Very good start for the discussion - thanks bernardo!

Only one correction - the flaw is not apparent in Yes/No markets. In multiple choice markets the mentioned strategie was to buy very low percentages of unwanted options to reduce the buy-percentage of wanted options.
If you were right you could gain much more. This flaw was never used (IMO) to win by buy/sell operations only to buy at better terms.

e.g. on Friday I had 750K each on two options DAX and FTSE. With 3M risked I gained 3M each market. I wouldn´t put down the last 700K each at 40%/50% or 70% but at 20-25% yes. That was a high risc operation the dax traded shortly positive and about -0,2% in both markets would ahve killed my 3M.

Very good start for the discussion - thanks bernardo!

Only one correction - the flaw is not apparent in Yes/No markets. In multiple choice markets the mentioned strategie was to buy very low percentages of unwanted options to reduce the buy-percentage of wanted options.
If you were right you could gain much more. This flaw was never used (IMO) to win by buy/sell operations only to buy at better terms.

e.g. on Friday I had 750K each on two options DAX and FTSE. With 3M risked I gained 3M each market. I wouldn´t put down the last 700K each at 40%/50% or 70% but at 20-25% yes. That was a high risc operation the dax traded shortly positive and about -0,2% in both markets would ahve killed my 3M.

Just because you didn't do it doesn't mean it wouldn't work with Yes/No markets :) It's the same principle and you could get a sure 95% option at a much lower percentage.

I can't really remember how it has been on Hubdub, but I tink a few hours of cash-in ban could have done the job to never let this come up as a big problem.

HD had no cash-in bans, as far as I remember.
The initial odds manipulation was effective on HD, but may have been punished (fines) as "gaming".
However, some HD'ers wrote that the effective odds immediately after heavy bets were not as good as the odds one could get 1 hour after the large bet. If you dare the delay, of course...

@bernardo - you are right it works but like in your example you mostly have no 5% option but a 30-50% one. This reduces the gains of that strategy much. And markets with less than 15% starting odds for yes and no may be realistic but don´t make for a successful Yes/No-design.

To emphasize my previous point: the multimillionaires got richer here because of buying cheaper than we should have - we still had to chose the right options. We didn´t cruise to the lead by buying/selling all the time and living from transactions gains with no real risc - we bettet huge amounts at good odds and were monstly right. Last week our betting provided many oportunities on the indices markets for everyone. The generally raised activity can be seen in both weeeks gain in the leaderboard.

Well, you can turn it around as often as you want to, but in the end you made a lot of money by just buying and cashing in (and by doing this generating huge gains). The fact that you had to have a big share on the winning option doesn't change this.
But the blame is on the system here.

@charlesf: Since the actual problem is the very low percentage of the winning bet, I guess a cash-in ban wouldn't do much about it. And I wouldn't want to see a general ban on multiple bets placed at once.
An investment limit on the other hand could effectively reduce this kind of gains to a minimum and has the nice side effect to be an incentive to create new markets.

@charlesf - Selling only in the last 3 or 4 hours in the daily indices could give everyone chances to bet how they like, but make no mistake 100K or 250K an option could still change the markets as long as the model stays sensitive.

Bernardo,
I have been concerned about this problem for a while and was going to post my thoughts in the forum but I am glad that you beat me to it. First, I don't type very well and second, I am sure that you explained it better than I would have.

Some other posts have also suggested limiting the amount of $ that a player can wager on a market. I don't think that this is neccessary. If you can afford to bet big to win big you can also lose big. This is how the game is meant to work. If someone makes a large wager on one or more options I can make a small wager on 1 of the other options with the potential for a large gain or wager on a different market. This is fair to everyone. Also, at this time we don't have enough markets for the wealthier players to invest in if there are limits.

The solution I came up with is a waiting period just like we have for people who create a market. They can't make a wager on their own markets for at least 2 hours. In this case any wagers made on any market can't be cashed out for at least 2 or 4 hours. The only place where I have noticed this occuring is in the daily financial markets. These attract enough players both large and small so that after having 2 or 4 hours for everyone take advantage of the generosity of the dummy bets they won't be worth very much when the time comes and they can be cashed in.

The financial markets are only open for 8 to 10 hours so for these and other short term markets this rule could be suspended for the last 4 hours that the market is open. By then every one who was interest in the market would have had time to make their wagers and there would be enough $ in the market to make this betting technique useless. For longer term markets maybe the 2 hour time limit could be increased to 6 or 12 hours and only be in effect for the first 24 or 48 hours that the market is open.

Bernardo,
I have been concerned about this problem for a while and was going to post my thoughts in the forum but I am glad that you beat me to it. First, I don't type very well and second, I am sure that you explained it better than I would have.

Some other posts have also suggested limiting the amount of $ that a player can wager on a market. I don't think that this is neccessary. If you can afford to bet big to win big you can also lose big. This is how the game is meant to work. If someone makes a large wager on one or more options I can make a small wager on 1 of the other options with the potential for a large gain or wager on a different market. This is fair to everyone. Also, at this time we don't have enough markets for the wealthier players to invest in if there are limits.

The solution I came up with is a waiting period just like we have for people who create a market. They can't make a wager on their own markets for at least 2 hours. In this case any wagers made on any market can't be cashed out for at least 2 or 4 hours. The only place where I have noticed this occuring is in the daily financial markets. These attract enough players both large and small so that after having 2 or 4 hours for everyone take advantage of the generosity of the dummy bets they won't be worth very much when the time comes and they can be cashed in.

The financial markets are only open for 8 to 10 hours so for these and other short term markets this rule could be suspended for the last 4 hours that the market is open. By then every one who was interest in the market would have had time to make their wagers and there would be enough $ in the market to make this betting technique useless. For longer term markets maybe the 2 hour time limit could be increased to 6 or 12 hours and only be in effect for the first 24 or 48 hours that the market is open.

Bernardo,
I have been concerned about this problem for a while and was going to post my thoughts in the forum but I am glad that you beat me to it. First, I don't type very well and second, I am sure that you explained it better than I would have.

Some other posts have also suggested limiting the amount of $ that a player can wager on a market. I don't think that this is neccessary. If you can afford to bet big to win big you can also lose big. This is how the game is meant to work. If someone makes a large wager on one or more options I can make a small wager on 1 of the other options with the potential for a large gain or wager on a different market. This is fair to everyone. Also, at this time we don't have enough markets for the wealthier players to invest in if there are limits.

The solution I came up with is a waiting period just like we have for people who create a market. They can't make a wager on their own markets for at least 2 hours. In this case any wagers made on any market can't be cashed out for at least 2 or 4 hours. The only place where I have noticed this occuring is in the daily financial markets. These attract enough players both large and small so that after having 2 or 4 hours for everyone take advantage of the generosity of the dummy bets they won't be worth very much when the time comes and they can be cashed in.

The financial markets are only open for 8 to 10 hours so for these and other short term markets this rule could be suspended for the last 4 hours that the market is open. By then every one who was interest in the market would have had time to make their wagers and there would be enough $ in the market to make this betting technique useless. For longer term markets maybe the 2 hour time limit could be increased to 6 or 12 hours and only be in effect for the first 24 or 48 hours that the market is open.

@bernardo: "but in the end you made a lot of money by just buying and cashing in (and by doing this generating huge gains)."
Actually, no.
The initial-odds-deformation betting does not necessarily involve cash-in at all.
The profit is taken at maturity, i.e. market completion.
The player cashes in only the deforming bets to save on expenses, and trusts his/her (or oddschecker.com's) judgement to have made the right choice.
The point of market deformation is to ensure a low point of entry (at a moment of low sensitivity of the market, so you can invest a lot at this low point) for the bid that is expected to win.
The bets that prepare the market for the "investment" may be cashed in (average recovery is about 25% for me), but also may be left in the market.

Bernardo,
I have been concerned about this problem for a while and was going to post my thoughts in the forum but I am glad that you beat me to it. First, I don't type very well and second, I am sure that you explained it better than I would have.

Some other posts have also suggested limiting the amount of $ that a player can wager on a market. I don't think that this is neccessary. If you can afford to bet big to win big you can also lose big. This is how the game is meant to work. If someone makes a large wager on one or more options I can make a small wager on 1 of the other options with the potential for a large gain or wager on a different market. This is fair to everyone. Also, at this time we don't have enough markets for the wealthier players to invest in if there are limits.

The solution I came up with is a waiting period just like we have for people who create a market. They can't make a wager on their own markets for at least 2 hours. In this case any wagers made on any market can't be cashed out for at least 2 or 4 hours. The only place where I have noticed this occuring is in the daily financial markets. These attract enough players both large and small so that after having 2 or 4 hours for everyone take advantage of the generosity of the dummy bets they won't be worth very much when the time comes and they can be cashed in.

The financial markets are only open for 8 to 10 hours so for these and other short term markets this rule could be suspended for the last 4 hours that the market is open. By then every one who was interest in the market would have had time to make their wagers and there would be enough $ in the market to make this betting technique useless. For longer term markets maybe the 2 hour time limit could be increased to 6 or 12 hours and only be in effect for the first 24 or 48 hours that the market is open.

Bernado, et al. , very good start of discussion, etc... I think we need to understand more fully "where" the money is coming from... There is a function which evaluates the probabilities, that relates to the cost function of the
various options. Hanson: C = b * ln(exp (q1/b)+exp (q2/b) ) where exp means e to the power of xxx where b is a constant related to the "depth" or amount the market can lose maximally..). As you were saying a shallow market is one where b is low, and a deep one where b is large...

If the losses are coming from Knew the News really giving $ to people who are able to utilize this aspect, then it is good it is discovered... (and it is appropriate that all this WAS DONE, and that it is now being addressed, in due course).. The losses, however, may be in part, be from others who have made poor bets and/or not totally
from the question, or that there were poor choices of starting percentages on the question... If the former (people with poor bets), then we do not want to change anything, in this regard.., if the latter, poor starting percentages, then again, this just requires us to be more diligent in forming questions, and nothing needs to be changed. So, we need to understand it a little bit better, and how best to set the "b value" in the above percentages, without necessarily any complicated rule change..

So, let's take our time about it and understand the various factors first...

Hanson has clearly shown the importance of the initial odds to assure a limit on market maker's loss.
A player, however, can bet to augment the initial or current odds (at the cost of these auxiliary bets), then place a bet at much better odds, and even partially recover the auxiliary bets.
I understand the question is not about the market maker's loss limits, but whether this, and possibly other, betting tactics should be limited/taxed/puniushed.

About where the money is coming from - if the market maker's loss is limited, then all the money come from the betters (plus the limited loss of the maker).

Bernardo,
I have been concerned about this problem for a while and was going to post my thoughts in the forum but I am glad that you beat me to it. First, I don't type very well and second, I am sure that you explained it better than I would have.

Some other posts have also suggested limiting the amount of $ that a player can wager on a market. I don't think that this is neccessary. If you can afford to bet big to win big you can also lose big. This is how the game is meant to work. If someone makes a large wager on one or more options I can make a small wager on 1 of the other options with the potential for a large gain or wager on a different market. This is fair to everyone. Also, at this time we don't have enough markets for the wealthier players to invest in if there are limits.

The solution I came up with is a waiting period just like we have for people who create a market. They can't make a wager on their own markets for at least 2 hours. In this case any wagers made on any market can't be cashed out for at least 2 or 4 hours. The only place where I have noticed this occuring is in the daily financial markets. These attract enough players both large and small so that after having 2 or 4 hours for everyone take advantage of the generosity of the dummy bets they won't be worth very much when the time comes and they can be cashed in.

The financial markets are only open for 8 to 10 hours so for these and other short term markets this rule could be suspended for the last 4 hours that the market is open. By then every one who was interest in the market would have had time to make their wagers and there would be enough $ in the market to make this betting technique useless. For longer term markets maybe the 2 hour time limit could be increased to 6 or 12 hours and only be in effect for the first 24 or 48 hours that the market is open.

I do not believe there is a problem,as long as the initial probabilities
are correct. It might be advantageous to lower the starting Hanson B value that
KtN uses, so as not to lose too much in the Market.

I also have advocated elsewhere, about not having too much $ bet , as this stifles people
who do not have so much $, from investing, and the big fish gobbles up all the goodies..IMHO.
I am just more egalitarian, which has served science , knowledge, and mankind well . It is
the opposite in politics, where money, influence, etc. make a sham of many of our democracies,
creating a lot of unrest in the world, IMHO, which becomes more apparent, when people cannot
get their basic needs met, IMHO.
ken

Bernardo,
I have been concerned about this problem for a while and was going to post my thoughts in the forum but I am glad that you beat me to it. First, I don't type very well and second, I am sure that you explained it better than I would have.

Some other posts have also suggested limiting the amount of $ that a player can wager on a market. I don't think that this is neccessary. If you can afford to bet big to win big you can also lose big. This is how the game is meant to work. If someone makes a large wager on one or more options I can make a small wager on 1 of the other options with the potential for a large gain or wager on a different market. This is fair to everyone. Also, at this time we don't have enough markets for the wealthier players to invest in if there are limits.

The solution I came up with is a waiting period just like we have for people who create a market. They can't make a wager on their own markets for at least 2 hours. In this case any wagers made on any market can't be cashed out for at least 2 or 4 hours. The only place where I have noticed this occuring is in the daily financial markets. These attract enough players both large and small so that after having 2 or 4 hours for everyone take advantage of the generosity of the dummy bets they won't be worth very much when the time comes and they can be cashed in.

The financial markets are only open for 8 to 10 hours so for these and other short term markets this rule could be suspended for the last 4 hours that the market is open. By then every one who was interest in the market would have had time to make their wagers and there would be enough $ in the market to make this betting technique useless. For longer term markets maybe the 2 hour time limit could be increased to 6 or 12 hours and only be in effect for the first 24 or 48 hours that the market is open.

@All
I only posted the above comment once and I am not sure how it got here a six more times. I don't mean to offend anyone. It appears to have happened to other people too.

I left the tab open, came back a few times and refreshed the page to see if anyone posted a reply.

Is it a bug in the system or am I doing something wrong?

@charlesf
when you refreshed your page, you resubmitted your post reply. normally, the browser askes you if you are sure to do so. anyway, I've changed the system to make a redirect when a reply is posted. so, from now on, if you refresh the page after posting a reply, you will refresh the redirect, and not any longer the submit of your reply. hope the issue is solved this way.

@ kruijs

Thank you.

@ hlf13

"To emphasize my previous point: the multimillionaires got richer here because of buying cheaper than we should have - we still had to chose the right options. We didn´t cruise to the lead by buying/selling all the time and living from transactions gains with no real risc - we bettet huge amounts at good odds and were monstly right. Last week our betting provided many oportunities on the indices markets for everyone. The generally raised activity can be seen in both weeeks gain in the leaderboard."

This excerpt from one of your above posts is a perfect illustration of why this betting system is so unfair to everyone else.

You buy your positions at lower odds than you should have.

Here is a simple example to show why this is unfair to everyone who plays the game fair.

I make a 100$ wager at 20%. If I am correct I will receive 500$.

You manipulate the market so you get to make your 100$ wager at 10%. If you are correct you receive 1000$.

If you receive twice as much for every correct prediction as everyone else in the game you would only have to be right half as often as everyone else to earn the same return. Sounds unfair to me.

It would be like the guy sitting next to you at work, doing the exact same job as you, but he gets paid twice as much as you because he found a way to rig the payroll system. How would you like that? It just doesn't seem fair to me.

As to your other statement about 100k and 250k wagers having a major effect on any market. Fine. All markets are first come first served and anyone can wager as much as they are comfortable with as long as they play by the same rules as everyone else.

How about 2 simple possible solutions that will solve the problem and avoid any time limits or betting limits that no one wants and system changes that the Admins probably don't want.

I will post them separately so people can vote thumbs up or down.

Option #1

Rule odds manipulation legal but create a separate leaderboard for players who choose to use this betting system because it would be unfair for everyone else who plays the game straight to share the leaderboard with players who get twice as much return for their correct predictions.

Option #2

Call odds manipulation what it really is, Gaming the system, and ban it.

This way the wealthy players and the not so wealthy players will all be playing by the same rules.

@charlesf:
Both options require that all editors actively sniff out who plays how - it's not easy, and the more players, the worse it gets.
Here is an example:
In a daily (8 hours) market with current total volume 5K (and already very far from initial odds state) a player invests 500K at Option_A. The market is further distorted and is already "deep", that is all further bets make small changes.
1. Then the same player bets 200K at Option_B, at low odds. At this stage, is it manipulation?
2. Half an hour later, the same player bets an other 200K at Option_C, and caches-in what is remaining form the 500K bet at Option_A. At this stage, is it manipulation? BTW when asked, the player explains the erroneous bet as typing error, or current events at the market, or a friend's recommendation - you name it, you shall hear it (i.e., the editor will).
3a. The market closes with Option_A winning (and a complete loss for the player). Was it gaming?
3b. The market closes with Option_C winning (and good profits). Was it gaming, or market state tracking bets?
3c. The market closes with Option_B winning (and huge profits). Was it gaming?

What I mean is, a market-maker function is good because there is (almost) no human intervention when calculating profits/loss, and if it guarantees that the market maker makes limited profit, or a limited loss, it should be fair - all the money comes from players.
Last, this is a game. I wonder what gaming the system really means ;)

The problem is with the model that allows this strategy. This was reported 3 weeks ago. And as you see in this thread there are many solutions suggested. After a new stable set of rules will be found there will be a networth reduction at the multimillionaires.

In between we enjoy the daily fight between us.

I have a problem with new markets moving 20% at 40%/50% at betting only 1200. IMO the mrket modell allows too little amounts to be bet at reasonable odds.

@charlesf: Against banning - the rules must be changed in a way that solve the models flaw and there are many possibilities.

1) No selling on indices
2) make the model linear instead of progressive
3) give the starting odds a base value (e.g. parts of 100K)
4) higher fines by time or for indeces
5) only allow selling indeces in the last two hours before suspend.
6) limit the playing capital
Zvassil already adapted the possibilities on DJIA; Nasdaq & VIX as wella s the weeklys to cover the high volatility and give more options.

re comments by HFL and CHARLESF;

I do not agree that there is a problem of any significant magnitude... I have investigated the situation with a spreadsheet to simulate KtN, and also have made some strategic bets of the type on markets: 8492 and 7868, to try and "game the system", in the ways suggested by Charlesf. I find that the so called "gains" are predominantly offset by losses, in those outside bets which then must be sold. (I later sold all my bets in these markets to start fresh, for the purposes of illustrating that it didnt do me any good).. Whatever gains I made in the markets (which appear to exist are predominantly "on paper", but are not real). Let me explain this "paper value":

Let's say for example, there is a question; who whill be the next President of the United States, and one
option is Sarah Palin, and it is down at 0.1% level, so i keep betting on this options.. Assuming I have enough
money to drive it to a HIGH percent of winning, then "ON PAPER" it will show that my NET WORTH is going up,
because i have driven it (the value of the item into the non-linear range, where it thinks I am going to
win).So now it estimates my value based upon the chances it thinks are appropriate (say i get her to be winning by 95%; all my stupid bets are NOW VALUED AS VALUABLE BY THE SYSTEM!!! HOORAY!!!.
NOW , if I am right, and I DO WIN (she becomes pres.) then i do make the money that was predicted by the system as it has shown ( I haven't "gamed" the system; i just knew something nobody else knew). If she loses, then the "system" was wrong to think my bet was good (on PALIN), but all it has is the average chance of winning
of that particular bet (it has no way of knowing what the real outcome will be)..So, then I lose $,
and that is fine.. , because i was DUMB, beting so much on her.. the Truth will out..

In my tests on betting strategies, i found that all the losses you reported charlesf, (which you mentioned in your
%, which sounds like you can get double % winnings, ARE FALSE! they are offset (predominantly ) by the losses
you endure by having to sell the other options... and by the 10% loss in selling... There are some small gains, but those are consistent with the amount the computer (KtN) puts in to pay for the market... The majority a person makes is due to the losses incurred by others who be on the wrong option... The only way to test this, is on a market which is NOT being bet on, or via a spreadsheet which mimics the rules...

Nevertheless, as far as changing some aspects of the system, YES, there are aspects that could be improved.. I believe going on a witch-hunt for those who have bet in ways to alter the odds, would not be a good use of time. People have done these techniques of betting in ways to invest money as you outline, but IMHO, they have made money by BEING PREDOMINANTLY RIGHT, and although this method allows one to invest more money (driving the % down first), that is PREDOMINANTLY DUE TO THE FACT THAT one can bet smaller % (fore example in the $100 bets rather than $1000 bets, since the betting equation has this funny (non-linear) behavior, but it is all withing the equations of Hanson's on betting probabilities and their relation to cost functions, etc.

So, happy to consider a democratic solution as you suggest Charlesf, although "truth" is not a democratic process, nevertheless, there are aspects on betting that could stand being improved.. If one were to consider the search for banned activities, one needs a good committee.
anyway , sorry for long email... i wanted to address the many issues..

Had a follow up idea...

Question: Would people agree to test how severe the "problem" is:
by 1) having a "commission" to investigate how much $ can be made? - By a person changing odds, on
a question designed to illustrate how the odds are manipulated. It can investigate the cases where other people are "betting", and/or "not betting." Naturally, some money IS made by people who DO bet correctly, but IMHO,
their wins, should predominantly be a result of the initial odds (and how much is made when they bet on a question which wins (if the odds were low , say 10%, in theory, they could win 10x as much money as they put in, but the actual amount depends on the thinness of the market: the b value, which affects how much $ KtN puts into the market..)

One idea, that might be worthwhile before undergoing drastic changes. Naturally, if there are clear easy things to change, that would be good for the site, these should be made first.

1. At the moment this kind of betting described above is legal/clever
2. The gains are high but the risks are high too (90% invested last 3 weekdays) without getting the winning result the strategy is dobbling the loses.
3. The early bird catches the worm
4. The strategy used the last days lead to cheaper options for everyone because the unwanted options were kept to buy later again and were sold very late. The millions in the market this week (revised since monday) gave opportunities to everyone. IMO a 4M market is much better than a 100K market.
5. Everyone agrees that the chances with this strategy are too high and a fix needs to be found (purpose of the discussion)

Any kind of gaming or baning police is shure to suck the fun out of the site and will start a whole lot of hostilities.

My grudges with the current model:
1. very low percentages schould be cut off - buying 300K -1,4M for $100 is rediculous = no option should be below 0,5% in markets. Initial design requires at least 1% each option and there is always the possibilty that the FED sells gold or Warren Buffet buys Italy ;-) In most crises there is a day/s when there are huge gains because people come back and the volatility is generally high.
2. gains are too high with the clever strategy - change model &/ fine or block selling total/partially
3. model allows too little bets in some markets with Yes/No you can only bet 20K if everyone shares one opinion or without the "clever" strategy
4. I hate it when early bets make the percentage rise like a rocket (from 40% to 60% = $1200) or from 5%to 90% with $5000. = because of time constrains and the many bets to make my 100 are dying out and my 2000/5000 are not so nice for me and the others who have to buy after me.

I agree with MANY , but not all of the things HFL says.
First, let me point out some numerics that tend to support the "system" is working properly (it doesn't mean
it cannot be changed to be improved for our purposes, however, Kruijs has incorporated the software brilliantly,
and is to be GREATLY COMMENDED FOR THAT..
The calculation is as follows: Consider a rich user,A, who has obtained, say 20,000,000$. Let us say, the market "B value" , which means the typical setting used by the site for all the question is $10,000. (although it could vary on different questions), this means that on any question, if a user plays it correctly, this is the most he/she can make. Let us say, the user A, has played 10,000 questions, then he/she could MAKE/EARN by excellent play an upper value of 100,000,000. Our top user , HFL, has about 20,000,000. So, that is 1/5 of the 100,000,000, certainly a value within legal bounds. (Of course, all the money comes from (originally KtN), but the top dog, cannot get it by the daily allowance (the $25 per day, etc.). Only those big sums can be obtained by betting. If the top dog, outplays the little dogs, then their daily money given to them, and the money they cull by investing wisely on questions(limited to the b value per question, just gets gathered into the smart dog's pockets.. ). 20%, as mentioned above by that calculation is like a 200 batting average in American baseball, not great, but for this kind of market, super, since the average is >but near 0% (people gain predominantly from others, and some from the market).. IF the market were truly broken, our top winner would win more than the 20%, i just calculated, IMHO..

Now, with regard to the items HFL discusses: 1) I do believe, that any betting, within the system, even if it misleads other players (as some watch what our top players bid, and then just follow like sheep), which is why i think NEWBIES, should stick to WHAT THEY KNOW, as opposed to mimicing top players.
2) I agree, that low % should be cut off (but not too low, since they can be REAL RARE probabilities that only a shrewd person will see). (and if cut off, they all need to be gathered together, or included IMO , no offense to anyone, in a catchall - "other, not by suspention date, etc. option", since otherwise,if the rare event occurs then nobody wins, what happens? or what will KtN do?
3) I area i disagree with, and it may be pedantic, deals with the item 5 "Everyone agrees...", but perhaps it is just the wording. I have no problem with changes, but I do not agree the system is broken..
4) with regard to the sums rising quickly with smallish bets (that is because they are smallish to you, but to a newbie, they take all one's money..., so therein IS a problem (the vast differences which accumulate)..

BY the way, I made a bunch of money recently, NOT by using the above strategy, but from betting against what others had bet by using it.. Then i lost about 100,000+ today, because i couldnt get on tthe system for >1 hour, and i do not know why (international cables down, or slow, or my computer internets system??? i do not know), but this is as HFL SAYS, we risk a LOT when we bet aggressively... and this money probably is in HFL's pockets now.. (do you have lots of big greenbacks? I can send you the serial numbers..)..

Well, just try and have fun on this, that is the purpose.. it's all play money..

@kenneth1: Here is an example (#8495 DAX today, now at suspended state) -
Market vol before bets: 0
Bets: 1242000
Cache-in result: -2378055
Tax: -362005.51
Total market vol at suspend time: 5342477
I will post net result later

Note that today is not the best day for large bets - all indices are very volatile

@zvassil
Neither option requires the editors to sniff out anything. You can't hide what you do and you don't have to because at this time there doesn't appear to be any rules against gaming. Whats going on is obvious to anyone that clicks on the transactions window right after the big wagers are made and imediately sold.

Because of this a simple change to the rules to ban this betting technique would end it. I believe 100% that you and the others would obey the rule and stop. Therefore no need to make major changes to the system just to deal with this issue. Am I correct in my assumption?

For your example:
1. No. There is nothing wrong with betting on multiple options in the same market. Any number of options and any size wagers would be fine.
2. Without the explanations, probably not. With one of your explanations, if true, No.
3a, 3b, 3c. The final outcome and if you win or lose has nothing to do with if you tried to game the question.
If you lose it is because of either poor predicting or just bad luck.

The only thing that determines if you are trying to game the market is your intention when you make the first wager. If the wager is made with even a small expectation that it could win and the intention to leave it until settlement as long as you still believe it may win it would not be gaming even if you later cash it in because of a market change or any other valid reason like you gave above.

What is trying to game the market is when you make the first wager with the sole intention of driving down the odds for the wagers that you actually think will win. You make your intentions quite clear when you cash in your first wager within seconds of placing the wagers that you think will win. You may be able to convince the editors that you made a mistake once or twice but no one would get away with it beyond that if a rule was added to ban the gaming of markets.

@kenneth1: 4) but my problem gets everyones else because now I am forced to bet more to acchieve my aimes --> I could ruin new markets for everyon else by betting up to 75 or 90% easily. Then your only option would be to follow (no high rewards) or bet against ( and in the long run my track record is pretty good if I am willing to go to 90%)

@ken

Your analysis of your Sarah Palin example doesn't quite cover the point I was trying to make. I will expand a little on your example to explain my point and keep the numbers small because I am unable to calculate the expected return of larger wagers. Lets assume that we are both equal and very good in our predicting abilities and that I get to the market first. There are 5 options all at 20% and I wager 100$. My return if I win will be 500$. You get to the market as soon as I get done and before anyone else. As in your example you make a sham wager on candidate X suficient enough to lower the odds on Sarah Palin to .1% and make a 100$ wager on her just like I did. Next you cash in your original sham bet on candidate X and receive the whole amount minus the 10% tax. (your 100$ wager would have a negligible effect on it.) If you win your return would be 100,000$.

If we both lose, I lose 100$ and you lose 100$ plus 10% of your sham wager.
If we both win, I receive 500$ and you receive 100,000$.

Seems like a significant difference to me.

If in your example your sham wager only lowered the odds to a more reasonable and likely 5% your return would still be 2000$ to my 500$. Still a significant difference.

@hlf13

I agree with you that allowing people to game the system by odds manipulation is legal for now but I can't understand how you would call allowing a few wealthy players to have an unfair advantage over 90% of the players who can't afford to use the system.

A little more than a week ago after careful thought and a bit of research I found a market where at 11:55 on July 31st I could have made a wager of under 500$ that would have promoted 1 player from Gray to Pink. This would have allowed me to alter the results of 2 of your markets at the same time resulting in a huge payout for me at real good odds.

I didn't do it because I thought that it would be unfair to the other players, just like allowing a few players to manipulate the odds to get bigger payoffs is unfair to the other players.

If I had done it, would you have considered it legal? I didn't find anything in the rules that said that I couldn't do it.

Also, would you have called me very clever because I was able to game 2 markets at the same time?

If you can't answer Yes to both of my questions you should recommend that your betting system be banned along with my idea to promote inactive players.

@charlesf: "The only thing that determines if you are trying to game the market is your intention when you make the first wager." - well, this is what I call sniffing out (a witch, or a contrarian investor ;) )
Intentions are quite hard to guess, at least for me.
Having rules that rely mostly on referee judgement, however esteemed the judges be, can compromise any game.
Thank you for the time to illustrate your opinion, however I do not agree the game can benefit from such rules. I think most restrictions should be explicit and constructive in nature - i.e. their execution should not rely on opinions whereever possible.
You mention that "You make your intentions quite clear when you cash in your first wager within seconds of placing the wagers that you think will win." - what would you say to a system feature that will block cashing-in on each bet that is at least say, 25% of current market volume, for some time (say 1 or 2 hours)?
I can agree to such a restriction, just because it can be enforced automatically.

@kenneth1: Here is an example (#8495 DAX today, now at suspended state) -
Market vol before bets: 0
Bets: 1242000
Cache-in result: -2378055
Tax: -362005.51
Total market vol at suspend time: 5342477
Sold price: 5187671.12
Net gains: 1205610.61

@charlesf: thank you for your fairness - this kind of problems comes from open bets - I am ok with influencing the colour markets by own accievement (or a timely loss) - what I didn´t like is using inactive players for that. After frustrating bets (and a KtN question too) in June I changed the design and watch every change. Last time I blocked a try with bets against. You need big moves to influence the market and I have more money and the void option. I intend to protect the mayority as long as there are no news justifying the changed positions.

IMO the clever strategy is not robbing other players IMO it even offers great opportunities - we take our money from KtN/kruijs not others.

3 weeks ago I watched zvassil on my own question and finally was able to understand his strategy. I posted it to the admins the same day and the consensus was clever but legal. I had 1,3m at that point and was unable to follow zvassils progress, maybe that was part of the reason it was not taken seriosly then. To test the strategy and to keep up I started myself on Monday 18th. All the time I reported my experiences about the flaw and when I made 7M last Friday it was obvious that there is some reason to reform. One easy fix (markets can no longer be brocken by a million was already made). Since then I had lots of fun and encountered all kind of intense forms of strategy. E.g. it is still my guess that he wanted a decission last Friday by betting 2m in DAX & FTSE each. I had to follow or let him pass. That day I tried to block his investments a bit (no easy way to sell). In the end he deinvested and I made 3m each market. But that realy was fun. Yesterday he was more succesful. Competing with your peers is the most fun.

By the way zvassil & I have agreed to give up the vast majority of our networth as soon as the problem is solved. Then we will fight with kenneth1 & cheesenips :-)

My guess is that I would have about 5m now without the strategy.

We are still in beta mode!

let's cool down a bit and lets have a look what our problem really is.

the problem appears to be that some players are able to make large gains in short time.
how do they do so? they play stock index markets.
why do stock market questions allow people to get rich quickly? the risk to lose money is very small, you simply follow the development of a stock index market and buy/sell predictions accordingly.
why is it so easy? because not many people play these kind of questions (or are active at all on the site), it is no problem to sell a prediction at a good value.
what also makes it easy? there are at least four markets which allow this kind of trading, daily.
do other markets also offer this kind of opportunities? no, not really.
why are these markets allowed? because they are fun.

IMO, we should return to what this site is all about: News prediction. With "news prediction", people really had to buy predictions, and were no longer hunting to follow a market index development.

so, how can we prevent any kind of "strategy" right from the start?
- limit or reduce the number of index markets to reduce the opportunities
- make it impossible or non-lucrative to heavily buy/sell
- make markets more diverse to make predictions harder
- make markets close sooner, or completely disallow the markets to run in-play, to enforce an actual prediction, rather than buying stakes based on market index movements.

@kruijs:
Currently, like in HD, the most popular leaderboard is Networth.
The restrictions above actually intend to change that.
Instead of trying to enforce restrictions, I think you could try to change the leaderboard.
For example, the networth one could be dropped entirely, and replaced by Monthly gains, or some other periodically re-set function.
I think that enforcing restrictions will always be much more difficult and unpopular than changing the target function

@zvassil
"The restrictions above actually intend to change that." no, that's not my intention.
in fact, someone good in "predicting news" should have the opportunity to be on top of just that list.
do you thinks that you and hfl13 have proven to be good in "predicting news" during the last week?
how many different topics did you correctly forecast? is it in relation to what your gains were?

This is exactly what I had in mind - a networth target function is not very compatible with your actual criterium, "news predicting".
I think "predictions" would be better measured as "number of successful predictions", not "value won/lost" or "return on investment".
This, of course, is an opinion and some suggestions of a player, who is interested in the game.

well, if you click on "Leaderboard", you will notice that it's not the Net Worth leaderboard which is shown most prominently. this is just because Knew The News want to focus much more on interaction between the players.
but still, a networth target still is a valid number to represent success in predicting news. otherwise, we could reduce the Knew The News trading aspect to a mere vote for or against (without the currency aspect at all).

but anyway, it might be a good idea to add those leaderboards.

1) Thanks zvassil, for your comment on numbers about
the effect you are talking about..... I shall look into it..

2) Re the changes to the gray market, that caused trouble to people
who lost money on that market...... Let me acknowledge
responsibility for doing this once with an inactive player,
and accept responsibility for that. Am sorry that this caused so much trouble to players.
If it is deemed I should be docked for that, then fine.. I did not see anything
in the rules against it, and just did it to see if it could be done...
Regret that decision now..

@ zvassil

I still think a ban would work because I believe that everyone in the game with enough wealth to use the system is honorable enough to obey any writen rule. Because of this I don't think it would be difficult to enforce.

Now on the other side I know that the Team has a limited amount of time each day to donate to the game and I would prefer that they be able to use that time to create new markets, settle markets, and try to find ways to get more players instead of having to play policeman or refferee our differences of opinion.

As to your solution of a short short time limit before being able to cash out large wagers. I like it. Already gave it my thumbs up. It is simple and will only affect a very small number of players.

You will still be able to use your system but it will be a little bit less profitable and it adresses most of my concerns. You still get the best odds and first choice of which options to wager on. Any other player that plays the market for the next few hours can still get real good odds on the options that you didn't like. You are very good but not perfect so sometimes a little guy will get a big return on one of the other options. This will help him to grow and stay in the game. Also with large wagers on 1 more option players who follow you will still get reasonable odds on the options that are more likely to win. Yes, the ones that you wagered on.

Your solution is very good but still I offer a few minor additions. Let me know what you think.

The 2 hour limit should be fine for short term markets but make it longer for longer term markets. Maybe 12 hours for 2-3 day markets and 24 hours for markets lasting 5 days or more. This is so more players, especially those who may only visit the site once a day will have the opportunity to find the market and play it under the more favorable conditions.

I know that each market has a certain amount of $ built into it before any wagers are placed. Does anyone know how much and can this amount be used as the current market value until the amount wagered reaches a certain point. I ask because the first bettor is looking at a market with a current market value of 0 and would trigger the time limit every time. This could also apply to the first half dozen bettors if they all make small wagers. I am sure that the intent is not to impose the limit on the small bettors. If the built in starting amount can't be used how about exempting any bets under 250$(subject to discussion).

One final suggestion to protect the big bettors. Remove the time limits at the end of all markets for the exact same # of hours that the time limit is for that market. From the above suggestion the time limit to cash out large wagers would be waived for the last 2 hours for daily markets, the last 12 hours for 2-3 day markets and the last 24 hours for 5 day and longer markets.

Lets say you made a 500,000$ wager on Severe Loss on the DAX market an hour and a half before the market was to close with with the intent that it would Win and it was big enough to trigger the time limit. If the DAX market were to rise up to Slim Gain over the next hour and you were sure that your 500,000$ wager was going to lose you would be prevented from cashing it out by the time limit. This would be unfair to you. This final sugestion would allow you to liqidate any position for any reason before the market closed.

Again great idea!

re zvassil's and kruijs' suggestions.....
I like the spirit of them... There is ONE aspect which has not been
discussed.... , which i think needs to be considered too. I am not sure
the proposed solutions does that, but it could be i do not understand
exactly what will be done mathematically..

The "problem", let us call it the "the wealth problem", which
i think is largely ignored, except by those who suffer from "too much wealth".
It is this... With the wealth comes a burden... The wealth can be
like a millstone around one's neck.... One is limited in the amount, percentage-wise,
one can earn in the same period of time, compared with one's poorer compatriots.
This is because, say on a particular question, it "saturates" at say 100 K, and perhaps at
that price, one can obtain, let us say another 100 K. So, one gains 100% on that option.
So that if one wants to go not higher than 90% probability of winning one option,
that is all one can MAKE, say 100 K on that question. IF one risks going higher than 90%,
one would gain but little beyond the 100 K, and risk a LOT more..
I do not know what the exact value is, at 90%. It likely depends on the starting %, and
how the other bets go, etc... But let us just use these #s. Now, If one's wealth is, say 200 K,
then one can make 1/3 , 33.33% in that time interval..IF however, one's wealth is 2 M, rather than
200 K, now one can only make 3.33..%, 1/10th as much percentage-wise...

So, the reset, you are suggesting (would need to reset the amounts people have).
So, one's wealth would be wiped out... This is why people with lots of $, also "leave"
the game a bit, when they get too high, there are no good questions where they can
gain a good betting environment... So, i am in favor of the suggestions offered about
some kind of reset, if it addresses this problem , "the wealth problem.".

So, if the wealth is not reset, it is not possible for wealthy people to play with the same
zeal, as ones who are not so burdened...

@ hlf13

I agree with most of your above statement. Effecting markets by ones own abilities is expected but using inactive accounts to do so is unfair to the game.

I disagree with the part where you are not robbing other players. You are right that you aren't actually taking their $ but you are depriving the other players of the opportunity to get a fair return on their winning wagers in any market where you use your clever system.

hlf13, your quote: "the multimillionaires got richer here because of buying cheaper than we should have"

If you are buying cheaper than you should have, that means that everyone else is forced to pay more than they should have to get the same $ return in any market where you use your stratagy. Most new players have more losing predictions than winning ones. Lets say that New Player A makes 5 100$ wagers in 5 fair markets on positions at 20%. If he loses 4 he is down 400$. He wins 1 and receives 500$. He breaks even for the day, gets his 25$ for showing up, and comes back the next day to do it again. If he makes the same 5 wagers at markets where you used your system, he loses 4 and is down 400$. Next he wins 1. The fair odds should be 20% but since you manipulated this market he has to buy his position at 40%. He only receives 250$ for his win. He gets his 25$ for showing up but still loses 225$ for the day. After a few days of playing your markets either he gets discouraged and doesn't come back or he loses all of his $ and doesn't come back. Not only are you hurting New Player A, you are also hurting every player in the game because the one thing that this game needs the most is more new players that succeed and stay.

I have been in sales many times over the years and have learned that the hardest thing to do is to get a new customer. This is why successful businesses will do everything that they can to keep every customer that they can.

I think that you and zvassil willingly giving up most of your wealth is a great thing. It will give many players hope that they may one day reach the top.

What is up with this statement: "Then we will fight with kenneth1 & cheesenips :-)" I would like you to know that at this moment I am #4 on the leaderboard eventhough most of my gain today is still in open markets. I am debating weather to sell and take the profits or let it ride and go for a larger gain. I may be optomistic but I would like to think that with a little time and a little luck I may be able to move up further on the leader board.

We may soon be able to end the debate on this topic because zvassil has posted a simple solution for this situation and I offered a few additions. Please let us know what you think.

@ kruijs

After reading the comments about news vs financial markets and about altering or adding new leaderboards I have a suggestion that may benefit everyone.

Keep the current Networth Leaderboard just like it is and include everything.

Create a new leaderboard just for wealth gained from the financial indicies markets.

Also create a new wealth leaderboard called the News Networth Leaderboard which will combine the wealth gained from all catagories of markets except for the financial indicies.

This way the news only fans can compare their predicting abilities without being overshadowed by those who make large profits in the stock markets and the financial indicies markets can continue to run like they do now.

Both sides should be happy because more players will be able to see their names on one of the leaderboards.

Yes add more Leaderboards. Me toping Science was an unexpected thrill ;-)

@ kenneth1

In reguards to your gray market experiment: Lets call it a real world scientific experiment to test a theory and it had educational value and no bad intent. Remember, you were just testing the system.

As someone above said: "We are still in beta mode!"

Play on and have fun.

@: charlesf: "Rich robbing others"

I watched zvassils raise for 6 weeks and learned to live with his three options bets in indices before I found his strategy on one of my markets. In Average he bought his favorite three options and nothing was wrong with the markets. One had to bet before him to get first dibs. The strategy could be a problem if it is concentrated on only one option making it 90% (risk to high) than everyone can´t buy that and there could be real disadvantages. I made more money betting against him and using slightly different strategies but my gains were not enough to keep the lead.

My guess it that I was more responsible for blocking out markets till July 18th without the strategy - when I watched Nasdaq or Dow in last options territory I know that in at least 50% of cases I would win - When a "-1% or worse option" got better for me -like 2,07% I could go to 70-80% if it was close to suspension and if it was -3% and more I could go all in and buy up to 90%. If we only get new selling rules that kill the strategy but don´t get a deeper market modell new problems with blocked markets will arise. Than richer and earlier players can make many markets unattractive for most players. Then we would need a maximal bet per market.

I watched today DAX and FTSE after I betted my 2 million on my options the rates were better than when only 7000/13000 were in the markets. Afterwards the percentages were 15/30/30/15 and pretty good for everyone.
Had I betted 50K without the strategy it would look like 5/45/45/5. IMO the model (without strategy) allows too little bets at low percentages (raises too quick). The second advanage is if there are 2 million in a market than the percentages raises very slowly giving all the oportunity to buy without exploding percentages.

It was a "big win - win" strategy. (loser was the bank - and to stop the bank robbery the model/rules should be fixed). The early players always have an advantage & a higher risk.

I just beted $500 in a market and the percentage rose from 25% to 40% - my thoughts "Fuck"!

I am very shure that bettung only a few 10Ks without the strategy would block out our options - that would allow only bets against. I have watched zvassils indices strategy a long time. I made lots of money on betting against him or by bying his options.

The only advantage of the rich is they can always buy their options with the strategy and get better percentages.

With a fair model it is possible to make about 1,5m in three month. If the market & question design offers big opportunities like last days crises than it is posiible to dobble or triple the money with all in. In my fight with zvassil we both were 80-90% invested daily. Last Friday I could have crashed down to 2m (cash was 300K 6,3m invested) instead I was right and up to 12m. The battles were as much fun as climbing the laederboard is.

___

I choose Ken and cheesenips because when I looked yesterdays these two where millionaires (by trying out the strategy or using the crisis).

I would love to compete with you for a "Non business leaderboard" and your "Prediction medal" - congrats - can´t post on your site.

@charlesf, hfl13:
1. I do not insist on using the "strategy" (I call it "pre-bet odds manipulation tactics"). And, it is not mine. It was used on HD on various markets and with widely varying results. What I would like is an explicit regulation or a change to the market-maker scoring function, and some degree of consensus on it from the community - I believe we are all here to have fun.
2. There is nothing very spacial about index markets - any market that is short-term, so you do not tie up $ for too long, and have 4 or more options are good targets (yes, election markets, sports winner markets, etc).
3. Every tactic has it's application area.
4. About robbing the bank - if you mean using the tactics causes larger bank loss, Hanson says it should not be so. If this is the case, there seems to be an implementation problem with the scoring function. HD also was plagued with scoring problems at extreme market volumes - I remember some precision loss errors.

@hfl13:
1. I am sorry I went MIA today - had to leave very early, and could not publish the US markets. Will catch up later, so you have some fun next week ;)
2. You all should already know that I tend to laissez-faire when markets are concerned. I did not share this intentionally, nor intended to enter a political argument - I am here for the fun of playing and winning, only.
If somehow I have offended anyone by presenting something more political - I apologize, it has not been intentional.
So, I promise I will try harder keeping to the technical side.
3. About the inactive accounts - I don't think there is anything wrong if anyone's open position's values change while he/she is away. This value is open position value - "money" on the table, they should change (else there is no chance of winnings).

@kenneth1:
Do you mean "lack of good investment opportunities"? It seems everyone with some real (or imaginary) capital to lose says that ;)

Nice now we know its the pbomt-strategy and you are not the inventor. IMO Hubdub had never this kind of problems if IIRC there only was one player after two years with 10m and he reached that by pushing one market with all his money to increase the networth.

In comparison to that the junps to 10M+ happen too fast and the modell is not behaving as it should. Rules/Fees should fix this. But markets/Some markets should be generally enlarged to allow bigger bets.


@zvassil last Friday I made 8m - nobody lost so it is the banks money or kruijs. Sorry for being flappy - I like to joke. What was political? Did I miss something? Please PM.

@hfl13: laissez-faire is a very political belief. This is not my game to make rules based on my beliefs.

@hfl13: "enlarge market" - do you mean market should be not so sensitive initially, behaving like initial odds are, say, 100K+?

"There is nothing very spacial about index markets"
well there is: they are in-play and along this, you can clearly follow the developments and trade predictions on that.
"Hubdub had never this kind of problems"
is wrong. Hubdub also limited the number of index markets as they were cash cow markets there also, alhtouhg there were more people playing them as heavily as you two do here. and there have been debates about them being valid prediction markets or not:
http://getsatisfaction.com/hubdub/topics/no_more_djia_the_morons_approach_to_climbing_the_leaderboard
http://getsatisfaction.com/hubdub/topics/business_indices_questions_restrictions_have_become_necessary
(etc)

also interesting:
http://getsatisfaction.com/hubdub/topics/the_royal_bank_of_hubdub

@zvassil1: I know laissez-faire but I thought i had overread a political statement
@zvassil2: Yes!
@kruijs: Are you trying to muddy the water = making lots more innocent but complicated suggestions like bank and interest or parallel markets for gray and black players?

@kruijs: Sorry, I have missed this. Do you remember when was this discussion?

@kruijs: I don't think in-play really matters, I can try to test that this weekend.

@hlf13
Maybe you were better at predicting who your main competion would be than I was. Most of my large inplay gains fell right along with the stock market on Wed. and I dropped from #4 down to #6. If I could have been at my computer I could have gotten out before it was all gone. Guess that is the price we pay for having a life away from KtN.

It can be all the exitement of the stock market without risking real money.
I didn´t play DAX or FTSE today.

@hfl, cheesenips...
I have also been trying to understand the "problems" on the site... And to this end, i have been investigating
some gains and losses, to see "where the money is coming from", like in Watergate, where Woodword and Bernstein are told "to follow the money" -- trying to see where the source of the problem was... so forgive me for looking at some aspects.. Cheesenips had a great day about 1 1/2 days ago (winning a LOT on the DAX market; the big loser, (apologies; we all make mistakes) was another big winner),so a LOT of money was redistributed, and this IS as it should BE; in other words, money was interchanged by people who won it, and gradually accumulates in the hands of smart people , like cheesenips.. Keep eating those cheesenips.. He is a supermouse..... People who bet big, can lose it to others who outsmart them at that ONE TIME...

But KtN also supplies all this money; now perhaps it is supplying too much, that is a question left to people "higher than my pay grade scale" as they say, like Kruijs, and superusers, etc.. .
In any case, I do not see a "problem" in Kruij's use of the software, although the probability function (see my recent note about "where the money comes from" ) to understand some of these aspects..
ken

@kenneth1: that was the exception because cheesenips played the very low percentages and won. IMO this could be one of the few cases were the loses of me and you (?) may have covered the winnings. Usually on indices markets zavssil and I won much more than the Ks on other options. The own substantial losses were the other options held by us. This is money creation like the American Fed just printing money. ;-) A month ago all KtN money was 3m now its about 75m. Case closed.

To my way of thinking , the amount that one may earn relates to various factors:
( My list may be incomplete, but listed are some of what I see).
(One important criterion will be the B value, a value that determines the amount one can earn per question)
1) How many questions settle how fast, In each market.
2) The amount one may "make" in each question (this relates to the B value).
I understand using a "high" B value is used....
3) Related to the B value, is the understanding that having the " high" B value reduces the pre-buying problem,
which sometimes is called "gaming the system". It is really just using the system, which exists now in place.
4) ONE BIG factor is how much money a person has, let us call this their "wealth factor", that he/she can invest. This is a big factor for the following reason.

The wealth factor, (for high B values), allows the player to invest a good fraction of his wealth, on just a few questions he/she may prefer, and make a certain % (a fairly high %, particularly , when high B's exist), when that question is settled.. Thus the person can obtain $ proportional to the amount he has divided by the time
the question settles in. Thus the rate of rise dW/dt (the rate of change of wealth) = K W,
where K is a factor depending upon the B factor, and the frequency of question settlement. The above is a simple, well known equation. It has a solution: W = Wo * exp (Kt) , where Wo is the wealth at some initial time, to. This is an exponential rise! So, this is what, i believe we are seeing, when we see the wealth of some individuals rising wealth.. They are getting better at using those questions, they can gain exponential wealth from. It is my opinion, that lowering the B value on the questions, would perhaps lower the amount that wealthy people can gain per question, but it is others' opinions that this allows more gaming.

It is my opinion that right now, the high B value, allows all of us to be "sheep" , just making money by eating the grass, because with a high B value, there is a lot of Sun shining on the grass, and the rich folks do not need to outsmart the little poor folks to gain money.. They can just eat the grass, but then there is less learning going on. When B is small (to my mind), it makes the rich no longer have the capability (unless there is a problem with the system, that allows the rich to make money on "pre-betting options" is a serious problem), and without it, they would then ONLY be able to make sums to get richer by becoming predators (outplaying the opponents).

Now they can get rich simply by eating the grass... so, we are all sheep... I do not mean to disparage people with lesser wealth.. partly the wealthy just have more time to "waste" on this system, rather than being productive elsewhere; the wealthy (on this system) are just more hooked..

Now I have a headache !

My ability to find the right option has not changed the last 18 month. The only differences are:
1. time spent to follow indices
2. Amount of luck (will turn)
3. clever Strategy alows playing more options
4. intensity of play = battle for position. In April and May I only was 10-20% invested the last successful days 80-95%.

Hi all, time to get a decision going in this.
Please vote up/down on the suggestions here:
Decisions

I like the suggestion of limiting the total individual investment in a market and want to advertise it a bit more:
Remember that this would not necessarily be a fixed value but in my opinion should be linked to the market volume in some way. This would advance the idea of betting against each other, reduce the "early bird" advantage a little and curb the described odds-manipulation (which is real, but inherent in all multiple bets on a single market - so simply banning it and assigning police duties to category editors and super users doesn't work).

Hi all, I just discovered this discussion and I agree that some issues need to be addressed. My enormous recent gains came about because while following the FTSE and DAX markets while in play during a volatile day, I happened upon several very low prices so of course, I bought those options. And I was lucky enough that the market moved in my favor before settlement. Actually I would have already become a millionaire the day before, except that my bets on low-priced options created 9-figure "will pay" values so that I was unable to sell losing positions because the "Sell" buttons were pushed underneath the vertical slider! (After I asked Kruijs for help, he reminded me that I can sell from my Profile screen, but that advice came too late-- anyway, I did so well the following day that it didn't matter.)

I didn't understand why such low-priced options had become available but, as a reasonably aggressive player, I bought them, especially because the price was so low that my risk was very small. Last night while playing the VIX, several low-priced options also became available. I did buy and sell them immediately, not having read this discussion. (I would have done better had I kept the ones I sold!)

I have not yet taken the time to read or consider the various excellent thoughts posted here already. My gut instinct is that something needs to be changed in the pricing model so that very low-priced options do not become available. Such low prices are not realistic in comparison to real world market prices. Is the problem that the index markets are so thin that certain options can from time to time have nothing invested, so that the pricing model creates outrageously low prices? If so then as someone has already suggested, perhaps a floor like 1% can be placed on prices.

I also agree that the goals and public recognition for this game should surpass net worth. I support Kruijs's creation of Community Points and will redirect some of my future efforts toward market creation and communication. Still, the successful prediction of outcomes and the resulting "financial" success seem to me to be the core attraction of this game.

I am happy that others are contributing their valuable ideas toward making this wonderful game more attractive and fair. Others have offered to give back some of their net worth toward such goals and I will do the same as the group comes to a consensus on such matters.

I created that VIX oportunities by betting about 300K till 600K on my four favorite options each the rest became incredible cheap 0%. With a few thousand K each I could buy insurances that were worth millions.
Yesterday my bets on Vix, nasdaq, DJIA and FTSE where my main options won but the Dax "about even" option won unexpected luckily for the 0,6% on even. DAX and FTSE were the easiest markets once - due to no corrrection to Wallstreet opening is being possible (suspend) and the euro crisis these are very hard to predict.

To repeat the problems that still are in the modell I repeat an old post:
My grudges with the current model:
1. very low percentages schould be cut off - buying 300K -1,4M for $100 is rediculous = no option should be below 0,5% in markets. Initial design requires at least 1% each option and there is always the possibilty that the FED sells gold or Warren Buffet buys Italy ;-) In most crises there is a day/s when there are huge gains because people come back and the volatility is generally high.
2. gains are too high with the clever strategy - change model &/ fine or block selling total/partially
3. model allows too little bets in most markets - you can only bet 20K if everyone shares one opinion or without the "clever" strategy IMO the markets should take about twice or threetimes the money that they take without the clever strategy.
4. I hate it when early bets make the percentage rise like a rocket (from 40% to 60% = $1200) or from 5%to 90% with $5000. = because of time constrains and the many bets to make, my 100 are dying out and my 2000/5000 are not so nice for me and the others who have to buy after me.

Yesterday I betted a few millions and made 14m with the strategy on the five indeces markets (4*right option + insurance limiting the loses on DAX).

This really needs fixing soon that we can continue playing normally.

Hi all,
My thoughts on the current suggestions
- Making index markets harder (or at least more expensive) to predict is good, and I have tried to go in that direction. However, I don't think we have to overdo it - there are already some lottery-style markets, if they are popular enough - let's create some more.
- If the main goal is to abolish "unlikely high return" odds - let's put a bottom on that. However, don't forget that Warren Buffet may really decide to buy Italy (or at least, Berlusconi's businesses) and a lot of people will suddenly find that a very interesting opportunity.
I don't think this measure would have been necessary if there were enough players betting on index questions. Maybe limit the rights to create such markets, or require formal approval?
- If the main goal is to abolish high bets, it seems best just to do that. Cash-in bans will not cut all high bets, but will cut all bets from players low on cash.
- I like most the "Cash-in ban for predictions which make up 25% or more of total market volume" with some changes:
1. cash-in block for 2 hours, for all bets of a player on a given market (i.e. player's position is "frozen"), when -
A) a bet increases the cumulative player's position on all options of this market past 25% of current market volume. (BTW this means the block could be lifted earlier if market volume increases sufficiently) and
B) market volume should be at least 10x the initial "pseudo" volume of the starting options for the "freeze" to be applied.

The intention is to allow smaller bets without any restrictions, and to automatically adjust for markets that are highly sensitive from the beginning.
I have made some calculations by hand, but if you consider this version interesting and not very expensive to implement, some experimentation will be necessary.
- I don't think the "early bird better" is a problem - the market creator is restricted from early betting, and he/she can chose the time of publication so that "early birds" would have to stay very vigilant in order to succeed.

"I don't think this measure would have been necessary if there were enough players betting on index questions. Maybe limit the rights to create such markets, or require formal approval?"
Good point. Hubdub also limited inplay index market questions besides US markets because of the lack of interest by players.

I also like the market volume-related cash-in ban, but there are still markets that are very thin for quite a long time, where you have days or weeks of no interference by others. That's where a 2 hours ban won't mean anything.
If I understand B) correctly, you don't want a freeze to be applied on a newly created/thin market, but this is where manipulation is cheapest (with relatively small gains, I agree). Is the initial "pseudo volume" really that small? Ten times as much sounds quite a lot to me.

By "early bird" advantage I mostly mean people getting breaking news on a subject and being the only big winner by investing several thousand $, which is ok but it wouldn't hurt too much to let others have a chance as well. And like I said, I'm all for abolishing extremely high bets, but in relation to the market volume. In case some more people agree on that we could talk about the exact numbers and functions later.

Please think - I alone can put the market volume to 5m or 10m and that is only with 100K bets.

"Breaking news" can be caught by finding the earliest news. If I see markets that moved 20% or visable the last day I suspect betting after the fact and sort google by date - than I define when the news broke - with that time I settle - biggest part of the problem solved.

I do think. What I considered a problem was the fact that you can exactly do that: make a 100k bet while this makes up like 90% of the market volume.

And for the news part: Not every news story carries a fact that's sufficient to settle a market.
By the way: I sometimes found newstimeline.googlelabs.com useful to look for the day first and then use the "sort by date" function.

Charlesf...Was looking over the writings because of the questions posted separately..
re your example on the Sarah Palin question... I think the results you site, are somewhat off..
because once the betting reverts (whether in small 100$ bets or 1 large bet FOR HER), you will no
longer be able to get the money you posted against her BACK... This is my understanding...

I did some tests and found this to be right.. Can you run a test (say on a question which is new before
anyone places any money)? I did this, and found it seems to hold (roughly in the cases i tried..).
Tell me how the odds moved, and your money moved, and how much you would gain in one particular choice made from a direct bet, and via the strategy of preselection betting...I could not get a significant difference..
(I did a couple of these tests)..(and i then took out my $, and incurred the 10% loss, so it looks like everything
was working fine)...

To me, ideally, and how i view it, and I would be interested in how others view the "betting odds/returns, etc.," I view it as follows ---

KtN puts in a chunk of money (and this chunk relates to the so-called B value) in the program, and this money
is available, say for N options controlled by the % chance of winning...
Now, if an option, say M, has a small chance, then a person who bets it, and if it wins, stands to gain a whole lot (but only up to the B value).. If others put $KtN in , and say they bet on other values, but a person still believes in M, and puts more money on M, and then wins, now he gets the $KtN money , but also the money that others put in... but $KtN doesn't lose any more money..

It is because of this, that i question the efficacy of bidding the price down, and then buying that option, etc..
(NEvertheless, I do see people's assets rising at very high rates, but I guess they know stuff i don't know)..
I sure could use some of their brains.. Glad though I am limiting my betting to KtN, and now with $US or euros or anything of the sort..

Yes, it is NOT working "IDEALLY"... I did some experimenting on hurricane issue questions with their predictions.... since i know the distribution follows an exponential decay, same as radioactivity... a
Poisson distribution exp -t/tau.... It is a slowly decaying function, however, hurricanes can be forecast
a few days to a week ahead, and there are already some brewing near the Caribbean, so i mayhave gotten there a little too late...

In any case, I was able to drive up the odds (for later ones), since big chunks were already invested in the earlier names, down to roughly 4,000 to ONE, so for 100$, i could get back $380,000, and since exp -4 is only about 2%, i was getting near 20:1 odds greater than i should have gotten...
NOW IDEALLY, the peoople who previously bet should have paid me that money, BUT THERE is no way, they are putting that amount of money into the market, as hfl13 says (and i believe him).. So, it is coming out
of my taxes (NO wrong country), no i mean out of KtN.

So, i think the formula could use tweaking or one of the suggestions (i think originally by Hfl13 or perhaps kruijs)
to limit the low betting probabilities might be good, particularly the starting ones, because they are most subject to abuse (if done by others)... or creatively (if done by ourselves).

You want to limit us low bidders???

Yes! Republican theory 101- millionaires can´t be wrong :-)

Seriously my proposal was no option could fall under 0,5% or 0.8% to deny someone the chance to make 20K, 80K or wvwn 1,6m with a simple 100.

@hfl13: As far as I can tell, it's Billionaires can't be wrong, you millionaires - don't jump the queue, you paisans ;)

Hi all,
what if the offered odds were not based on bets-on-this-option / all-bets, but on
(shares-bet-on-this-option + $to_win-for-shares-bet-on-this-option) / (all-bets + all-$to_win)

This does not seem to be a Hanson marketmaker function, but surely can augment the offered-odds extremes i the right direction.

Yes to zvassil's overall suggestion of a modified formulae. I have been working on this. This was at the basis of some of my other comments about how i think the market "should work" , (didn't express it as shares, because we have been using %, or fractions, or really "real numbers", which is at the heart of this basis). So "no" (if i understand the suggestion properly of shares, for the following reason) -

To me, shares implies a fixed number of quantized bits (a share), which someone must be willing to sell. We have been using Real numbers. Real numbers are infinitely divisible, whereas "shares" are essentially integers, and therefore come quantized in little bundles that are not divisible. When someone buys one, then they get to sell it, and KtN would need to sell their share, so they would have to put in the price they would be willing to sell at, etc.. If this is what is being suggested..... Unless KtN, prints "new shares", then until someone is willing to sell shares, they remain unsold, etc. and it is too radical a change (if this is what is envisioned). Perhaps i am misinterpreting what you have in mind Zvassil.. If so , sorry..

What i am working on, would be a modified formula, that would leave "things" as they are, (so we could continue in the same vein), but the formula would be different, so $ would not go through the roof, via market manipulation..., etc..

I had in mind the share of all bids on a given option, not the company/financial meaning of share.
Real numbers in computers are not infinitely divisible - they are represented as a finate sequence of bits, so one has to chose from a finite set of representations (rounding).
The other thing that is possible when a bid (or a series of bids) results in generation of extra $ - the bank could probably invest (by betting) an equal (or proportionate) amount in the lowest-odds options. Something like the bank buying the best bets. (QE KtN ;) )

Actually, how are the points for "predicting" awarded? I could not guess it from the notes at the bottom of the page.

What would you say to a measure of "precision predicting", like -

for each market
(number of winning options with players bids) / (number of options with players bids)

and the weekly/monthly sum of these numbers is the score?

From comments with Kruijs, it seems that modifying the Hanson formula might prove problematic
to this market, so i am forgoing that option and have asked it to be withdrawn, so feel free to either vote against it, or not for it, as suits you... I can recognize that the Hanson market maker formalism although problematic, is one that has a history... I am totally supportive of this viewpoint, and shall go and vote against it myself,now..
Ken

There is only one solution to this problem.
Changing the market model from Hanson's log market to a conventional bookmaker where the betting is against the bookie and not against each other.
I made a spread-sheet that allows fixed market volume adjustable to the number of players and their bets, from 100 players up to millions.
Although I advocated this model from several angles, and the current model is proven a failure that allows users to get very rich by manipulation, contrarily to the initial use of the log market, my arguments reached the eyes of our 'boss', but they, very surprisingly, did not reach the game play part of his brain.

I have the following questions about the validity of this poll:

1. Why was the alternative of another market model, eg.a conventional bookie, not part of one of the initial options ? The current system has been proven a failure of considerable size. Any amendments made to this model will end in a failure that's also sinking into a swamp. I am beginning to understand why it is assumed that it will take new members months to understand the game.

2. Why was I not informed about this poll until yesterday (by zvassil) ?

"The current system has been proven a failure of considerable size" ummm, no?

"Why was I not informed about this poll until yesterday (by zvassil) ?" it has been on the frontpage all the time (as a wall post), and additionally in the originating discussion thread.

but the question about the validity remains, of course.

It really puzzles me how you still want to defend the current market model.

1. It was installed originally by your own saying to favour small bets to keep the amounts on the leaderboard within the reach of starting members to compete. A good, understandable and respectable idea. But it backfired as never expected.

The top five of the leaderboard own 50, 34,12, 7 and 6 million $. Not by brilliant predicting, but by manipulating the markets because of a flawed model.

2. All kind of actions like early suspending, banning popular markets, forum discussions and a poll about amendments are now required to keep the current model afloat. These actions do not help generating traffic or transparency of the game at all. Let alone the system that cannot be repaired by whatever new rules you are going to apply. It just does not work.

The possibility of manipulation in itself should be reason enough to get rid of this failing system.

You are without any doubt a brilliant IT-engineer and a great organiser, but to me, at the moment, it looks like you have a blind spot for this miserable system comparable to a total eclipse.

If you do not open your game play part of your mind you will keep on struggling with this model to no result whatsoever.

It's like flogging a dead horse.

Do you realise that by just stating that the system doesn't work and that every change to it won't make it better you don't discuss it at all? Although I wasn't exactly happy with how the discussion went, there was a number of arguments brought up and you had/have the chance to reply to them.

Please explain who this bookmaker would be exactly and how he would guarantee "valid" odds at any given time, so that wealthy people and very regular players don't get exponentially richer even quicker than with the Hanson market maker.

I see a lot of heated discussions.... As far as the market wealth, and finding the "kinks", i think it simple that we regard the current wealth as a "breaking in period" of the new system, and after we reach some kind of "consensus", with appropriate homage to Kruijs for doing 99% of the work, so the people who do the work, should be able to have some greater say, as to how it goes, because we do not pay his salary, etc.

I have been studying some of the problems, some in consternation to others. nevertheless, also have been reading hanson's and others' reports of these markets. Much to my surprise, they can allow such manipulation (as we have found out).. Here are some- http://hanson.gmu.edu/combobet.pdf
and this is a particularly good one - http://hanson.gmu.edu/combobet.pdf
It points out for example - The.. market maker also stands ready to make any large bets that can be constructed from tiny bets. The only catch is that the prices p change as tiny bets are made.. and ...It seems good to minimize the extent to which people making bets on some dimensions of the prob..distrib.regularly cause unintended changes to other unrelated dimensions of the distribution... It goes on to elaborate.. He also discusses how the number of states grows , when there are multiple markets with multiple states, and this (involves calculations , which has led to some time delays, perhaps)..

Anyway, aside from working on trying to find an approach similar to Hendricks, but also considering other options which are less radical, i came across an idea, that might be easy to implement. Am not a great software engineer, but here is the idea. Since money can be made with the current setup, using the following strategy, say ONE wants to buy option 1, because he thinks it will win, or even sell it later, by first buying other options, say 2-N, and thus driving the price of 1 down, then buys 1. then sells 2-N, and 1 shoots up in value, this creates a false kind of wealth which shows up on one's value estimate as real wealth, (until the market pays off, etc. or one sells with the 10% loss, etc.). One simple way to avoid this, is the following. If one has some % of market, say >10%, and one has more than one holding in the market, then if he sells ONE, the value of his others is held constant (they do not go up).. He gets his money back from the sale, but the value of the others is simply frozen. This seems fairly straightforward to implement, and i think totally does away with market manipulation, unless i am missing something. One could even eliminate the 10% rule.. If this looks like a good idea. feel free to post some simple wording of it....
Let me add, we all owe Kruijs a vote of thanks for trying to resurrect Hubdub, and he has worked tirelessly in these efforts, and having all the shenanigans lead to lots of pressure cannot be fun.. so let's tone down the rhetoric, and find our way together towards a reasonable solution... Sorry for the length..

@ bernardo
I have discussed with Stijn from the start ca. 9 months ago, that the current system of distributing the odds and pay-outs is not the way to calculate an internet game in these days.
The current Hanson model looks to be a pari-mutuel model that divides the pot between the winners of a given option. This only works for thin markets where no big bets are made.
This pari-mutuel model also results in raising the price before the bet has been placed. The increase of the total pot is percentage-wise less than the decrease of the share of the pot of the individual winner (winning group and bets is smaller than the total of players and bets).

This makes the changing of the odds of the option breaking the golden rule of the bookmaker: do not change the price until the bet has been placed. For the regular better this feels like ' cheating ', although according to pari-mutuel rules it isn't.
Therefore a 'long time ago' I made a spread-sheet for Stijn to consider. This way of calculating the odds and pay-outs allows adjusting the distribution to the number of players and their bets.
Eg. a fixed market volume of 100 Million $ combined with a maximum of 100 K per bet decreases the volatility in such a way that makes it impossible to manipulate the market, and allows all kind of markets that are currently banned, or suspended long before closing time, or not even made anymore (eg. will Dow open up ?.
It is not my choice but the children of the now ruling culture want winning, and they want it now. To satisfy their needs, and increase traffic of people and new ideas, you have to use the popular markets that pay-out within hours. Even markets that pay-out eg. every hour is very desirable and possible with a conventional bookie.
Even when a player gets so wealthy or the number of players and bets come near to manipulating we increase the market volume to eg. 500 Million.
And all the banning and early suspending and other actions to mend the current system become obsolete once the conventional bookie is installed.
Stijn as well as the admins all have this quite simple spread-sheet. Feel free to try this system.
---

I have tried to advocate my point in several ways during the last months but it time and again hit a wall.

That's why my recent post uses the rhetoric of a tired man, who nevertheless admires the great efforts and results that have been made.

@kenneth1: It does away with the market manupulation to increase immediately the value of players' other bets; it does not afaict prevent from getting better odds at the other options (high sensitivity)

@zvassil, I do not understand what "it" refers to.. so cannot follow your argument..
@hendrink ,, Not 100% sure what yours does .. havent studied parimutual... Happy to look at your
model should you wish...

I messed up, with the Hanson two articles, and looks like i only sent the same site 2x.
i read a few of his, and some others... I think this is the one i referred to : http://hanson.gmu.edu/mktscore.pdf

Happy with anything that works, and I will forward a spreadsheet to anyone who asks, about
the methodology i suggest (am going to have it ready in a couple of days.. .. dealing with the aftermath of
earthquake, and grandkids -- which is worse than the earthquake, as that only lasts < 1 minute... :) )..

I really think the new value of the "liquidity" parameter makes the markets react far too slowly. In markets with a volume of some thousand dollars a 500 $ bet doesn't have any real impact anymore, which means the initial odds and the first bets together carry too much weight and afterwards only very heavy bets change the overall prediction values significantly. This is somehow inconsistent with the idea of "wisdom of the crowds", which I know many of you don't really care about but in my mind it's a central part of the game.
Maybe it's possible to find a compromise that lets markets "live" longer over time (meaning that they can still react to new developments of long-term news stories) but still doesn't make the initial values too easy to manipulate.

I agree with bernardo. Before the change, in the Dow Poker market a 1000$ position on an option at 25% would raise it about 20% to around 45% (too much). Today in the same market a 2000$ position in the same option at 25% only raised it 1% to 26% (too little). Going from 1000$ = 20% to 1000$ = .5% is going from too volatile to too stable. How about if the settings are adjusted by a factor of 5 so in this example a 1000$ position would move the option up by 2.5%. Also, after the roll back for the top 5 there will be much less need to protect the markets from huge positions.

With a little bit of luck this may be the final adjustment needed. We need to bring in more new members. That, combined with the growth in wealth of todays white and gray players should be enough to stablize the markets to compensate for the increase in wealth of the top players. This increase should be more moderate than it has been for the past few months.

you are right, I experimenting a bit right now. what would the correct behavior be in your eyes, for a 1000$ stake on a 25% option?

25% + 1000$ -> 30%

25% + 1000$ -> 35%

25% + 1000$ -> 40%

25% + 1000$ -> less than 30%

25% + 1000$ -> more than 40%

I gave my thumbs up to 35%, a happy medium. 35% puts it nearer to where it was in the spring when things seemed to be going smoothly and before some of us started growing too fast. The higher % will also help to compensate for any markets with less that perfect starting odds. This should be fine for the financial markets and other markets with a large number of options. These tend to draw the largest amount of $ but most larger wagerers will make large wagers on more than 1 position keeping the favorite option below 50%.

Will these adjustments also work well with markets with only 2 or 3 options? In these, a large wager would seem to have a bigger effect and also a large wagerer would be more likely to only bet on 1 option.

If the new settings don't work well for all markets would it be possible to have different settings for different markets? This could keep all markets equaly flexible. If it is possible they could be based on the number of options that the market has. The arithmetic part should be easy but would it be possible and practical to do it?

IMO a +5% for only 1000 $ in much too high maybe 1% till 1,25% higher.

"for only 1000 $"

remember: that's an "all-in" for a newbee...

I'm a $100 player usually. But, that's just my mentality. $1,000 bet would be betting the farm.

1000 on 25% now makes it move to 31.8%

let me know how it feels.

Something is wrong with the odds for both of the new financial markets. First wager of 100$ moves odds to 100% and all others to 0%. A 100$ wager on the others registers at 1%.

Same problem with the 2 new entertainment markets.

you're correct thanks. I fixed it.

You better be careful. We might expect you to fix every problem this fast.

Thank you.

What will be the daily change of the DAX on 15.09.2011?
Still something weird - 25000@6% -> 73%, buy-at price 73% - all of it?

no, not specifically weird. the 25,000 are simply a huge trade.

aha

@kruijs - Net Worth:+673,391,432,602.22$

We just got done with the "clever stratagy". What do you call your new system?
I am impressed but guess that you are doing things that the rest of us can't.

Hi Kruijs,
what about offering a grouping withing Profile/Positions to show all engagement of each market in a single comprehensive line?