Page 1 of 2

Manipulate the stat price

Posted: Wed Nov 21, 2007 5:23 am
by Guest
Dear all,

I'm thinking about the method of manipulating the stat price in order to earn better profit.

The concept is simple:
:arrow: reload goods in the high stat price period
:arrow: dump goods in certain time but never kill the future price

Following is an example for illustration.
Settlement stands for the settlement of stat price.

:) Settlement as high price (initial)
:!: Reload as high price
:arrow: Sale
:) Settlement as high price (*)
:arrow: Sale (Key point, this sale must cause following settlement as high price)
:!: Reload as low price
:idea: Do not visit store at all
:) Settlement as high price
:twisted: Dump
:twisted: Dump
:twisted: Dump
:!: Reload as high price
:( Settlement as low price
:arrow: Sale
:arrow: Sale
:) Settlement as high price
:arrow: loop to the (*) position

There are several necessary elements for such method:
1. the one (or group) must dominate the industry.
2. must have enough stocks for 48 hours dumping.
3. revolving cash must be enough since there is a period we can't visit stalls.

I am experimenting this method now...

Posted: Wed Nov 21, 2007 5:45 am
by Guest
Could you please explain why you think this method will raise the stat price more than simply selling as high as possible?

Posted: Wed Nov 21, 2007 6:33 am
by Guest
This method will "not" raise the stat price more than simply selling as high. However, it is possible more efficient than simplex one.

There is an assumption here: the settlement is only related to the stall behavior in last 24 hours.

Actually, I didn't realize what is the exact meaning of "selling as high as possible". Higher price stands for not only higher incomes but also lower efficiency. That is a trade-off.

If we always concern about stat price, the efficiency will be poor. We have to build many stalls for selling little amount of goods with high price. On the other hand, if we always calculate the best efficiency, the stat price will go down in most cases. Either policy may not stand for the optimal solution for this game.

I purpose to find a method to earn better efficiency by dumping but never kill the economy in future. There are two stages in this method.

1. normal stage: selling as high as possible, just like you mentioned before, whatever it is. My behavior is no different to other players who concerned about stat price.
2. dumping stage: I calculate the best efficiency to sell my goods. Such behavior must cause the stat price going down.

Above two stages work together alternatively, one after another. After dumping, it does cause a worse stat price but it doesn't matter. If you trace the progress carefully, you may found that I could always reload my goods into stalls at the high price period. Even dumping stage caused a poor stat price, there will be no reload in the period. Actually, the normal stage will pull up the stat price in next day.

There are several features under such method:
* I earn the same profit in the normal stage.
* I earn the better profit in the dumping stage.
* I didn't bear the consequence caused by dumping since normal stage revive the stat price after that.

That is the reason I thought it might work. After all, everything is under experiment.

Posted: Wed Nov 21, 2007 12:37 pm
by Guest
Does this theory involve all those people that always undercut prices?

Posted: Wed Nov 21, 2007 12:58 pm
by Guest
Greenhouse wrote:Does this theory involve all those people that always undercut prices?
There is an important condition I mentioned before:
"the one (or group) must dominate the industry. "

If you have only minor effect to stat price, it won't work.

In that case, it is very difficult to manipulate the Beer industry since even to keep a stable price is already very tough. On the other hand, it might be possible to control some goods which is not involved by many cities.

Posted: Wed Nov 21, 2007 2:09 pm
by Guest
This method of price manipulation I think will work less in a cartel, because there is the opportunity for members to dump before the set time.

Generally, I don't think cartels will work, especially in the long run. You are better off keeping your price just a little higher if you have the power to make more profit that way.

Posted: Wed Nov 21, 2007 3:18 pm
by Guest
:cry: i dont get it? just raise prices for say 3 days. wham stats price goes up

Posted: Thu Nov 22, 2007 2:40 am
by Guest
i aint sure how dumpnig at certain times wil raise the price up????
anyway like ganma says realy, just all have to sell slightly above statistic and price wil go up eventualy....so no dumping!!!!
i wish i could effect it but even though i have 30 taverns up i only hold
like 2% of the beer industry, so my gues is that there must be 50 people here that have atleast 30 taverns each or people sell in 12 hour cycles?
and dump them, since 328k beer per day is alot

Posted: Thu Nov 22, 2007 4:33 am
by Guest
First of all, dumping will never raise the price up.

The reason why members won't get hurt by dumping is "do not reload stalls during the low price day".

[Initial day: high price]=>[Sale as high price]=>[Reload as low price, but do not sale until next day]=>[next day: High price]=>[Dumping]=>[Reload as high price]=>[next day: Low price]=> :!: [Sale as high price]=> :idea: [next day: High price]=>[Reload]

Watch the Exclamation mark, we could still sale as high price even in a low price day. After that, it cause another high price sattlement. Under such method, members could dumping a whole day but no one get hurt. Even future economy won't get hurt as well.

In addition, I have to emphasize again. It is necessary to have the capability to dominate the price result. If you have only minor effect in the industry, such method shall fail.

Posted: Thu Nov 22, 2007 8:04 am
by Guest
wel for beer prolly noone dominate it since as i say 328k beer and increasing each day is alot and i doubt anyone holds 5-10% of that amount cause they would make 4 mil a day then just on beer

Posted: Thu Nov 22, 2007 8:38 am
by Guest
If anyone can dominate the beer market they win the game. :cry:

Posted: Thu Nov 22, 2007 8:39 am
by Guest
miauwally wrote:There is an assumption here: the settlement is only related to the stall behavior in last 24 hours.
And that is an incorrect assumption. Stats are somewhat mysterious, but we do know they are averaged over several days.

Posted: Thu Nov 22, 2007 9:02 am
by Guest
Knolls wrote:
miauwally wrote:There is an assumption here: the settlement is only related to the stall behavior in last 24 hours.
And that is an incorrect assumption. Stats are somewhat mysterious, but we do know they are averaged over several days.
You are right, I'm aware of it already. The average price may not only related to last 24 hours. I'm trying to finger it out.

I guess, it might be the combination of both previous stat and behavior today. For example:

Current stat: raiment 2,352(pieces) 510.10(¢τ).

If there are 3,000 pieces of raiment sell as 520¢τ, the new stat price might be calculated by following:

(2352*510.10)*F+(3000*520)*(1-F) where F is a portion factor between 0 to 1 (probably 0.5).

Of course, it might not be linear anyway...

Posted: Thu Nov 22, 2007 9:29 am
by Guest
its roughly

half of today's(t) sales plus a quarters of yesterdays sales(t-1) plus a eighth of the day before thats sales (t-2) and so on.

Posted: Thu Nov 22, 2007 10:10 am
by Guest
I'm appreciated. It is a good news since the formula might be linear combination of previous stat.

If so, there should be some kind of simplex form.

We could defind f(i) = number * price in i-th day.

f(t) = f(t-1)*(0.5) + f(t-1)(0.5)^2 + f(t-3)*(0.5)^3 +...

In that case, any behavior three days before only contributed a minor effect which is less than 3%. Then, it is possible to form a rotation for manipulating the price and dumping in certain period...