + b2x2 + ... + bnxn where y is the predicted label (or score), b0 is the y-intercept (or default value when all x's are equal to zero) and each b1-bn are the weights (or penalties) assigned to each x1-xn feature or variable (i.e. touchdowns, field goals or in the credit world: TLs, COs, Repos etc). However, while this statistical model proves quite useful for many applications, it is not very useful for predicting which people are likely to default on a loan for mathematical reasons beyond the scope of this post. Fortunately for statisticians, there are many other useful machine learning tools that we can use to more accurately predict who is likely to default on a loan. In the topic of credit scoring, we are mainly interested in the identification of rare instances (i.e. - instances where a borrower defaults on a loan). Creating an algorithm to detect rare instances among large sets of data poses some complex mathematical challenges. Fortunately, armed with super-computers, some linear algebra and IQs that would give our credit scores a run for their money, brilliant mathematicians (Robert Tibshriani, Jerome Friedman and others) have developed several machine learning techniques that prove quite useful for calculating credit scores. Among these techniques are Support Vector Machines (SVM), Stochastic Gradient Boosting (SGB) and Regression Trees would work as well. In the arena of credit scores, a Support Vector Machine is a classification system that attempts to classify whether or not someone is going to default on a loan. This SVM model works by taking borrower instances (vectors which inlclude their credit history information) as inputs and compares these vectors with other vectors using an algebraic operation known as the dot-product. The algorithm then groups these vectors together with other similar vectors (similar credit reports). It then attempts to divide these various vectors by forming a hyper-plane that separates them categorically. Measures of fitness can be added to the algorithm and mathematical transformations are used to return Credit Scores which can then be used by lenders to make decisions based on your likelihood of repaying a given loan.

 

Explanation:

You could envision the above process by taking a piece of paper and drawing a bunch of O's on the top half and a bunch of X's on the bottom half. Next, draw a few O's in with the X's and a few X's in with the O's. Then draw a line that separates the top half from the bottom half, thus grouping/separating the X's and O's categorically. This is a simplification of what the statistical algorithms that score your reports do (though they do it in multi-dimensional space that you can't envision instead of on a 2-dimensional piece of paper). They effectively group you categorically with other people that have similar credit reports (or "vectors" or X's and O's). There will be some misclassified credit reports (those would be the few O's among the X's and a few X's among the O's) that the SVM classifier gets wrong (i.e. - borrowers with high credit scores that default on a loan & borrowers with low scores that don't). This is why you often read replies such as "it depends" when people ask questions like "how much will this effect my score?" The truth is, no one knows the answers to those questions because they can't see into the classifier's algorithm and predict which vectors your credit report will most closely identify with after your report changes (not even a math nerd like me). What we do know is that with every improvement you make on your reports, you begin to look more and more like the borrowers that lenders want to loan money to and your scores will reflect that. However, a fairly good thing to keep in mind is that if you have several baddies on your report, your scores will likely climb more slowly at first because you will still look quite a bit like other borrowers that also have low scores (is there really much difference between 17 baddies and 18 baddies? - lenders think not). But as your report begins to clean up and more tradelines drop off, you will likely start to see some steady climbs in scoring. This is also why those of you whose scores start to get mixed in with the higher scores (high 700s and 800s), will have difficulty separating yourselves from others who also have high scores and your scoring increases begin to slow back down. You can think of it like a bell curve, where most of the elevation occurs in the middle and the two tails on the ends don't see much change in elevation (though this is more of a visual explanation than a correct usage of the bell curve). To really separate yourself from the pack of others with high scores, you will likely have to qualify for some rather impressive loans (and be responsible with them of course).

 

 

Cliff Notes:

If you walk away from this post learning one thing, let it be this: The TLs on your credit report are not assigned a particular value or weight that causes increases or decreases in your score when changes are made. The increases and decreases to your scores come from how closely your credit report resembles other reports after these changes. For those of you aiming to maximize your scores, you should pay close attention to the statistics that are printed in your CRs - the ones that tell you that "High Achievers" carry a 7% revolving balance to available credit limit or that "High Achievers" have no collections on their CRs. You want to look like "High Achievers" as much as possible. Remember, maximizing your credit score is not the same thing as maximizing financial responsibility. All other things being equal (and zero balance), the borrower who carries a revolving balance of 7% on his CCs is going to have a higher score than the borrower who pays his CCs in full every month. If your aim is to maximize your scores, then you want to emulate the "high achievers" as much as possible.

Share this post


Link to post
Share on other sites

Eigenvectors are sexy

 

Conjoint analysis rocks

 

Logit trumps OLS

 

Haha, yes, yes indeed. Eigenvectors are sexy and OLS is lame, log-odds rule!

Share this post


Link to post
Share on other sites

donkeyshark,

 

Nice. It would be a good post to point people to when they ask for the "formula" that FICO uses.

 

By using some of the information in your post I found this link to a rather nice description of many of these techniques as applied to credit scoring.

 

http://www.statsoft.com/textbook/credit-scoring/

 

Overview

Credit scoring is perhaps one of the most "classic" applications for predictive modeling, to predict whether or not credit extended to an applicant will likely result in profit or losses for the lending institution. There are many variations and complexities regarding how exactly credit is extended to individuals, businesses, and other organizations for various purposes (purchasing equipment, real estate, consumer items, and so on), and using various methods of credit (credit card, loan, delayed payment plan). But in all cases, a lender provides money to an individual or institution, and expects to be paid back in time with interest commensurate with the risk of default.

 

Of course this mathematical description would need to be relieved of attributes that, no matter how predictive, are either not a part of the info in consumer CRAs or are variables disallowed by law (age).

 

 

Again, thanks for the post.

Share this post


Link to post
Share on other sites

I've been reading through these forums now for a couple months and there seems to be a common theme woven into several questions that I think I can help shed some light on. I have read numerous posts that read something along these lines: "How much will paying off TL X or Y affect my credit score?" or "What will happen to my scores if my reports split?" These questions (and often their accompanying responses) are framed upon a misunderstanding as to how credit scores are actually calculated. As a statistician that works directly with these same types of algorithms (though not for FICO), I thought I'd try to clarify how these scores are calculated mathematically in an attempt to help you guys gain a better understanding as to how a company like FICO computes your scores. First I will explain the mathematics of machine learning from a 50,000 foot view for those of you who care to learn more, then I will provide cliff notes below for the mathematically challenged (or at least mathematically uninterested).

 

If you were to ask most people to guess how their scores are calculated, you will likely find that the most common guess would be that FICO assigns a certain weight to each TL and a certain penalty to 120-day lates, a certain penalty for BKs, a penalty for COs, etc. and that these penalties would decrease in weight as time progresses. However, this is not true, this is not at all how scores are calculated.

 

Math Nerd Explanation:

The science of credit scoring actually stems from a branch of artificial intelligence known as "Machine Learning" or "Predictive Analytics" where computers are trained how to think. There are many different modeling methods that we as statisticians use to make predictions (after-all, this is precisely what a credit score is - a numerical prediction of how likely you are to default on a loan). One of these modeling methods is called "Linear Regression". This is the type of linear mathematical model that most people think of when they are asked how something is scored - ie each feature or variable is assigned a weight that is used in calculating a particular score (e.g. - a certain penalty for BKs, a certain positive weight for good TLs etc.) You see this scoring method every time you watch a football game. A touchdown is assigned the weight of 6 points, a field goal is assigned the weight of 3 points, etc. A linear regression model would look something like this: y = b0 + b1x1 + b2x2 + ... + bnxn where y is the predicted label (or score), b0 is the y-intercept (or default value when all x's are equal to zero) and each b1-bn are the weights (or penalties) assigned to each x1-xn feature or variable (i.e. touchdowns, field goals or in the credit world: TLs, COs, Repos etc). However, while this statistical model proves quite useful for many applications, it is not very useful for predicting which people are likely to default on a loan for mathematical reasons beyond the scope of this post. Fortunately for statisticians, there are many other useful machine learning tools that we can use to more accurately predict who is likely to default on a loan. In the topic of credit scoring, we are mainly interested in the identification of rare instances (i.e. - instances where a borrower defaults on a loan). Creating an algorithm to detect rare instances among large sets of data poses some complex mathematical challenges. Fortunately, armed with super-computers, some linear algebra and IQs that would give our credit scores a run for their money, brilliant mathematicians (Robert Tibshriani, Jerome Friedman and others) have developed several machine learning techniques that prove quite useful for calculating credit scores. Among these techniques are Support Vector Machines (SVM), Stochastic Gradient Boosting (SGB) and Regression Trees would work as well. In the arena of credit scores, a Support Vector Machine is a classification system that attempts to classify whether or not someone is going to default on a loan. This SVM model works by taking borrower instances (vectors which inlclude their credit history information) as inputs and compares these vectors with other vectors using an algebraic operation known as the dot-product. The algorithm then groups these vectors together with other similar vectors (similar credit reports). It then attempts to divide these various vectors by forming a hyper-plane that separates them categorically. Measures of fitness can be added to the algorithm and mathematical transformations are used to return Credit Scores which can then be used by lenders to make decisions based on your likelihood of repaying a given loan.

 

Explanation:

You could envision the above process by taking a piece of paper and drawing a bunch of O's on the top half and a bunch of X's on the bottom half. Next, draw a few O's in with the X's and a few X's in with the O's. Then draw a line that separates the top half from the bottom half, thus grouping/separating the X's and O's categorically. This is a simplification of what the statistical algorithms that score your reports do (though they do it in multi-dimensional space that you can't envision instead of on a 2-dimensional piece of paper). They effectively group you categorically with other people that have similar credit reports (or "vectors" or X's and O's). There will be some misclassified credit reports (those would be the few O's among the X's and a few X's among the O's) that the SVM classifier gets wrong (i.e. - borrowers with high credit scores that default on a loan & borrowers with low scores that don't). This is why you often read replies such as "it depends" when people ask questions like "how much will this effect my score?" The truth is, no one knows the answers to those questions because they can't see into the classifier's algorithm and predict which vectors your credit report will most closely identify with after your report changes (not even a math nerd like me). What we do know is that with every improvement you make on your reports, you begin to look more and more like the borrowers that lenders want to loan money to and your scores will reflect that. However, a fairly good thing to keep in mind is that if you have several baddies on your report, your scores will likely climb more slowly at first because you will still look quite a bit like other borrowers that also have low scores (is there really much difference between 17 baddies and 18 baddies? - lenders think not). But as your report begins to clean up and more tradelines drop off, you will likely start to see some steady climbs in scoring. This is also why those of you whose scores start to get mixed in with the higher scores (high 700s and 800s), will have difficulty separating yourselves from others who also have high scores and your scoring increases begin to slow back down. You can think of it like a bell curve, where most of the elevation occurs in the middle and the two tails on the ends don't see much change in elevation (though this is more of a visual explanation than a correct usage of the bell curve). To really separate yourself from the pack of others with high scores, you will likely have to qualify for some rather impressive loans (and be responsible with them of course).

 

 

Cliff Notes:

If you walk away from this post learning one thing, let it be this: The TLs on your credit report are not assigned a particular value or weight that causes increases or decreases in your score when changes are made. The increases and decreases to your scores come from how closely your credit report resembles other reports after these changes. For those of you aiming to maximize your scores, you should pay close attention to the statistics that are printed in your CRs - the ones that tell you that "High Achievers" carry a 7% revolving balance to available credit limit or that "High Achievers" have no collections on their CRs. You want to look like "High Achievers" as much as possible. Remember, maximizing your credit score is not the same thing as maximizing financial responsibility. All other things being equal (and zero balance), the borrower who carries a revolving balance of 7% on his CCs is going to have a higher score than the borrower who pays his CCs in full every month. If your aim is to maximize your scores, then you want to emulate the "high achievers" as much as possible.

 

 

 

This is interesting. It makes me wonder if "FICO" is punitive indirectly upon your score if you have credit with "low brow/subprime" lenders even if you have good payment history. For example...using your description above, if I have a subprime card with a $1000.00 liimit, make good use of it and pay it off every month vs someone that has a good prime card with a $1000.00 limit who has the same use and payment history with all other credit report items being the same would the prime card holders score be higher because their report is "more" similar to those who hold prime cards and most likely have higher scores? I wonder if the calculation is/could be tied to lenders i.e. being with some lenders would keep you in the higher end "blocks" and therefore give you a higher score? Or if it could just be based on limits...I have a $500 limit and even though I use it and pay it I am penalized becasue I am grouped with other $500 limit holders who may not be so diligent in their payments. The whole thing kinda reeks of an Animal Farm / Oligarachy situation... "All animals are equal but some animals are more equal than others"

 

Makes me want to go and cancel my subprime CapOne card....

Share this post


Link to post
Share on other sites

This is interesting. It makes me wonder if "FICO" is punitive indirectly upon your score if you have credit with "low brow/subprime" lenders even if you have good payment history. For example...using your description above, if I have a subprime card with a $1000.00 liimit, make good use of it and pay it off every month vs someone that has a good prime card with a $1000.00 limit who has the same use and payment history with all other credit report items being the same would the prime card holders score be higher because their report is "more" similar to those who hold prime cards and most likely have higher scores? I wonder if the calculation is/could be tied to lenders i.e. being with some lenders would keep you in the higher end "blocks" and therefore give you a higher score? Or if it could just be based on limits...I have a $500 limit and even though I use it and pay it I am penalized becasue I am grouped with other $500 limit holders who may not be so diligent in their payments. The whole thing kinda reeks of an Animal Farm / Oligarachy situation... "All animals are equal but some animals are more equal than others"

 

Makes me want to go and cancel my subprime CapOne card....

 

This is certainly possible from the perspective of the algorithm. In order to answer that question with confidence however, I would have to know with certainty that the different cards are being reported differently on your CRs, e.g. - our C1 secured cards are being reported by a company named "Capital One Secured" and our prime borrower C1 cards are being reported by a separate company named "Capital One Venture". I don't know if this is the case or not though. It is quite possible (and perhaps someone here can verify this) that these cards are all being reported simply as "Capital One". However, even if it is the case that they are being reported as separate named companies, which could effectively provide another feature (or variable) for the learning algorithm, it doesn't necessarily follow that it will consider this information to be of value (and thus would be excluded from the scoring algorithm). In fact, I would guess that it would discard this information as it is likely to be highly correlated with other negative information found in the TL. This is another way of saying the algorithm would be receiving redundant information on the same TL and penalizing you twice for it, which is something the algorithm desperately tries to avoid for mathematical reasons. Highly correlated data in machine learning translates into poor predictive power for the classifier. The algorithm would aim to avoid this at all costs as predictive power is their number one goal. That being said, I cannot say with 100% confidence that the name of a company will in fact be highly correlated with the other data. It may be the case that if you take a loan with a sub-prime company, that your scores are reflective of being associated with others who also borrow from that company. Another problem with this is that if the algorithm grouped borrowers by who they borrow money from, this could be seen as a binning technique that aims to segregate people geographically as there are many lenders that only serve small geographic regions. I believe there are laws against the practice of using zip codes as part of the learning algorithm, just like there are laws against using gender or race even though the statistical algorithm would find them to have highly predictive power. These algorithms must be "supervised" by statisticians so that they follow certain laws that are in place. It used to be the case that even the color of your skin was calculated into the scoring algorithms. If you've ever heard of the welcome wagon ladies from 50+ years ago, this was a big part of their job. They would come to "welcome" you to the neighborhood and knock on your door bringing you cookies etc. However, what they were really there for was to write down information about your family (single mom? messy house? skin color? etc.) All these factors went into scoring your credit. We've come a long way since then and have put many protective measures in place to ensure that all people are given an equal chance, even if it costs the lenders predictive power.

Share this post


Link to post
Share on other sites

Dear Big Brother,

 

I didn't mean to offend anyone by titling the 3rd paragraph "Sarah Palin Explanation". However, when you edited my post, you left my explanatory section with a nonsensical title. This reflects negatively on my writing style as it now appears as though I title my sections with simple nouns. This would be tantamount to writing an article and titling one of your sections as "Tree" or "Dog" rather than "Oak Trees" or "Wild Dogs". Could you please rename the section now titled "Explanation" to something like "Simple Explanation" or "Layman's Terms"? I would appreciate it greatly.

 

Thanks,

Little Brother

 

PS - I find it humorous (perhaps ironic) that you took offense to the "Sarah Palin" title, but found nothing wrong with calling the mathematically inclined "nerds".

 

Note: If you now decide to change the "Math Nerd Explanation" to remove the word "Nerd" from the title, I would also appreciate it if you changed that title to "Mathematical Explanation" as "Math Explanation" would again be nonsensical.

 

Double thanks,

Little Brother

Edited by donkeyshark

Share this post


Link to post
Share on other sites

This is interesting. It makes me wonder if "FICO" is punitive indirectly upon your score if you have credit with "low brow/subprime" lenders even if you have good payment history. For example...using your description above, if I have a subprime card with a $1000.00 liimit, make good use of it and pay it off every month vs someone that has a good prime card with a $1000.00 limit who has the same use and payment history with all other credit report items being the same would the prime card holders score be higher because their report is "more" similar to those who hold prime cards and most likely have higher scores? I wonder if the calculation is/could be tied to lenders i.e. being with some lenders would keep you in the higher end "blocks" and therefore give you a higher score? Or if it could just be based on limits...I have a $500 limit and even though I use it and pay it I am penalized becasue I am grouped with other $500 limit holders who may not be so diligent in their payments. The whole thing kinda reeks of an Animal Farm / Oligarachy situation... "All animals are equal but some animals are more equal than others"

 

Makes me want to go and cancel my subprime CapOne card....

 

This is certainly possible from the perspective of the algorithm. In order to answer that question with confidence however, I would have to know with certainty that the different cards are being reported differently on your CRs, e.g. - our C1 secured cards are being reported by a company named "Capital One Secured" and our prime borrower C1 cards are being reported by a separate company named "Capital One Venture". I don't know if this is the case or not though. It is quite possible (and perhaps someone here can verify this) that these cards are all being reported simply as "Capital One". However, even if it is the case that they are being reported as separate named companies, which could effectively provide another feature (or variable) for the learning algorithm, it doesn't necessarily follow that it will consider this information to be of value (and thus would be excluded from the scoring algorithm). In fact, I would guess that it would discard this information as it is likely to be highly correlated with other negative information found in the TL. This is another way of saying the algorithm would be receiving redundant information on the same TL and penalizing you twice for it, which is something the algorithm desperately tries to avoid for mathematical reasons. Highly correlated data in machine learning translates into poor predictive power for the classifier. The algorithm would aim to avoid this at all costs as predictive power is their number one goal. That being said, I cannot say with 100% confidence that the name of a company will in fact be highly correlated with the other data. It may be the case that if you take a loan with a sub-prime company, that your scores are reflective of being associated with others who also borrow from that company. Another problem with this is that if the algorithm grouped borrowers by who they borrow money from, this could be seen as a binning technique that aims to segregate people geographically as there are many lenders that only serve small geographic regions. I believe there are laws against the practice of using zip codes as part of the learning algorithm, just like there are laws against using gender or race even though the statistical algorithm would find them to have highly predictive power. These algorithms must be "supervised" by statisticians so that they follow certain laws that are in place. It used to be the case that even the color of your skin was calculated into the scoring algorithms. If you've ever heard of the welcome wagon ladies from 50+ years ago, this was a big part of their job. They would come to "welcome" you to the neighborhood and knock on your door bringing you cookies etc. However, what they were really there for was to write down information about your family (single mom? messy house? skin color? etc.) All these factors went into scoring your credit. We've come a long way since then and have put many protective measures in place to ensure that all people are given an equal chance, even if it costs the lenders predictive power.

 

It is not illegal for credit card comapnies to use data about where and how you use your credit card to make decisons about credit limits, clis, and even card cancelation. I saw something last week...cant remember where maybe msnbc or reuters...that AMEX is now using behavioral scoring to restrict certain types of consummers (those that shop at discount stores, live in certain zip codes,...) from getting or even keeping their cards. The article said that this practrice will become more and more prevalent and have a greater impact on card holders... If credit card companies can use this type of data I cant see the evil empire (FICO) from making use of it in their scoring algorithms. I would love to see a fico compare between two exact credit report holders for two months where one holder does nothing but live the good life (fly first class, top notch dining, and shop at high end stores...think cartier...) and the other low brows it (Greyhound, white castle (they ahve these where I live!!!), pawn shops...). I feel fairlt confident that all factors considered equal that the high end card holders score will be higher at the end of the 2 month experiment.

Share this post


Link to post
Share on other sites

Dear Big Brother,

 

I didn't mean to offend anyone by titling the 3rd paragraph "Sarah Palin Explanation". However, when you edited my post, you left my explanatory section with a nonsensical title. This reflects negatively on my writing style as it now appears as though I title my sections with simple nouns. This would be tantamount to writing an article and titling one of your sections as "Tree" or "Dog" rather than "Oak Trees" or "Wild Dogs". Could you please rename the section now titled "Explanation" to something like "Simple Explanation" or "Layman's Terms"? I would appreciate it greatly.

 

Thanks,

Little Brother

 

PS - I find it humorous (perhaps ironic) that you took offense to the "Sarah Palin" title, but found nothing wrong with calling the mathematically inclined "nerds".

 

Note: If you now decide to change the "Math Nerd Explanation" to remove the word "Nerd" from the title, I would also appreciate it if you changed that title to "Mathematical Explanation" as "Math Explanation" would again be nonsensical.

 

Double thanks,

Little Brother

Read the TOS and perhaps you will see why nerds is not a violation and denegrating a political figure is. Just my 2 pesos

Share this post


Link to post
Share on other sites

Paragraphs are wonderful inventions.

Not much missed in reading every other line :lol:

Or just the opening and closing ones. :D

Share this post


Link to post
Share on other sites

Dear Big Brother,

 

I didn't mean to offend anyone by titling the 3rd paragraph "Sarah Palin Explanation". However, when you edited my post, you left my explanatory section with a nonsensical title. This reflects negatively on my writing style as it now appears as though I title my sections with simple nouns. This would be tantamount to writing an article and titling one of your sections as "Tree" or "Dog" rather than "Oak Trees" or "Wild Dogs". Could you please rename the section now titled "Explanation" to something like "Simple Explanation" or "Layman's Terms"? I would appreciate it greatly.

 

Thanks,

Little Brother

 

PS - I find it humorous (perhaps ironic) that you took offense to the "Sarah Palin" title, but found nothing wrong with calling the mathematically inclined "nerds".

 

Note: If you now decide to change the "Math Nerd Explanation" to remove the word "Nerd" from the title, I would also appreciate it if you changed that title to "Mathematical Explanation" as "Math Explanation" would again be nonsensical.

 

Double thanks,

Little Brother

 

 

No politics on CB - it's in the TOS.

 

All admins and mods have active, working PM's. Don't "get up in our faces" in public, please. Feel free to PM if you have a question.

Share this post


Link to post
Share on other sites

Dear Big Brother,

 

I didn't mean to offend anyone by titling the 3rd paragraph "Sarah Palin Explanation". However, when you edited my post, you left my explanatory section with a nonsensical title. This reflects negatively on my writing style as it now appears as though I title my sections with simple nouns. This would be tantamount to writing an article and titling one of your sections as "Tree" or "Dog" rather than "Oak Trees" or "Wild Dogs". Could you please rename the section now titled "Explanation" to something like "Simple Explanation" or "Layman's Terms"? I would appreciate it greatly.

 

Thanks,

Little Brother

 

PS - I find it humorous (perhaps ironic) that you took offense to the "Sarah Palin" title, but found nothing wrong with calling the mathematically inclined "nerds".

 

Note: If you now decide to change the "Math Nerd Explanation" to remove the word "Nerd" from the title, I would also appreciate it if you changed that title to "Mathematical Explanation" as "Math Explanation" would again be nonsensical.

 

Double thanks,

Little Brother

 

 

No politics on CB - it's in the TOS.

 

All admins and mods have active, working PM's. Don't "get up in our faces" in public, please. Feel free to PM if you have a question.

 

No worries, my bad. I didn't mean to ruffle any feathers. Feel free to delete my big brother post. It was only meant to be light-hearted. You guys all work hard here for the benefit of others and all your hard work is greatly appreciated. It wasn't my intention to flame anyone. I just got a little chuckle out of the fact that "Sarah Palin" was removed but "Math Nerd" held its ground. I'll avoid political comments in the future.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  




  • Today's Birthdays

    No users celebrating today.
  • Member Statistics

    • Total Members
      177,739
    • Most Online
      1,528

    Newest Member
    mikeh714
    Joined

About Us

Since 2003, creditboards.com has helped thousands of people repair their credit, force abusive collection agents to follow the law, ensure proper reporting by credit reporting agencies, and provided financial education to help avoid the pitfalls that can lead to negative tradelines.
×
×
  • Create New...

Important Information

Guidelines