why does a member have almost 400 neg feedbacks in a month and have 98%

[bob2839]

i get some idiot that after 3 months decides his item didnt work when it arrived ,especially when i had it professionally packed in foam,gives neg after i refuse to refund, so after  700 dealings with a  perfect 100% rating i am now 98% because of this fool decided he didnt want it and wanted a refund ,and does ebay look after me ?, not likely very disapointed    bob2839

[*tippy*toes*]

That seller has over 900,000 feedback. You have 716. As Joe mentioned, they have received nearly 300,000 feedback in the last 12 months, you have received 86. When it comes to percentage, they are about the same.

Your feedback page will tell you how that number is calculated if you hover over the link under your username. This one is what shows for you:

And this is how they worked the percentage out for the high volume seller:

[dazzledayz]

I have a favourite quote from Mark Twain that fits this situation perfectly:

"There are lies, and tthere are damned lies, but then there are statistics"

Bob is quite correct in what he says inasmuch as he has upset 1 customer whereas the large seller has upset over 7000

(4036 neg and 2973 neut)

A simple statistic often does not tell the whole story but is a whole lot simpler to encode into an algorithm.

[davewil1964]

---

@dazzledayz wrote:

I have a favourite quote from Mark Twain that fits this situation perfectly:

"There are lies, and tthere are damned lies, but then there are statistics"

Bob is quite correct in what he says inasmuch as he has upset 1 customer whereas the large seller has upset over 7000

(4036 neg and 2973 neut)

A simple statistic often does not tell the whole story but is a whole lot simpler to encode into an algorithm.

---

Benjamin Disraeli, actually

[dazzledayz]

Or possibly Sir Charles Dilke?

http://www.york.ac.uk/depts/maths/histstat/lies.htm

It seems the attribution isreally  quite uncertain.

[joethenuts]

you can also say   seller b  has 23.000 happy buyers   and seller a  only has 80  happy buyers  mmmm   at the end they both have the aprox the same % of happy buyers

[springyzone]

I think everyone is right, in a way.

It's perfectly true that a small seller can be hit in the percentage a lot more by just one stroppy or unreasonable buyer, it can make them look 'worse' than another seller who has a bigger turnover.

But is the % the be all and end all, that's the question.

I suppose everyone has their own way of approaching a purchase. The way I go about it is if a % looks a bit low, I go straight to the negatives and read those, to try to get a feel for what went wrong. Sometimes you can just see the buyer was a a fruit loop. (sorry to put it so rudely, my mind's a blank at the moment for a politically correct term.)

If I see a % dragged down by that sort of thing, I take no notice, I have a look at the positives.

I am more inclined to buy from an aussie seller or smaller business where I can.

On the other hand, if someone is a big seller, has a huge turnover, I tend to be suspicious they are selling cheap junk. (Yes, I know, that is unfair but anyway....) and what i do is I busily scroll the negs to see if I can find someone with a comment for the same item I am looking at, especially for any comments re quality.

And that's the case even if their FB says 99.9999%

[cathyjoyjoy]

And this is exactly why it is important for small sellers to name the buyer ID within the response to feedback and ensure the response is professional.  Response to negative or neutral feedback is more for the information of future buyers and other sellers.

[gec2002]

While FB is based on % it doesn't necessarily tell the whole story or even a particularly accurate story at low numbers, due to the Law of Large Numbers ie the bigger the sample size the closer the result is to the correct answer..  Eg if the seller only had 4 FB and gets 1 neg they are at 75% however their next 96 FB maybe positive giving them a score of 99%.  However someone with 290000 FB are close to their true statistical score as small aberrations have little effect on their FB  ie 290989 +ve 4036 -ve 98.6% they unfortunately get a very nasty buyer who leaves 100 -ve their FB remains unchanged ie 290989 =  4136 -ve = 98.6%.  So this is close to their true FB value, whereas with our OP their score is almost meaningless, their next 500 FB might all be positive or littered with negs.  However any negatives will have a disproportionately large effect on their score.  Feedback at low numbers is a poor indicator of seller reliability, just another reason to scrap it.

[dazzledayz]

I can't argue your stats argument which is spot on.

But one would wonder why a large seller with presumably years of experience is allowed to continue disaffecting such a large cohort.

[*tippy*toes*]

I suppose when you're selling close to 300,000 things a year, 400 a month doesn't seem that much. A high volume seller like that is probably protected anyway, so they could upset 4,000 a month and nothing would happen.