Topic: Bets of Bitcoin - Bitcoin betting on real world events - page 20. (Read 62149 times)

forgot my password. Anything I can do to recover it?
Thanks

Doh!  Sorry all.  (sucks being OCD sometimes)


Ya, this one had a few different barbs to it.


That's why I was asking how they will be resolved in the future.

Fortunately, so far this bet had a relatively small amount wagered.  So this is the best time to experiment and see what works and what doesn't.  You're doing a fantastic job, and I really enjoy both the topic (a bitcoin-enabled predictions market) and the content itself (the bets I can place, and seeing the progress of bet statements that I've submitted myself)

In fact, it is progressing so well, I'm now concerned about something else -- hacked accounts.  Will there be an option for two-factor authentication (as an option only, of course)?  I see that google two-factor is what GLBSE added recently.
 - http://lwn.net/Articles/470764
 

I did not and will not intentionally change the bet descriptions. In this particular case defining the price of something as the lowest you should pay to get seemed natural to me.
I am sure there will be conflicts like this in the future. If we can not seem to reach a mutual viewpoint, canceling (and possibly re-posting) would be the solution, which I just did for this case.

I have closed the statement, refunded all the bets and the submission fee. Feel free to open a new bet if you wish.

So what happened was the bet was approved without realizing that it violates one of the criteria that is used in evaluating each new bet.  That criteria is that there must be an incentive to bet both sides.  This criteria was used to dismiss the "World will end in 2012" bet because there was no incentive to bet "Agree":
 - https://bitcointalksearch.org/topic/m.756912 (can't stop chuckling about that one)

So in this instance, if the bet can be manipulated easily to guarantee that the outcome for the statement is true (i.e., "Agree" wins) then there is no incentive to bet "Disagree".  And thus the only ones that would do so would be those that don't understand how the manipulation is possible.  To publish that bet would be predatory and I'm happy to see that BetsOfBitco.in did not allow it, in its original form.

The only question was ... in the future, will the details of bets continue to be changed after betting has commenced or will they be cancelled and re-posted after correcting for the flaw?

I am so sorry that you felt being enforced. I emailed you before publishing my interpretation and you've said you are OK either way. If there is so much concern among both statement submitter and the bettors, it is a trivial matter to cancel the statement and open new one without these concerns in the way you want. I am here to help.

I'm not sure.  The problem I see is that when there is an ambiguity in a bet, those betting earlier are disadvantaged then when at a later time that ambiguity gets cleared up.

In this specific instance, the bet was fairly clear the way I read it.  If a trade occurred at 1.10 BTC or higher, the statement would be true.  

The way these "share auctions" work there isn't a single IPO price.  The range of prices will be from the highest bid on down to the price paid for last share that is sold:
 - https://bitcointalksearch.org/topic/--76594

And yes, since even a single share can be bid on with any price, the range used in determining the outcome for this bet can be easily manipulated.

So the bet itself was flawed.  It probably should have been cancelled, the bets returned, and possibly a new bet created without the flaw corrected.   Or to let it continue as-is, without comment or correction and deal with the manipulation issue if one exists at the time the site moderators decide the outcome.

Of course prior to the change the grand total of 1.4 BTC had been bet so if anyone has a real problem with the change, it isn't a big financial deal at stake.

But let's say this is like the real world and there are significant amounts of money riding on this.  If that IPO range ends up being 1.09 BTC to 1.15 with the majority of shares trading at or over 1.10 then your change just made the outcome of the bet to be False.  Any rational person functioning as a site moderator would have determined that the IPO price of 1.10 was definitely reached.  Those who had already placed bets on Agree prior to your change get cheated in this instance, essentially.

I trust the site moderators to be rational at the end of the bet had this conflict occurred.  Whatever decision they come up with would be the best decision based on the information at the time it is made.  That's what I am counting on when I place a bet.

So my point was simply, if a bet is bad enough that it needs clarifying, then it is bad enough to be cancelled.  I don't think this bet reached that threshold of needing to be canceled and thus should have been left as-is, ... in my opinion.

People asked for clarification here at the forum, I told them how I interpret the claim and put the same statements for others to see on the site. What would be the appropriate thing to do in your opinion?

PS: If you are curious total open bets is over 850 BTC, almost 1/10000 of all bitcoins in existence.

But you did that after there already had been wagers placed, right?

I guess part of the challenge to this is in predicting the outcome of the event, and then there's the additional challenge in predicting what material statements coinjedi will provide after the betting has already commenced.

p.s., Would be great to see the site put up a total of all wagers (agree + disagree) that are open at the present time.  (not exact, but in the range of "More than X00 BTCs in NN open bets".

As a clarification I put the "Price is defined as the lowest price paid for the bonds." note on the page. So it is the second definition.

I agree that it is very easy to decide. But if we allow such bets many people will start submitting those kind of bets ("This statement will get more bets than this other one" etc). We would like to keep the site on real world events. There are many events (sports games etc) in a few days. You may want to try those.

We did created many of the early statements, but we do it very rarely these days. So yes we are allowed to put statements.

I am glad that people put thought on the details of the procedure. We basically have two choices. We can either wait until the pre-set event date to be 100% sure, or we can use our best judgment and pay the winners. One day in the future we might be wrong but overall I think this way is more fun and profitable for everyone.

Also please spell out more rules on submitting bets and collect already rejected bets.

I got a bet rejected because "We decided not to have self-referential bets" - (http://betsofbitco.in/item?id=335). There is no mention on the page anywhere that this is against any rule. Actually it's one of the most easy ways to decide if a bet has won or not and I had even more ideas that I can now probably no longer submit ("This statement will have an even amount of Bitcoins as bets" and similar).

Currently lots of bets are long in the future - if I want to bet, I want to see some result in a few hours or days. Not months.

Are you as the operators allowed to put up statements too?


Edit:
Also, http://betsofbitco.in/item?id=337 is far too easy to manipulate if the criteria is "1 share or more sold at 1.1 BTC from the IPO" - having "all shares must be sold at 1.1 BTC or above" would make more sense to me... it is not 100% clear to me which of the 2 is meant - if it is the first, I'd like you to reject the statement.

What if something that makes it incredibly unlikely, but not 0%? The loss to the early bettors is almost the same, but you surely cannot intervene right?

What if Santorum changes his mind or runs under the new party "Christians" after America undergoes a sweeping religious revival?

If Obama falls down a 30 foot well and is unconscious but not yet confirmed dead do you honor bets made after that saying he will not be re-elected? Or gets hit by a truck and is in a coma, when do you pull the plug (on the wagers). What if he picks his nose and flicks it on a reporter?

I'm having a little fun, but I'm also curious and it does seem like an important thing to know.

One of the factors considered when the site reviews newly submitted bets is whether or not there is an incentive to bet on both sides.  I've only seen one report of a bet being rejected for that reason though ("The world will end in 2012" which was rejected for it offered no economic incentive for anyone to bet "Agree".)

So, I'm guessing there has never yet been a bet that made it to expiration with bets sitting on only one side.

But as I recently learned, a bet could be closed if the outcome is known early.

So it is possible that a bet would close with bets on only one side.  The "Hugo Chavez will be re-elected" bet is one such bet:
 - http://betsofbitco.in/item?id=302

If he were to die today, there are no "disagree" bets at the present time.

How would that situation be handled?

One hurdle to most anyone who newly registered an account is that funds need to exist in the account before a new bet entry can be submitted.

However, the submission needs reviewed by an admin before it is accepted.  Why can't submissions then be accepted regardless of balance?  Then the verification that the account has funds occurrs at the same time that the review is performed.  Or perhaps at the time of submission, in addition to looking at the account balance consider any 0/unconfirmed payments as well -- those would in nearly all instances be confirmed by the time that the entry's review occurs at a later time.  And if they didn't confirm, then the entry doesn't get approved.

To improve participation levels ... try to remove the reasons people don't participate.

Here is what happened. When I saw the news I intended to close the bet by setting the deadline to the previous day but put the wrong date (4/10). Couple hours later noticed that it is still available, so I fixed the deadline (4/9) and system automatically cancelled and refunded the bets you did in the mean time.

Moral of the story: If you see it on the news it is too late to bet. Sooner or later your bets will be cancelled. You wouldn't want others to jump on your bets after the news, right?

PS: We are honored that you put us in your signature.

Also this is what happened when I placed a big bet on http://betsofbitco.in/item?id=209 as soon as I read the news that he stepped back:

1st bet:
Deadline still months away

2nd bet (after transferring more money):
Deadline changed to April 10th, 5 hours to go (probably my rather large 1st bet made them check the news? )

3rd bet (only 10 Bitcents that I wanted to use on a statement of my own):
Bet still open

Today:
All 3 bets cancelled and refunded (1st bet was refunded twice, I'm not sure if out of error or as a kind of reconsiliation), bet didn't close on April 10th but on April 9th according to the website now.

I don't know if anyone else was affested too, but something like a better overview who dit bet when how much might be useful for transparency.

Something like:
"Agree"
1BTC@1234weight@UNIXTIME12345678
2BTC@1200weight@UNIXTIME12365789
(in a nicer chart etc. of course)

If you don't want to do this, please at least have an API, then I could pull the data myself and display it on my own. I already have quite an extensive set of spreadsheets that calculates payout amounts if I place a bet this high on that statement and this data could be used to refine it.

Here is what is written on the help page:

Still is ambiguous to me.  Well, not only ambiguous but clear enough to me (before you clarified what the intention was) that I felt comfortable placing a bet on something that seemed to nearly be a "sure thing".

Had that initially been written so that there was no doubt the intention, it would have included the word "either".

Such as:
"This statement will be true if there is an official statement from the OPERA collaboration stating that either there was an error in the previous analysis or that the results are refuted by more data before the end of April.

This will be a good test here for seeing how this gets resolved.  Observing the process I wonder if improvements to policy for the service should be considered.

Specifically, the policy on when the service will weigh in on something like this.  Bets Of Bitcoin just provided clarification about how to interpret the details regarding an event after bets had already been taken but before the betting has ended.  In my opinion (as a layman) I don't think there should have been any comment other than to direct the user to existing published policy regarding what the judging process will be once the event occurs.


So an event can occur before the event date?  This should probably be referenced in the help:
 - http://betsofbitco.in/help