Anyone good with patterns (calculations?) here?

[Lord Antares]

For anyone who has a little bit of time to spend and likes a challenge, I would be thankful if you went to:

 

https://member.expireddomains.net/domains/combinedexpired/?start=0

 

And took a good look at the results. Make sure there are no filters and they are sorted by ''dropped''. You are looking at domains which have most recently expired. We want to see if there is pattern to expiration of those domains and maybe, be able to estimate when a certain domain would expire given its name or whatever other metric might be relevant.

 

It's a longshot and probably impossible to figure out, but feel free to give it a go. Thanks.

[Raider5678]

You have to log in.

[Sensei]

This website requires registration and logging..

Better copy'n'paste results, or make screen-shot, to not bother people too much.

I would expect rarely updated abandoned websites expiring the most frequently. So algorithm should collect the all registered domains in database, visit them e.g. one time per day (the all billions of websites!), check whether they were changed (it might be difficult, if website is dynamically generated). If it's all the time the same content, entire year, it has the highest chance to expire (because it has the highest chance to be abandoned).

Edited December 18, 2018 by Sensei

[Lord Antares]

6 minutes ago, Raider5678 said:

You have to log in.

 

3 minutes ago, Sensei said:

This website requires registration and logging..

Better copy'n'paste results, or make screen-shot, to not bother people too much.

 

Apologies, I forgot. There are millions upon millions of domains there and I don't think anyone could find a pattern on the sample size of just 50. Registering takes 10 seconds tho and you don't need to confirm the mail. Sorry about that tho.

 

5 minutes ago, Sensei said:

I would expect rarely updated abandoned websites expiring the most frequently. So algorithm should collect the all registered domains, visit them e.g. one time per day (the all billions websites!), check whether they were changed (it might be difficult, if website is dynamically generated). If it's all the time the same content, it has the highest chance to expire (because it has the highest chance to be abandoned).

Nonono, you misunderstood. This is about domains which have already expired. All of the domains on that website are already expired. But expired does not equal deleted. Once it's expired, it goes through several stages. The last one is the pending delete stage. In that stage, it cannot be renewed by the owner and will be released to the public registry in 5 days. We want that. Release to public registry.

The domains drop in a 3 hour window each day. I wanted to know if there was a pattern to their dropping. Again, all of these domains are already bound to drop, it's just a question of when.
I doubt it's practically possible to find out as it's probably hidden behind a good RNG, but maybe not. I do notice some weirdness. Like some minutes will have a lot of domains dropping, then a pause. Some minutes will only have one domain dropping etc. Also, I find that a lot of the domains dropping in the same minute are in the same TLD (domain extension, like .org for example). It just might be possible to figure something out.

Here's an example:

Weird, no?

 

[Raider5678]

22 minutes ago, Lord Antares said:

Nonono, you misunderstood. This is about domains which have already expired. All of the domains on that website are already expired. But expired does not equal deleted. Once it's expired, it goes through several stages. The last one is the pending delete stage. In that stage, it cannot be renewed by the owner and will be released to the public registry in 5 days. We want that. Release to public registry.

May inquire as to the reason you want to know?

[Lord Antares]

7 hours ago, Raider5678 said:

May inquire as to the reason you want to know?

For a good purpose, trust me.

[John Cuthber]

6 hours ago, Lord Antares said:

For a good purpose, trust me.

Your view of "good" and mine (or Donald Trump's) may not tally

[Lord Antares]

3 hours ago, John Cuthber said:

Your view of "good" and mine (or Donald Trump's) may not tally

What does it matter anyway? The solution to the ''puzzle'' has nothing to do with why I want to know.

[John Cuthber]

2 hours ago, Lord Antares said:

What does it matter anyway? The solution to the ''puzzle'' has nothing to do with why I want to know.

Really? You had to ask?

Our willingness to help would depend on that.

[StringJunky]

2 hours ago, Lord Antares said:

What does it matter anyway? The solution to the ''puzzle'' has nothing to do with why I want to know.

When people ask for help then you go all secretive is annoying and a bit insulting because it can make people feel used. This site is about curiosity and you are stamping on it. You can't ask questions without being asked questions; that's the deal.

Edited December 18, 2018 by StringJunky

[Lord Antares]

9 minutes ago, John Cuthber said:

Really? You had to ask?

Our willingness to help would depend on that.

 

2 minutes ago, StringJunky said:

When people ask for help then you go all secretive is annoying and a bit insulting because it can make people feel used. This site is about curiosity and you are stamping on it.

Alright then. There are thousands of people who lurk dropping domains each day and even more trade domain names. Good and valuable domains drop frequently, even every day. Occasionally, even 6 figure domains can drop. All of the good ones are being caught by ''dropcatchers'' and then auctioned off to investors. I won't bother explaining how all of that works because it's not really relevant.

I looked into it and investigated and I know exactly how dropcatching is done and I could do it in theory, but I would need a very large (read: impossible) investment to be able to compete with the others realistically. If I (or you) could figure out a pattern and know or be able to estimate when the names will drop, it would be a lot easier and cheaper and the ''randomness'' with which these domains drop seems really weird. 

But all of this is extremely far fetched, as it's probably highly unlikely that this would turn out to be even remotely simple. Besides, it would be useless if everyone knew how to do it which is why I didn't want to say anything.

[StringJunky]

10 minutes ago, Lord Antares said:

 

Alright then. There are thousands of people who lurk dropping domains each day and even more trade domain names. Good and valuable domains drop frequently, even every day. Occasionally, even 6 figure domains can drop. All of the good ones are being caught by ''dropcatchers'' and then auctioned off to investors. I won't bother explaining how all of that works because it's not really relevant.

I looked into it and investigated and I know exactly how dropcatching is done and I could do it in theory, but I would need a very large (read: impossible) investment to be able to compete with the others realistically. If I (or you) could figure out a pattern and know or be able to estimate when the names will drop, it would be a lot easier and cheaper and the ''randomness'' with which these domains drop seems really weird. 

But all of this is extremely far fetched, as it's probably highly unlikely that this would turn out to be even remotely simple. Besides, it would be useless if everyone knew how to do it which is why I didn't want to say anything.

Cheers for explaining. I'm no mathematician but wouldn't it be like catching the stock market? I think you need to be very aware of the health of companies, track their decline in real time by some consistent parameter, then try and predict that endpoint. You'd need a massive database to follow and compute it all.

Edited December 18, 2018 by StringJunky

[Lord Antares]

1 minute ago, StringJunky said:

Cheers for explaining. I'm no mathematician but wouldn't it be like catching the stock market? I think you need to be very aware of the health of companies, track their decline in real time by some consistent parameter then try and predict that endpoint. You'd need a massive database to follow and compute it all.

I don't really understand, or rather, I don't think you understand.

You can easily see which domains names are due to drop in the coming days. The link I gave in the OP is of the domains which are gone and deleted from the database, free for registration. Dropcatchers will want to register them as soon as they hit that stage.

But there is another section called ''pending delete'' domains. These are domains which WILL drop in the next 5 days max and the previous owner can't do anything to save them. They will expire and someone will catch them if they are good.

Before that, they are in the grace deletion period where they are expired and can still be renewed by the original owner but are unlikely to be, because the owner wouldn't have let them expire in the first place.

Some names that are due to drop in the following days:

- securtyonline.com

- businessnow.com

- megastore.net

- esop.com

etc etc. All of these are worth thousands or tens of thousands of dollars and they drop all the time. I personally witnessed the drop of sushi.com and the ending bid was something like 350k.

This has nothing to do with tracking the market or being aware of companies. It's just about being able to register them the fastest after they drop.

Due to the fact that there's such money involved, I find it highly unlikely that anyone would  be able to figure this out and the chance wasn't worth the time typing this, but there you go.

[StringJunky]

2 minutes ago, Lord Antares said:

I don't really understand, or rather, I don't think you understand.

You can easily see which domains names are due to drop in the coming days. The link I gave in the OP is of the domains which are gone and deleted from the database, free for registration. Dropcatchers will want to register them as soon as they hit that stage.

But there is another section called ''pending delete'' domains. These are domains which WILL drop in the next 5 days max and the previous owner can't do anything to save them. They will expire and someone will catch them if they are good.

Before that, they are in the grace deletion period where they are expired and can still be renewed by the original owner but are unlikely to be, because the owner wouldn't have let them expire in the first place.

Some names that are due to drop in the following days:

- securtyonline.com

- businessnow.com

- megastore.net

- esop.com

etc etc. All of these are worth thousands or tens of thousands of dollars and they drop all the time. I personally witnessed the drop of sushi.com and the ending bid was something like 350k.

This has nothing to do with tracking the market or being aware of companies. It's just about being able to register them the fastest after they drop.

Due to the fact that there's such money involved, I find it highly unlikely that anyone would  be able to figure this out and the chance wasn't worth the time typing this, but there you go.

Yes, I've since read a bit and see what you mean. Can't see you beating the big boys.

[Lord Antares]

A large investment is needed unless a pattern can be deduced. I'm just looking to see if anyone could figure out a pattern for the dropping names from the first link. If that was figured out, then it's easy and I could get all the necessary software made.

[koti]

1 hour ago, Lord Antares said:

A large investment is needed unless a pattern can be deduced. I'm just looking to see if anyone could figure out a pattern for the dropping names from the first link. If that was figured out, then it's easy and I could get all the necessary software made.

What makes you think there is a pattern ?

 

[Lord Antares]

2 minutes ago, koti said:

What makes you think there is a pattern ?

 

 

22 hours ago, Lord Antares said:

I do notice some weirdness. Like some minutes will have a lot of domains dropping, then a pause. Some minutes will only have one domain dropping etc. Also, I find that a lot of the domains dropping in the same minute are in the same TLD (domain extension, like .org for example). It just might be possible to figure something out.

But maybe not.

[koti]

11 minutes ago, Lord Antares said:

 

But maybe not.

There might be patterns caused by factors like database refreshing or some other technology based glitch. I hope you're looking for that kind of pattern and not for market based one because if you are looking for the latter, you might be on your way to crackpot land...be careful  

Edited December 19, 2018 by koti

[Lord Antares]

4 minutes ago, koti said:

There might be patterns caused by factors like database refreshing or some other technology based glitch. I hope you're looking for that kind of pattern and not for market based one because if you are looking for the latter, you might be on your way to crackpot land...be careful  

The site says it updates every minute but it's possible.

[John Cuthber]

OK,  Thanks for the explanation. 

I must be missing something.

Imagine I work out some pattern.

Imagine that I somehow know that "reallyValuableName.com" has a 99.999 % chance of dropping out of use at noon on some particular day- let's say some time in the new year to give us time to act.

I might want to register it- but so will my competitors.

Like me, they can see "what's coming up" and, unlike me, they have enough money to win a bidding war.
How can I hope to win?
 

[Lord Antares]

On 12/19/2018 at 1:47 PM, John Cuthber said:

OK,  Thanks for the explanation. 

I must be missing something.

Imagine I work out some pattern.

Imagine that I somehow know that "reallyValuableName.com" has a 99.999 % chance of dropping out of use at noon on some particular day- let's say some time in the new year to give us time to act.

I might want to register it- but so will my competitors.

Like me, they can see "what's coming up" and, unlike me, they have enough money to win a bidding war.
How can I hope to win?
 

They won't know when it drops. It's a game of RNG where every dropcatcher will send many API reuqests throughout a 3 hour window (which is the window when domains drop each day), instead of a one minute window or whatever if you know when the name drops.

I have foregone the idea tho. I was crazy to think that there's even a low chance of finding out even though there are some weird things happening in the pattern. I have earned a bit and will keep earning a bit from expired domains but not in this way, I don't imagine.

I have talked to programmers who can make dropcatchers (none of them know what it was beforehand) but it's only worthwhile to go for much less competitive names. There's a guy who earns $125k+ per month offering backorders for relatively uncompetitive names so there's definitely a market there.

[wtf]

On 12/17/2018 at 6:13 PM, Lord Antares said:

Apologies, I forgot. There are millions upon millions of domains there and I don't think anyone could find a pattern on the sample size of just 50.

What kind of patterns are you looking for? It would help if you were specific. 

Remember that there are many domain registrars out there, and they each have their own policies for how they handle expired domains. 

Rather than look at the data 50 at a time, if I had this problem I would write a scraper to hit this site and collect a database of thousands or (if project has a serious budget, for example if I were doing this for a company) millions of records, along with the registrar, the date/time, the phase of the moon, whatever else you think might be of interest. Then I'd datamine the heck out of it, looking for ... whatever it is I'm looking for.

What exactly are you looking for? 

[Lord Antares]

14 hours ago, wtf said:

Remember that there are many domain registrars out there, and they each have their own policies for how they handle expired domains. 

They all need to follow some certain rules assigned by ICANN. While each registrar can handle expired domains differently, there is a limit to what they can do. Also, the fact that there is a fixed 3-hour window where all domains (from all registrars) drop implies to me that registrars have no influence over this. The owner of expireddomains.net is on a forum I'm on, I might ask him if he knows more about this. He would probably be the right person to ask.

14 hours ago, wtf said:

What kind of patterns are you looking for? It would help if you were specific. 

I don't know. I'm no coder. I assumed there might be a way to code something such that it can detect deviations from ''expected randomness'', if that makes sense. If you look over that site, the domains seem to drop very weirdly. The same TLDs tend to drop in a bunch and some minutes have hundreds of drops, while some others have none. This is under the assumption that the information on that site is correct. I might look over the registrars over those dropped names to see if there's anything there.

14 hours ago, wtf said:

Rather than look at the data 50 at a time, if I had this problem I would write a scraper to hit this site and collect a database of thousands or (if project has a serious budget, for example if I were doing this for a company) millions of records, along with the registrar, the date/time, the phase of the moon, whatever else you think might be of interest.

Why is millions so much more expensive than thousands? I would imagine that since the code is that, and it does the same thing in higher quantities, it wouldn't cost much more.

I have been researching more about this and I dropped (no pun intended) the idea of dropcatching expensive names. In the end, it all boils down to the fact that the big players like dropcatch.com wouldn't have spent millions on their software if it wasn't absolutely necessary. However, the big dropcatchers have no options for backordering .io, .co.uk, .co and some other domains which still have value. As I said, a guy who offers dropcatching services for .io and some others earns 6 figures per month by doing so and he has next to no competition.