The BizOp News

Robert L. Park's book on Voodoo science has a great deal of wisdom to offer to the prospective franchisee or distributor and their investigations about the income generating opportunity.

In Science, the "no free lunch rule" from the social world is recast as "conservation of energy", or "you cannot get something for nothing".

Instead of the income generating opportunities that promise an endless source of residual income for no work, in the scientific world we have an parade of perpetual motion machines.

Dr. Park writes fluidly about some of these characters, Joe Newman, Cold Fusion, and James Patterson. He also maintains a website here.

Dr. Park has an excellent insight about what the attraction is to these hare-brained schemes.

I want to flesh this out in some detail because it demonstrates an important flaw in our reasoning methods. Dr. Park discusses the case of Randell Mill's BlackLight Power, one of the Cold Fusion castoffs. Despite having a theory in which there was "a state below ground state", Mill's company got funding from two utility companies for a total of $10 million.

According to Dr. Park, in a conversation with the business editor of the Princeton Packet, "The woman listened ... and asked 'But isn't possible that the laws of physics are wrong in this case, and Randell Mills is right?' ... A better way to phrase the question is 'What are the odds that Randell Mills is right?' To a very high degree of accuracy, the odds are zero. It's Pascal's wager, again." (my emphasis)

What is Pascal's wager, and why is it relevant? It a nutshell, it is the type of argument that runs like this: well, I know that A is highly unlikely to be true, but if I only invest a small amount $Z, the pay-off might be huge.

What is wrong with this form of argument, and if it is wrong or a fallacy, why is it an attractive fallacy?

People are very poor at strategic reasoning, contemplating how to act in a universe that is planning to react to their choice. Consider the following two bets or games, where the numbers represent dollars returned.

G1	S1	S2
A	100	-1
B	-1	0

G2	S1	S2
A	2	-1
B	-1	0

How much would you pay to play G1 and how much you pay to play G2, given that you know nothing about the probabilities of S1 and S2? You might reason this way.

Well in G1, if I played A and S1 and S2 were equal, then I would get on average at little less than $50, while in G2 I would get about 50 cents. So probably G1 is worth at least 90 times more than G2. Whereas, I might pay very little to play G2, I might spend much more to play G2.

As attractive as this line of reasoning is, it is false if we assume that whatever is choosing S1 or S2 is watching you make your choice, ie a market.

The zero sum value of G2, is -25 cents. But the value of G1 is -1 cent! G1 is not even a profitable game to play, despite that lovely looming left hand corner in which you get a return of $100.

Why isn't G1 valuable?

In a zero sum game, where people are trying to outwit you, S2 is going to be the state of nature in proportion to S1 in the exact ratio of 101:1. S2 is going to played to limit your gains to zero, and S1 will be played rarely -but positively just in case you always played B.

This is what trips people up: in a zero sum game, once you have the pay-off matrix, you have an excellent estimate of the states of nature. But in a decision theoretic version of the problem, the probabilities for the state of nature are not dependent on the pay-off matrix.

Reasoning as if they are produces errors in judgment.

Comments

Park's assertions about cold fusion are complete nonsense. Cold fusion was was replicated by hundreds of world-class laboratories, and these replications were published in mainstream, peer-reviewed journals. You will find a bibliography of over 3,000 papers and the full text from over 500 papers here:

http://lenr-canr.org

I suggest you review the scientific literature before commenting on this subject.

- Jed Rothwell
Librarian, LENR-CANR.org

Posted by: Jed Rothwell | November 13, 2008 9:56 AM