Idea Math Proposal
"You say you want a revolution,
Well, you know
We all want to change the world...."
In the wake of the revaluation of the ideas market, I think the first key point of any new pricing system is that it has to tie idea pricing to the shares market. Before the revaluation there was basically no way to push enough cash through the shares market to make it worth playing compared to the passive interest of the ideas market. Since the revaluation it has become impossible to invest the gains from the share market into the undervalued ideas market. It is predictable under the current math that at some point the ideas market will grow to a balance point, but soon after will once again grow beyond the shares market.
The second key point considered is that the game deserves better than the investment decision 'more than 50 blogs, yes/no?' Now that I have experienced the far greater intricacies of share play I have no interest in returning to filling a portfolio with ideas and checking once a month for industries that have crept over 50 blogs. Every industry should respond in an individual manner.
Accomplishing this requires adjustment of prices, but also requires a system that will not only produce ideas, but consume them as well. The following system uses price and quantity controls in concert to make industries cycle.
No changes are required in the spiders that produce the 'drop', only in how the drop is processed. Rather than just adding ideas endlessly to the market, for each industry in which ideas are produced the system compares two quantities:
D=the number of ideas in the drop
M=the number of ideas on the market
to determine:
N=the new number of ideas on the market
There are three cases.
For M < n="M+D" n="M-D"> 10*D, N=.95*M
The drop will add to the availability of ideas if there are few or none available while consuming ideas if there are large quantities idle on the market. So "dead" industries will see their abandoned ideas get consumed away, while industries that are "in demand" will see continued production.
To create a cycle then, pricing needs to be tied to the actual quantity of ideas in existence. This is done in the first term of the pricing equation, the upward force term. This term also uses two quantities:
D= the number of ideas in the drop
E= the total number of ideas in existence
to compute:
U= upward industry force
1000*D
----------- = U
E+1000*D
So, if there are no ideas in existence, U=1, and as ideas accumulate it declines towards zero. After 1000 drops the average drop produces U=.5, after 2000 drops the average drop produces U=.25. In many cases the size of the drops can vary widely, so there is always variation in the value of U with smaller than normal drops generating lower than normal values and larger than normal drops generating larger values, but in the long run upward industry force will be affected by the trend of ideas accumulating or being consumed.
Obviously, with an upward force varying between 1 and 0 we need a downward force. In the next term idea pricing is tied to the rest of the game. Again two values are considered:
I = total value of all existing ideas
W = total worth of all the players
to calculate:
D1 = downward market force
I
-- = D1
W
Obviously, this is simply the fraction of the players wealth that is invested in ideas. If the players collectively invest more of their money into ideas, or the amount invested grows through price increases, that will produce a downward force, tending to drive prices down.
So the basic pricing function is the interaction between U and D1, where:
P0 = previous price
P = new price
P = P0 + P0(U - D1)(.1)
This generates a cyclic system. Take an industry with relatively few ideas compared to the average production, generating high values of U. The price will tend to rise, making it a good investment. Since it is a good investment it will tend to get purchased, so there will be few on the market. With few on the market drops will produce, adding to the accumulated ideas, which will eventually rise to a level where the values of U that are generated no longer maintain the rising trend in price. At that point investment in the industry will slow, causing ideas to languish on the market, reducing production and eventually actually consuming ideas.
This is sufficient to create the general market cycle, but since each industry will react individually there will always be extremes, and the extremes have to be managed to keep them from becoming too extreme. The upper extreme is a consistently growing industry, where ever rising drop size maintains consistently high values of U. This would be a good investment industry, but at some point could become such a large contributor to the total ideas value that it drives other idea prices down. To prevent this a limiting force is used, calculated from:
T = total value of the industry's existing ideas
N = the total number of industries in the game
I = total value of all existing ideas
T * N
---------- = D2
I * 100
Basically, the value of D2 becomes significant if the value of an industry reaches 100 times the average industry. At about 1000 times the average it will dominate the equation and force price to decline. Making the price equation:
P = P0 + P0(U - D1 - D2)(.1)
Limiting the opposite case is perhaps even more important. If a large quantity of ideas become 'locked' in portfolios then the value of U is held down, causing prices to drop endlessly. The easiest example of how this could happen is me. I hold over 90% of the ideas in a particular industry. If I quit playing those ideas would drive price down in that industry even if everyone else sold their ideas and they were consumed in the market. Taking a glance at reality we see the solution: a dead and decaying industry can be revitalized by a revolutionary new idea. When price drops below a minimum threshold, say .25B$, a new idea is created and all the existing ideas in that industry become obsolete and are removed from play. The low threshold ensures that the disappearance of all those ideas will not have a huge impact on anything or anybody. The next drop in the industry will effectively be in a new industry.
Which brings us to the question of pricing in a new industry. B$10. A new industry will generate high values of U and drive the price up. How fast and how often will depend on industry factors and will vary greatly, so they can all start out the same.
I'm no coder, but I think compared to the array of conditionals in the current pricing system the single equation for pricing and the small conditional for production/consumption should process relatively quickly, and since only industries that are in the drop need to be processed it should go very quickly.
I've run plenty of sample drops through a spreadsheet with the basic calculations built into it, and believe the system will work, and work well, but it is predictable that there will be some difficulties in implementation. These are all inherent in switching from a geometrically expanding unlimited system to a cyclic system. Basically, a lot of industries exist that are so far from their cyclic norm that they will have to undergo some radical shifts to get on track.
Catagory 1: An industry with a few blogs that has produced only a few ideas. This industry will have a very high price compared to new industries starting at B$10, and might always cycle at some higher than normal value.
Catagory 2: An industry with a few reasonably small blogs that has been reindexed daily or near daily for a long time. Price here is going to be very high, since it was a 'rare', but the huge number of ideas in existence will mean no upward force, and the high price and large number will generate a large D2...in some cases a really large D2. An industry with a total value greater than B$20b will generate a D2 greater than 1, as of this writing.
Catagory 3: An industry that has a lot of blogs, a lot of big blogs, and a huge pile of ideas. In most cases these prices have been pounded through the floor by the existing system, and since the number of ideas is so gross the new system is going to continue to drop price until enough ides get consumed to change that. In many cases it is likely that price will reach the revolution threshold before that happens. The revolution was intended to be a rare event, but during the transition it might be an everyday event.
Ongoingly there will be a need for adjustments. In the coding there should be constants; a multiplier on each of the terms, the '1000' in the U term, and the '.1' in the price change equation. The multipliers on the three terms are initially set at 1, but they can be used to shift the balance up or down by varying them between .5 and 2. Specifically, lowering the D2 constant initially will reduce the transition impact on the severely bloated false rares. Changing the 1000 in the U term will impact the upward force in all industries, and will also affect the frequency of cycles. Lowering the value will reduce the upforce immediately and tend to shorten the cycle. The .1 in the equation is a 'volatility adjust'. Initially, due to the transition, there will be some powerful forces generated in most industries. Over time we might find that the cyclic nature 'damps out', in which case this would be adjusted up.
That's my input.
Passer
Well, you know
We all want to change the world...."
In the wake of the revaluation of the ideas market, I think the first key point of any new pricing system is that it has to tie idea pricing to the shares market. Before the revaluation there was basically no way to push enough cash through the shares market to make it worth playing compared to the passive interest of the ideas market. Since the revaluation it has become impossible to invest the gains from the share market into the undervalued ideas market. It is predictable under the current math that at some point the ideas market will grow to a balance point, but soon after will once again grow beyond the shares market.
The second key point considered is that the game deserves better than the investment decision 'more than 50 blogs, yes/no?' Now that I have experienced the far greater intricacies of share play I have no interest in returning to filling a portfolio with ideas and checking once a month for industries that have crept over 50 blogs. Every industry should respond in an individual manner.
Accomplishing this requires adjustment of prices, but also requires a system that will not only produce ideas, but consume them as well. The following system uses price and quantity controls in concert to make industries cycle.
No changes are required in the spiders that produce the 'drop', only in how the drop is processed. Rather than just adding ideas endlessly to the market, for each industry in which ideas are produced the system compares two quantities:
D=the number of ideas in the drop
M=the number of ideas on the market
to determine:
N=the new number of ideas on the market
There are three cases.
For M < n="M+D" n="M-D"> 10*D, N=.95*M
The drop will add to the availability of ideas if there are few or none available while consuming ideas if there are large quantities idle on the market. So "dead" industries will see their abandoned ideas get consumed away, while industries that are "in demand" will see continued production.
To create a cycle then, pricing needs to be tied to the actual quantity of ideas in existence. This is done in the first term of the pricing equation, the upward force term. This term also uses two quantities:
D= the number of ideas in the drop
E= the total number of ideas in existence
to compute:
U= upward industry force
1000*D
----------- = U
E+1000*D
So, if there are no ideas in existence, U=1, and as ideas accumulate it declines towards zero. After 1000 drops the average drop produces U=.5, after 2000 drops the average drop produces U=.25. In many cases the size of the drops can vary widely, so there is always variation in the value of U with smaller than normal drops generating lower than normal values and larger than normal drops generating larger values, but in the long run upward industry force will be affected by the trend of ideas accumulating or being consumed.
Obviously, with an upward force varying between 1 and 0 we need a downward force. In the next term idea pricing is tied to the rest of the game. Again two values are considered:
I = total value of all existing ideas
W = total worth of all the players
to calculate:
D1 = downward market force
I
-- = D1
W
Obviously, this is simply the fraction of the players wealth that is invested in ideas. If the players collectively invest more of their money into ideas, or the amount invested grows through price increases, that will produce a downward force, tending to drive prices down.
So the basic pricing function is the interaction between U and D1, where:
P0 = previous price
P = new price
P = P0 + P0(U - D1)(.1)
This generates a cyclic system. Take an industry with relatively few ideas compared to the average production, generating high values of U. The price will tend to rise, making it a good investment. Since it is a good investment it will tend to get purchased, so there will be few on the market. With few on the market drops will produce, adding to the accumulated ideas, which will eventually rise to a level where the values of U that are generated no longer maintain the rising trend in price. At that point investment in the industry will slow, causing ideas to languish on the market, reducing production and eventually actually consuming ideas.
This is sufficient to create the general market cycle, but since each industry will react individually there will always be extremes, and the extremes have to be managed to keep them from becoming too extreme. The upper extreme is a consistently growing industry, where ever rising drop size maintains consistently high values of U. This would be a good investment industry, but at some point could become such a large contributor to the total ideas value that it drives other idea prices down. To prevent this a limiting force is used, calculated from:
T = total value of the industry's existing ideas
N = the total number of industries in the game
I = total value of all existing ideas
T * N
---------- = D2
I * 100
Basically, the value of D2 becomes significant if the value of an industry reaches 100 times the average industry. At about 1000 times the average it will dominate the equation and force price to decline. Making the price equation:
P = P0 + P0(U - D1 - D2)(.1)
Limiting the opposite case is perhaps even more important. If a large quantity of ideas become 'locked' in portfolios then the value of U is held down, causing prices to drop endlessly. The easiest example of how this could happen is me. I hold over 90% of the ideas in a particular industry. If I quit playing those ideas would drive price down in that industry even if everyone else sold their ideas and they were consumed in the market. Taking a glance at reality we see the solution: a dead and decaying industry can be revitalized by a revolutionary new idea. When price drops below a minimum threshold, say .25B$, a new idea is created and all the existing ideas in that industry become obsolete and are removed from play. The low threshold ensures that the disappearance of all those ideas will not have a huge impact on anything or anybody. The next drop in the industry will effectively be in a new industry.
Which brings us to the question of pricing in a new industry. B$10. A new industry will generate high values of U and drive the price up. How fast and how often will depend on industry factors and will vary greatly, so they can all start out the same.
I'm no coder, but I think compared to the array of conditionals in the current pricing system the single equation for pricing and the small conditional for production/consumption should process relatively quickly, and since only industries that are in the drop need to be processed it should go very quickly.
I've run plenty of sample drops through a spreadsheet with the basic calculations built into it, and believe the system will work, and work well, but it is predictable that there will be some difficulties in implementation. These are all inherent in switching from a geometrically expanding unlimited system to a cyclic system. Basically, a lot of industries exist that are so far from their cyclic norm that they will have to undergo some radical shifts to get on track.
Catagory 1: An industry with a few blogs that has produced only a few ideas. This industry will have a very high price compared to new industries starting at B$10, and might always cycle at some higher than normal value.
Catagory 2: An industry with a few reasonably small blogs that has been reindexed daily or near daily for a long time. Price here is going to be very high, since it was a 'rare', but the huge number of ideas in existence will mean no upward force, and the high price and large number will generate a large D2...in some cases a really large D2. An industry with a total value greater than B$20b will generate a D2 greater than 1, as of this writing.
Catagory 3: An industry that has a lot of blogs, a lot of big blogs, and a huge pile of ideas. In most cases these prices have been pounded through the floor by the existing system, and since the number of ideas is so gross the new system is going to continue to drop price until enough ides get consumed to change that. In many cases it is likely that price will reach the revolution threshold before that happens. The revolution was intended to be a rare event, but during the transition it might be an everyday event.
Ongoingly there will be a need for adjustments. In the coding there should be constants; a multiplier on each of the terms, the '1000' in the U term, and the '.1' in the price change equation. The multipliers on the three terms are initially set at 1, but they can be used to shift the balance up or down by varying them between .5 and 2. Specifically, lowering the D2 constant initially will reduce the transition impact on the severely bloated false rares. Changing the 1000 in the U term will impact the upward force in all industries, and will also affect the frequency of cycles. Lowering the value will reduce the upforce immediately and tend to shorten the cycle. The .1 in the equation is a 'volatility adjust'. Initially, due to the transition, there will be some powerful forces generated in most industries. Over time we might find that the cyclic nature 'damps out', in which case this would be adjusted up.
That's my input.
Passer
posted by Tim at 7:15 AM
4 Comments:
I don't understand it all (in detail), but in general I totally agree with tying the two markets together.
Thanks for putting so much thought and testing into this!
Sage
Thanks for the comments. You are right, there are a LOT of bloated industries which would initially decline...many of them very rapidly.
Would there be huge sell offs? No doubt. That's part of the plan. Putting those huge bloats on the market will rapidly dispose of the ideas through the consumption equation. Yes, the price will fall, but by rapidly consuming the ideas the industry will be returned to health.
Looking back I see that somehow Blogger ate part of the idea consumption equation when I pasted this together, so I fixed that. Basically, assuming a massive sell-off in the bloated industries, they will hit the qualifir for 5% consumption every time the industry appears in the drop. With that rapid decline in the number of ideas they will be generating some upforce very quickly.
I see that I should have been looking at this more often.
First off...yeah Street I sent the sheet to KWP and I'll send it to you as well if you want.
Now to the questions from Tom...
The balance point for total worth of ideas is determined by the values chosen here: "In the coding there should be constants; a multiplier on each of the terms"... if these values are left as ones the market will trend to half the total value being in ideas, if the multiplier on the upforce term is raised to 2 then it will trend towards two thirds of the value being in ideas. If the multiplier on the downforce term is raised to 2 it will trend towards one third being in ideas.
As to the length of the cycle, that is going to vary widely, since it is basically a function of how often ideas produce in the industry...and that is a function of the number of blogs and how they get reindexed. Industries like 'internet', 'english', and 'male' are going to be incredibly volatile...ideas in industries which no longer have producing blogs will be just like cash.
Will the market crash if this is implemented? Yeah, to a great extent it will. I'm not 'in favor' of that or opposed, it is just a consequence that is almost unavoidable. Anything we put into effect will come on top of however many years of unchecked geometric growth that was in a sense constantly being 'gamed'. The pricing algo based on number of blogs theoretically rewarded scarcity, but reindexing skewed the system. Trying not to produce a playable balance without tipping the old applecart is a tunnel collapsed from both sides.
Post a Comment
<< Home