
Originally posted by DomeBase
if it sold for 10k then, yes, there is a 100% probability that it can sell for 10k because it just did
If you flip a coin and it gets heads there is STILL only a 50% probability (ignoring the edge) that it *can* be heads even though it just landed. Sorry Domey  your quote does not make mathematical sense as snoopy pointed out.
DomainView.co.uk  60 day droplists. Hosted chasing. Backorders. Thousands of domains caught.
DomainOffer.co.uk  powerful domain parking and one page website hosting

if it sold for 10k then, yes, there is a 100% probability that it can sell for 10k because it just did
to add to what rob said:
there is only a 100% probability that it *did* sell for 10k
there is not a 100% probability that it *can/will* sell for 10k.
When using google for counts  use double quotes for usage counts for multiword terms and set "match type" to "exact" for all search volume lookups. Click here for more info

if it sold for 10k then, yes, there is a 100% probability that it can sell for 10k because it just did
We should not discount the degree of probability which must be given its due in this matter. That is, is it a "low probability or high probability".

We are not communicating here.
The value of something can be viewed as a single number, as an expectation (prob x outcome) or as a random variable distribution
The most accurate way to view it is as a random variable distribution. This is just plain old statistics.
The question at hand here is whether you revise the distribution (mean, variance, whatever parameters you wish) when you receive new information.
This is not new stuff. It is called Bayesian statistics. If you revise the distribution when you get new info, then you are behaving in a Bayesian manner. When the distribution is revised, it may affect the expected mean, variance, or both. This goes for coin flips, houses, sun spots, whatever.
The question here is whether you revise the expected probability distribuition of a random variable (e.g. domain value) based on new information or not. You do not have to set it equal to the new data point, but you should revise your distribution to some extent. If you ignore new data and hold to your original distribution, then that is your perogative... but I think it is odd to ignore new information.
Let me put this another way. What started this is that someone asked for appraisals on a domain. Someone else posted an appraised value of $###. The original poster then said that it sold for $##,###. Rather than view this as new information to adjust the appraisal value, there appears to be a view that the new information has no value and that the original appraisal remains as correct as it was before the new information. If this is correct? If it is, then what would it take for your to change your appraisal? Two sales at $12,000? Four sales at $12,000? Or would they all be wrong as well?
I think that it you cleared away all the fog and got to the heart of the matter, everyone would agree that new information matters. That is all I am saying.

Originally posted by Rob
If you flip a coin and it gets heads there is STILL only a 50% probability (ignoring the edge) that it *can* be heads even though it just landed. Sorry Domey  your quote does not make mathematical sense as snoopy pointed out.
Originally posted by safesys
to add to what rob said:
there is only a 100% probability that it *did* sell for 10k
there is not a 100% probability that it *can/will* sell for 10k.
Yes to both of you.
The next event is a probability distribution.
The last event is a certainty.
The question is whether your revise the probability distribution of the next event based on the evidence of the most recent one or not. Update or ignore.

A story.
John Doe posts a domain for appraisal.
Ed Smith says  "That domain is worth $100."
The next day John Doe reports that the domain sold for $100,000. "See, my domain was worth more than $100," says John.
"No," says Ed, "It is still worth just $100. That $100,000 sale was a fluke and there is no guarantee that it will sell for $100,000 again."
The next day the buyer of the domain sells it again, this time for $120,000.
Again John Doe posts on the forum, "Look, now the domain has sold for $120,000. What is its value?"
"Domain sales are highly volatile," responds Ed, "This might be a fluke as well. Can you guarantee that it will sell again for $120,000?"
"Ah, no, no guarantees," says John, "But don't these two sales mean *something*?"
"No," says Ed, "Unless you can guarantee that it will sell for $120,000, I stand by my original appraisal of $100."
The next day John buys a $100,000 car with the money from his sale.
"Nice car, isn't it?," John asks Ed.
"No," responds Ed, "It is only worth $100."
"How so," asks John in shock.
"Well... your $100,000 was not really worth $100,000 because the domain was only really worth $100, so your car that you bought with the profit from the domain is really only worth $100."
"Is see," says John, "Well... its a darn fine car for $100 then."

but I think it is odd to ignore new information.
Not suggesting it be ignored, but rather it only be used in relation to being one statistic used for the market valuation model  the model itself then applied back to valuing the domain  this way outliers are removed rather than being given undue weight.
When using google for counts  use double quotes for usage counts for multiword terms and set "match type" to "exact" for all search volume lookups. Click here for more info

Let me put this another way. What started this is that someone asked for appraisals on a domain. Someone else posted an appraised value of $###. The original poster then said that it sold for $##,###. Rather than view this as new information to adjust the appraisal value, there appears to be a view that the new information has no value and that the original appraisal remains as correct as it was before the new information. If this is correct? If it is, then what would it take for your to change your appraisal? Two sales at $12,000? Four sales at $12,000? Or would they all be wrong as well?
Let us attempt to break this down to its simplest terms. Let's suppose that the initial appraisal was for $xxx, based on the reasoned judgement of a domainer on this forum. Absent of the contravening fact that the owner claimed he sold it for $xx,xxx.
The issue to be addressed by all is "was the original appraisal reasonable, given all the various metrics that domainers subscribe to"
If the answer is to the affirmative, then there is really no issue here.
OTOH if it turned out after the initial appraisal that new information was obtained that the domain could be infinging on a TM, then that information can and must be used in the valuation process.
The fact that the owner sold it to a willing buyer for more, does not in and of itself establish the validity of the domain as a benchmark for similar names. It simply can be explained that he found a buyer willing to pay indiscriminately for the domain.
If one has ever been to a vehicle auction sale, you probably have noticed people bid outrageously for cars without engines or bad engines in them.
There are different reasons why this may happen, but a reasonable auctioneer would not use the sale of the first vehicle as a benchmark to sell an identical vehicle.
Last edited by Magick; 20060807 at 02:40 AM.

Originally posted by safesys
Not suggesting it be ignored, but rather it only be used in relation to being one statistic used for the market valuation model  the model itself then applied back to valuing the domain  this way outliers are removed rather than being given undue weight.
yes

Originally posted by Magick
Let us attempt to break this down to its simplest terms. Let's suppose that the initial appraisal was for $xxx, based on the reasoned judgement of a domainer on this forum. Absent of the contravening fact that the owner claimed he sold it for $xx,xxx.
The issue to be addressed by all is "was the original appraisal reasonable, given all the various metrics that domainers subscribe to"
No. That is not the issue. This issue is *not* whether the original appraisal was accurate given information at the time, but rather whether one incorporates new information and updates one's appraisal. I have no problem with the original appraisal. It is ignoring new information that puzzles me.
Originally posted by Magick The fact that the owner sold it to a willing buyer for more, does not in and of itself establish the validity of the domain as a benchmark for similar names.
This is a straw man argument. The issue is not "establishing a benchmark for similar names" (if, indeed, that is even possible given mountains of data) but rather what *this particular domain* is worth, now, given the fact that it sold for $##,###. You either take the sale into consideration, as SafeSys just suggested, or you do not. The choice is yours.

A man with red hair walks into a bar.
Sam and Bill in the bar whisper to each other. "What do you think that guy is like?" asks Sam.
"Looks like a nice enough fella to me," says Bill.
Suddenly the redhaired guy pulls out a gun, gives a blood curdling scream, and kills half the people in the bar.
"Wow... looks like you were wrong about that fella," says Sam.
"Nope, I was right. He is a nice guy," responds Bill.
"How do you figure that?," asks Sam, "He just killed half the people in the bar!"
"Well... I was right based on what we knew before that... " says Bill.
"Oh," Sam pauses, "But what about what happened after that? Don't that matter?"
"Hey," replies Bill with a snort, "Are you saying that what happened here estalishes a basis for saying that all red heads are killers?"
"Ahhh... noooo." says Sam, "I think I'd better have a drink. This is getting wierd."
Last edited by DomeBase; 20060807 at 02:52 AM.

OTOH if it turned out after the initial appraisal that new information was obtained that the domain could be infringing on a TM, then that information can and must be used in the valuation process.
Edited version by magick
I was at a house auction recently, when the bidding started, it got so over bid that afterwards it became obvious that the desperate bidder was hell bent on winning at all cost to make sure his friend didn't lose his property. Keep in mind that the realestate values of other properties was much lower on that street. Therefore, in the valuation of this house and similar ones on the same street, in light of this new information, how should it affect its valuation be it positive or negative.
In other words, going back to the original issue, if the owner sold the domain at $xx,xxx. Do we automatically assign the domain the value of what it was sold at. How do you suppose we use this "new information". Do we recognize it, discount it, or what.

Has it possibly occurred to anyone that Snoopy was incorrect respecting his appraisal. Why is the appraisal Snoopy gave more valid than the suggested valuation CertifiedOfferService.com gives the domain at NetworkSolutions.com? I have yet to see an accurate appraisal myself. Frankly, I don't think it is possible to have an absolute formula, but if I am wrong I am certain you guys will show me......and I say that with all due respect to all of the seasoned members.
Respecting home sales, I can assure you even if a home sells for more that what you think it should the taxing authorities will definitely use that new information when assessing and determining fair market values for the purpose of increasing your property taxes.
As far as domain sales................I have no doubt that Uncle Sam will be acknowledging the value of this domain as he will want his fair share. ()
All in all, I find this thread most interesting.
Cheers,
Shiowcase
Last edited by showcase; 20060807 at 06:18 AM.

Appraisals are opinions not statements of fact  the variables in domain sales transactions vary massively.
It's only when looking at the picture as a whole that you get to see distribution curves (complete with outliers) that can be used to look at what is probable.
When using google for counts  use double quotes for usage counts for multiword terms and set "match type" to "exact" for all search volume lookups. Click here for more info

Originally posted by showcase Has is possibly occurred to anyone that Snoopy was incorrect respecting his appraisal........ ............... Respecting home sales, I can assure you even if a home sells for more that what you think it should the taxing authorities will definitely use that new information when assessing and determining fair market values for the purpose of increasing your property taxes. Showcase
Little doubt Snoopy was incorrect on his definition of value. IMO, the probability of a sale really has a low relationship to value since the end user or buyer may take forever to realize the name is available for sale thus delaying the probability of sale, with much of the sale and its timing based on luck.
Good point about real estate sale price and tax and absolutely right information. In addition, once the sale closes that becomes your #1 comp in other appraisals since it is the most recent indication of value involving similar property.

Good point about real estate sale price and tax and absolutely right information. In addition, once the sale closes that becomes your #1 comp in other appraisals since it is the most recent indication of value involving similar property.
That is also the reason why most counties allow for the challenge of assessments.

IMO, the probability of a sale really has a low relationship to value since the end user or buyer may take forever to realize the name is available for sale thus delaying the probability of sale, with much of the sale and its timing based on luck.
Surely buyers realizing the name is for sale (either proactively or reactively) is a part of the probability of the name selling.
Good point about real estate sale price and tax and absolutely right information. In addition, once the sale closes that becomes your #1 comp in other appraisals since it is the most recent indication of value involving similar property.
Again, most of these house transactions include a market valuation before they can complete (to get the mortgage). Valuers look at overall sales patterns not just individual sales  that's how they lessen the effect of outliers.
When using google for counts  use double quotes for usage counts for multiword terms and set "match type" to "exact" for all search volume lookups. Click here for more info
Posting Permissions
 You may not post new threads
 You may not post replies
 You may not post attachments
 You may not edit your posts
Forum Rules


Sponsors
