Randi on dowsing:
‘As I’ve said before many times, I have found that dowsers are generally very honest folks, and their firm convictions about the reality of their dowsing powers are examples of genuine self-delusion. On only two occasions, with the literally hundreds of dowsing claims I’ve examined over the last 56 years all around the world, have I found dowsers attempting to cheat — and both were caught out immediately, easily, and definitively.’
peter morris said:
Let’s consider “the patience of the dowsers” argument. You have 8 water dowsers. Each one must take each iteration independently, away from all the others, so you must sequentially put eight people through a minimum of five passes apiece, for a total of 40 passes. There were actually 50 passes, so two opted for 10 passes. At 50 passes (i.e. different iterations of selecting the pipe with water), 30 minutes per pass, you have 1500 minutes, or 25 hours. Clearly that will not be a pleasant day. Say you cut to 10 minutes per pass. That’s still 8.33 hours. That’s over 8 hours that everyone has to be on site, hanging around. Some are waiting to take their tests, some have completed and are waiting for results. Some are busy monitoring the whole thing. Is 10 minutes a fair limitation for each person to spend walking the grid pattern and determining which pipe has the water? Then there’s time for shuffling people between the grid and the waiting area, resetting the grid, etc. I think it is a fair to conclude the rationale for limiting the tests to 10 interations per person maximum is the time delay on the overall test.
Note that this particular test was looking for a strong success rate (80%), so 5 tries per person provides enough variation to pick between 80% vs 10%. (How do you get 10% on 5 tries? If you got one right, that’s automatically 20%. Chance would say that one day you might get 1, and one day you might get 0, averaging out to 10%. But if everyone hit their 1 on the same day, the total statistics will automatically say 20%.)
It’s been repeatedly pointed out to you and you still fail to grasp that the test was not designed to accurately gauge a 20% vs 10% effectiveness. It was set for 80% vs 10%. If the dowsers had said they could get 20% regularly instead of 86% regularly, then the test would have had to span a lot longer time period to allow the necessary iterations to allow that fine a determination.
Mr. Miskatonic said:
peter morris replied:
Wrong. Some of them supported you that grouping the brass and gold statistics with the water statistics was invalid. Some of them did not agree. The numbers seemed evenly balanced to me, not “mostly” agreeing with flaws in Randi’s conclusions.
This is such an ironic statement coming from you, who still doesn’t understand this sentence.
-Randi at the Australian dowsing test above.
peter morris said:
Why should he have given them a second run? The statement of the rules were that they all get either 5 or 10 tries at their discretion, chosen before testing starts. Their results for those 5 or 10 tries determine whether they meet the required 80% level. IFF one reaches the required success level, it can be requested he reperform the test (5 or 10 iterations as before) “in order to establish a more satisfactory statistical sample or to ensure security of the protocol”. The results of their two sets of runs would then be averaged for their total results, and judged vs the 80% pass/fail criteria. NONE of them got near 80%. None qualified for continued testing, for the prize. (Note that request is limited to once. This is to limit Randi so there can’t be arguments that he kept asking them to repeat until they failed.)
Now if your intent of that comment was to see if the 22% was fluke or above chance for statistical relevance, then that’s another matter. As has been pointed out numerous times, the test was designed around the stated abilities of about 86% success rate, not 22% success.
He summarized the total results of the day with one average. He then also lists the results for each dowsing target. I don’t see any coverup if he actually lists the results you say he’s trying to cover up.
owlofcreamcheese said:
Maybe it is reasonable, maybe not. But it doesn’t matter, because Randi asks the dowsers if they can find a few gallons flowing in a pipe or not. They say yes. They even get to practice on the actual test setup where they know where the water is flowing, and get the successful results they claim. Only when the location of the water is unknown to them beforehand do they fail.
If the dowser were to state, “No, I can’t find a few gallons in a pipe, only a natural source or only thousands of gallons of underground water,” then you would have a point.
But consider the difficulty of trying to design a test for that situation. You must find (1) an underground river (2) that a limited and controlled few know the location of (3) that can be subsequently verified to everyone after the fact. If it is widespreadly known ahead of time (like some sort of geography database), then the dowser could identify the location and learn it beforehand. But if it is not widely confirmable afterwards, then the test is not conclusive or biased against the dowser (what do you mean it’s not here but there? Prove it!) Alternately, you pick an unsurveyed area and have the dowser go after it, then drill both his picked locations (to verify water is there) and several unpicked locations (to verify water is not there). This gets exceedingly expensive and difficult to reach conclusive results that the water is only where the dowser claims.
That is why tests use simplifications. If it is at all possible, it is better to reduce the complexity of the testing verification. Plus it is good to design protocols that allow statistical evaluation so you can compare vs. chance results.
Consider this situation:
Dowser: “I can do X.”
Skeptic: “X is difficult if not impossible to design a test for. Y is similar, but much simpler. Can you do Y?”
Dowser: “Of course I can do Y.”
Skeptic: “Good, then let’s set up a test for Y.”
Now is that a fair representation that the dowser did or did not claim to be able to do Y? Is that a fair way for the skeptic to find something that the dowser can do but can be tested and evaluated effectively?
In my example, what if dowser responds
D: “Um, I suppose I can do Y. I don’t see why not.”
That makes it a lot less clear. But again, in that case, you take them and give them a situation Y and let them practice. They then confirm they can do Y, repeatedly, under the proposed conditions, except the controls removed. Is it then appropriate to proceed with the actual test, implementing the controls? Why would it be unfair, if they just proved they could do it?
Okay, just for grins, try another response.
D: “I don’t think so. Y isn’t really like X at all.”
S: “Okay, let’s refine your limitations so we can find a Z. What about Y is different than X that prevents it from working? What other things can you do that are similar to X, but easier to control and setup?”
See, the skeptic proceeds into a dialogue to determine an appropriate test. Is that not a fair way to come up with a test protocol that works for everyone?
Please provide a cite for that. Give us even one case where Randi tests a claimant to a stronger claim than they actually make.
bnorton said:
What it does is shows this particular group of students how gullible they are. It shows how a generic set of situations can be fit by the receiver into something that seems very personal and specific - but the personal and specific aspects are all added subjectively. It explains how horoscopes can seem very personal and accurate but be general and inaccurate. It shows the problem with subjective evaluation.
This is another case of explaining how it appears to work, not proving that it doesn’t work.
I don’t think anyone is arguing that the point of the psychic agreeing to the test is that they agree to the scientific accuracy. The point of the psychic agreeing to the test is that the test accurately reflects something they claim to be able to do.
Shade said:
I think you’d be surprised. I’m sure he’d be willing to devise a test and administer it. The question becomes the determination of the claimant. Realize that all expenses for testing are paid by the claimant, not by JREF. So Randi can devise a test that runs a full 8 hrs a day for 5 full days to collect a large enough collection of data to mine small statistical results. But the cost of that test would be up to the claimant. Realize it probably wouldn’t be that much more, if people participating are volunteering their time (I assume this is the case, but don’t know). I mean once you’ve dug up the field and buried the pipes and hooked up the water flow, how much more expensive is it to run five days versus one? I don’t think much.
Libertarian said:
Funny how you fail to notice his appeal to authority or his inaccurate representation of the content of the TV show. Both were demonstrated in this thread. If you’re going to blast his critics for logical fallacies and bad arguments, you ought to point out his, too.
glee said:
and patchbunny and TVAA both make comments.
Folks, I think you misunderstood Libertarian. That comment was, I think, specifically directed at this remark:
Summertime said"
glee said:
Are you saying you move the pieces yourself and occasionally move one too far, or something different? Does that mean on a standard size board, and if I bring my travel chess kit you’ll be totally off? Or are you saying that only about 1% of the time will you misremember a piece position, and try to move through them or such?
IzzyR
Well, yes, there is a minimum level. The claim has to be distinguishable from chance, because that’s the whole point. It has to be a testable claim. If you claim to have a paranormal ability that performs exactly at chance, then how is that testable? And furthermore, how is that a useful psychic ability? So there is a threshhold value.
However, the claimant gives the criteria for their performance, and the test is designed around those criteria. The claimant has to plainly state what it is they can do, and the limits, so that an appropriate test protocol can be devised. Assuming the performance level is distinguishable from chance, then a test can be devised to whatever performance level the psychic claims. 90 to 100%? Easy. 40 to 50%? Okay, a little more work. 2% over chance? That’s going to be very difficult, time consuming, and probably expensive.
But if you doubt that the psychics have any say in their expected rate of success, I direct you to the same Australian test we’ve been discussing:
http://www.skeptics.com.au/journal/divining.htm
Or check out the JREF challenge rules page.
Look, I’ll ask you what I asked peter morris and he has so far failed to provide and ran away from my thread. Provide me a case of someone who backed out of the JREF Challenge with accusations that Randi was being unfair, with specifics on how Randi was unfair. I don’t just want a generic statement, I want something documentable. Maybe an email of an unfair request or letter with statement of requiring a stronger performance than the claimant makes. Because you can bet Randi keeps records of all his correspondence precisely for this reason. Put it in writing, and then you have later proof of what occurred. So find someone with documents showing unfairness on Randi’s part. Show me evidence, not just claims. Anybody can claim anything - I can claim you’re made of green cheese. I want evidence.
Well, one would assume that the students would have been selected at random to represent the population as a whole. Otherwise it’s not science at all - it’s more of a morality lesson.
Keep in mine that the OP had to do with questioning the scientific legitimacy of Randi’s tests - at least that’s how I saw it.
(Thanks, Davesink!)
If Randi is making no claim as to whether of not “dowsing really happens” (ie, is a paranormal phenomenon)–but only testing whether given self-described “dowsers” can actually do what they agree-in-writing that they can do–then his methods are not noxious (though his personality may be).
Is it the case that he has never claimed, on the basis of encounters of this character, that the so-called power of dowsing has been scientifically disproven (as opposed to, that this individual’s claim has been disproven)?
My guess is that he is innocent of having made such a sweeping assertion. The problem with the hypeful rhetoric and inconcealable disdain with which some “professional skeptics” approach claimed paranormality, is that it distracts the observer, and the potential critic, from the logical substance of the “test”–which substance depends in a crucial way on precisely what is being asserted on BOTH sides of the clipboard.
RickJay has already made the following points but they bear repeating, simply and clearly:
Randi has never said that dowsing is absolutely disproven. Dowsers like to set this up as a strawman, but that’s all it is.
Randi is loudly sceptical about dowsing etc because he never says any evidence of it. What is wrong with this attitude?
Many people allege that Randi’s tests are unfair, that he sets the tests to be impossible etc. When you ask for details, there are none. People can (in this thread and elsewhere) speculate for hours on whether Randi is or might be in a position to demand unfair procedures. It is all so much smoke and hot air. Fact is, there is no evidence whatever that Randi ever does
The false choice is set up by you.
Randi is not the only game in town. If a “real” dowser approached Randi and found that Randi would not agree to reasonable testing rules, that dowser has a golden opportunity in several ways.
Firstly, he can document his experience in detail and disseminate what a fraud Randi is. Given the number of psychics, dowsers etc out there to whom Randi is a thorn in the side, this information would be howled to the rooftops.
We have asked, nay, demanded cites of such information. None have been provided. What this implies is obvious.
Secondly, the “real” dowser can set up his own fair scientific double blind test. Have a few respectable observers, maybe film it, take the test, prove himself. That film would be valuable, and the dowser a hero.
Suggesting Randi has a monopoly on double blind testing is obtuse in the extreme.
It’s 1880. Jon Edwar in his book “Crossing Over”, in which he describes his experiences as the sailor with the most Atlantic crossings, has stated that no one will ever be able to build a flying machine, and certainly not one capable of crossing the Atlantic. He puts up a prize for the first person to fly a powered heavier than air machine 10 miles.
James “Randy” Wright, a well known ladies man and publicity hungry inventor is pretty sure that an Atlantic crossing is possible. He learnt to build light weight canvas and wood structures during his former career at sea, and he is well qualified to know that Edwar is wrong. To build his transatlantic machine would take more resources than he has. So James decides that he will prove his point, and make publicity by taking Edwars prize: thereby demonstrating how foolish is Edwar’s denial of what is possible.
James builds his machine then announces to the world that there is going to be a demonstration. Demonstrating flight with his fragile and temperamental plane is not easy. He knows that if there are any technical hitches, or bad weather prevents flight, or if luck is against him in any way the public will be all too ready to believe that he are wrong and flying is impossible.
So the week of the demonstrations arrives. Knowing that many people won’t give credit for a tasty cake with no icing, he surrounds his flights with dancing girls and brass bands. On the first day he flies four miles. On the next day with his Aussie cousin Dick “Smithy” Wright flying, he has mixed fortunes. One flight goes nine miles, in another the plane goes only a hundred yards before crashing but luckily Dick and the plane are salvageable. On the third day, and on several thereafter, the plane goes ten and even fourteen miles.
James stands before the crowd and points to his success. He mentions that the crash happened, but makes no further comment. He speculates about how the plane might work, and what his success tends to suggest about the possibility of transatlantic flight.
Many are impressed. But there is a small bunch of detractors.
Peter Morris stands on a podium and yells that the James is a fraud and liar. bnorton and a few others chip in some comments. Peter points out that:
[ul][li]the plane crashed once, and James does not mention this or acknowledge that this means that planes cannot fly.[/li]
[li]James says a plane may be able to cross the Atlantic but James has never done it so therefore he shouldn’t say that a plane can fly[/li]
[li] Peter says James is a fraud and a liar[/li]
[li]Edwar announced a prize for flying ten miles. James should not get that prize because Edwar speculated that no plane would fly across the Atlantic, and James has not done that.[/li]
[li]James’ speculation about how the plane might work is wrong, so therefore the plane did not fly[/li]
[li] Others have carried bird-like contraptions on trucks for ten miles, so they should get the prize, not James. James should stop playing word games and just accept that they were the first to scientifically demonstrate ten miles of powered flight, not him.[/li]
[li] Peter says James is a fraud and a liar[/li]
[li] James constantly self-promotes, grandstands and boasts. His demonstration was about making money for James Randi, not about any scientific endeavour, therefore his plane didn’t fly.[/li]
[li] The test was conducted from a field outside the town of Skeptic, along the road to the village of Putz. A white line was marked on the road at ten miles. James had insisted under the competition rules that he only had to fly one inch over the line and he could claim the prize. Peter points out that with surveying techniques being what they are, the line could well have been as much as 10 yards short of ten miles, and that the degree of accuracy insisted on by Randi was unreasonable. The fact that he flew the whole way to Putz, a distance of fourteen miles, on two occasions is irrelevant[/li]
[li] The rules provided that the flights must not be tail wind assisted. All agree that the best way of ensuring this requirement is to have a flight in both directions (known as double wind testing). On one day, James flew 11 miles but then broke down and could not fly back. His first flight was into a strong headwind so James suggests that it should count for something. Peter says this suggestion is a fraud, because there was no double wind testing.[/li]
[li] No scientists were present when the plane flew over the spectators lining the route. No scientists designed the test. Who is to say that the plane was actually flying?[/li]
[li] Peter says that James presented no information about the source of the parts of the plane, what sort of wheels it had, or the type of fuel the engine ran on. It is later pointed out that there was a brochure with all this information available, but Peter nonetheless says that lacking this important information, it is fraudulent to say that the plane flew.[/li]
[li] Peter says James is a fraud and a liar.[/li]
[li] James and his supporters are clearly zealots and fundamentalists and James himself is a polemicist. Therefore his plane did not fly[/li]
[li] Some people bet on James being unable to fly ten miles. On the day that he flew nine miles, many of these people withdrew their bets. Some people kept their bets on, but after the successful flight refused to pay up. They said that James’ plane was held up by fishing line strung from skyhooks, or that he had been assisted by the alignment of the planets. Clearly, Peter says, this shows that James must be a fraud, otherwise why would these people be complaining?[/li]
[li] James clearly believed that he could fly before he did his demonstration. Therefore, he was biased and the flight of his plane should be ignored.[/li]
[li] James jumped to conclusions, and only tested intensively over a week, not extensively over years. Who’s to say that he could fly next week? His demonstration of flight is invalid[/ul][/li]
While Peter opines and argues and points out flaws and imperfections, he totally misses the impressive sight of James, in his plane, buzzing around overhead.
That’s some nice posting there, Princhester.
Is this true? Are your opponents sitting there saying “knight to rook 8” and suchlike, and you keep the board position in your head? Wow.
Sorry, I should have explained blindfold chess.
The expert has no access to a set or board. (Traditionally you simply sit with your back to the opponent, who has a normal set + board.)
There are a couple of notations for describing chess moves (old = p-k4; newer = e4).
The opponent studies his board, moves his piece and announces his move in chess notation.
The expert visualises the game in his head and replies by also announcing his move in chess notation.
And the game proceeds till checkmate or a draw.
An illegal move by the expert would be, for example, if a piece is not on the square he thinks it is, or his move is ‘blocked’ in some way.
Note that the expert has to rely throughout on his mental picture of the board. He has no physical contact with any chess pieces or board.
That is pretty difficult (try it yourself!).
Now consider that the expert has to play well. This involves analysing the position i.e. ‘moving’ the pieces himself in his head, then ‘restoring’ the original position, then analysing another variation etc.
Now add on that the expert is playing two games at once - the simultaneous bit - (traditionally one with White, and one with Black). He has to hold two separate chess positions in his mind, and analyse each in turn.
That is really difficult.
My claim is that I am around 99% accurate in visualising the two positions and thus announcing legal moves. I can also beat club players, despite the obvious handicap.
Personally I don’t believe in dowsing (since there is zero evidence for it). But I also doubt, assuming it exists, that it is anywhere near as tiring as simultaneously blindfold chess. Since I can provably manage around a 99% success rate, why can’t dowsers match that?
I haven’t offered any excuses. I’ve offered a couple of possible reasons. (I didn’t use the word “maybe” for nothing.) The fact that you are using a euphemism means that your argument is merely rhetorical. The fact that you can’t think of any reasons that a test might fail means that your conclusion is predetermined.
Funny is as funny does, I reckon. Funny how you don’t know that appeal to authority is not necessarily a fallacy. Funny also how you presumed that Randi’s own book was automatically an accurate representation of his show. Maybe he has a bit of a Bob Eubanks memory lapse. Or maybe he is dishonest. Or maybe he is just mistaken. The blind faith some of you put in this man does not speak well of skepticism as a philosophy.
Is this fair comment? I certainly think that tests might fail and I don’t think anyone is suggesting, in any absolute sense, otherwise. But if the tests mostly seem fair, and if the testees regularly come out with pathetic excuses involving alignment of planets or the clouds being shaped funny (and didn’t suggest that to be a problem before the test) there is a conclusion to be reasonably drawn.
But Lib it’s hard to credit a person who appeals to authority, then picks out the fact that the authority supports one of that person’s minor premises, but ignores the fact that the authority contradicts that person’s major premise.
Blind faith or a not unreasonable weighing up of evidence? If I have to weigh up recollections of a TV show between:
[ul]- the ten year old recollections of an anonymous poster who saw the show once, who’s vehemence against Randi is, well, kinda intense and who’s factual reliability has not always been, errr, this ain’t the pit but you are invited to read my pit thread, Lib; and
I posted an account of a dowser’s test for the money:
http://www.randi.org/jr/032902.html
I summarised it ‘So that eliminates any unfairness in the tests (the dowser chose the materials and their placements), any problems on the day (the dowser scored 100% in the open tests - equivalent to Godhood?!) and showed an amazing difference when the dowser (and Randi) didn’t know where the material was.
And the scientific conclusion is?’
I assume Randi had a tape of the show (I got one when I appeared on TV.) Note that Peter Morris relied on his 10 year old memory to make serious allegations. (Of course you’re familiar with the standard procedure on this board to provide cites to back claims?)
Randi has raised a million dollars to encourage paranormal powers to be demonstrated. By contrast both the British and American Dowsing Societies have a policy not to be scientifically tested. Their dowsers, however, do charge money for their ‘services’. This raises serious doubts in my mind. Does it in yours?
here a lot to comment on, so I’ll limit it to a few choice remarks.
To Miller:
If that were true it would bother me, but scientists keep on agreeing with my conclusions. Doesn’t it bother YOU that nobody with any knowledge agrees with you?
Wrong. If the tester is hoping that the subject will fail his body language can inflence the subject to make the WRONG choice. It can work either way.
That is rubbish. The only possible interpretation of his words is that he was denying the EXISTENCE of underground rivers. Even if we accept your rather desperate reinterpretation of his words, you have not yet shown ONE dowser who makes the claim that “underground rivers are a common phenomenon in all types of terrain” so Randi is STILL wrong on that count.
To Princhester:
Read again. Clarke agrees with my main point, which is that Randi’s conclusions are way out. The remark I made that there is a difference between underground rivers and pipes was NEVER my main point, only a passing thought. YOU have made a big deal of it. Clarke backs my main points completely.
To Musicat :
He may not be an authority on dowsing, but he is an astronomer, which requires more than a basic understanding of mathematics. When he comments on Randi’s maths, adding the results of three seperate tests, and finds that Randi’s conclusions are wrong, that opinion has some weight. And if you consider Clarke’s open-mindedness unscientific, then you simply don’t understand science.
You are totallly wrong on that. The water dowsers did not agree to having the results combined with brass and gold. They were totally seperate tests taken by different people.
To Patchbunny
So I remembered four instead of three. Mea culpa.
In the book, maybe, I’ll take your word for that. I’m sure it wasn’t shown on the televised program.
In fact I never claimed that, I said she didn’t pick a saw.
So the map was smaller than I recalled. Mea culpa again.
Which is exactly what I said. Try reading.
And then you post a passage that CONFIRMS my statement.
To Princhester:
Actually, all you lot did was spew abuse and invective, calling me disgusting names like ‘fuckwit’ and ‘shitforbrains’ and ‘princhester’. You repeatedly presented me as a believer in dowsing, while I repeatedly said I wasn’t. You lot came up with statements such as ‘Randi says underground rivers are fictional’ then hastily backtracked when I proved them wrong. I am still waiting to see refutation.
To Patchbunny:
Tell me, Patch, if the dowsers had claimed a 20% sucess rate, do you think that Randi would have let them take the test? HE set the limit of 10 tries, and scoring 2 out of 10 is not unlikely. If someone sent in their application form, saying they would get 2 hits, would Randi have let them take the test, or would he have rejected the application? What do you think?
If they dont sign the form, they don’t get to take part. This does not prove they really think the test is fair.
To **Irishman **
Exactly, this is why Randi won’t take the long tests. But a couple of observations: a) I dont see any reason why several people should not test simultaneously, b) why not stagger the test over several days, no need to have them all waiting throughout the test.
I understand it perfectly. The Randi lovers have desperately tried to attach some alternate meaning to a simple direct statement. Even their alternative interpretation is factually wrong.
to Princhester:
Not a bad analogy, Princhester, except for a couple of things. First of all, lets say that he never makes his 10 mile flight, he keeps landing after 2 or 3 miles, thus fails to claim the prize. Then it is a good analogy.
Other than that, you have got the roles the wrong way round. I would be the one saying that setting a 10 mile minimum is unfair, that the flights of 2 or 3 miles show some flying ability even though its not enough to win the prize, and that it merits further study, to see if Randy can make the 3 miles EVERY time. I would be saying that if he can do a 2 mile flight several times, that should qualify him for the prize.
YOU would be the one calling Randy a deluded fool and a fraud, calling the 3 mile flight an abject failure, and declaring that if flight is possible, why is nobody claiming the prize. You would be screaming about why would Randy agree to a 10 mile minimum if he knew he could only do 2 miles. You would be saying that the two mile flight is just a fluke, that Randy just got caught on the wind, and it’s not really flying at all. You would be saying there’s no need for further tests, he tried and failed.
In fact, take those things you attribute to me, add some name-calling and a few straw men, and it’s EXACTLY what you have been saying all along. Good analogy, which clearly demonstrates the flaws in your own arguments. Thank you.
So they sign a statement saying they agree that the rules of a test are fair even when they don’t believe that’s true. Isn’t there a word for people like that? Oh yes - liars.
Only if random chance would indicate that a normal person going at the test without any flying ability could fly 1 mile, on average. But currently, humans have about a zero mile flight range, so a 2 mile flight is a MUCH larger margin of success than guessing right 20% of the time when expected to guess 10%.
To Patchbunny
I read your statement as the saw was one of the items she picked, that Randi said was incorrect. Which does bring up another question. If you claim that Randi didn’t reveal anything about the objects, how do you support your claim that you know the background of one of the items?
I did read. The tone of your statement was that there were the option of 5 or 10 trials, then there were only ten. A better phrasing on your part would have been “Setting the maximum of 10 tries was entirely for their benefit…”
In a test that was designed to test for a much higher success level? Lesse… If I get 2 out of 10, or 1out of 5, you pay me $10,000. I don’t believe that Randi is that stupid. He would suggest a different test so that their claimed abilities could be separated away from random chance.
There were eight dowsers trying for water, with 50 total trials (with 11 successes). That means two dowsers made ten trials, and six made five trials. I would expect the total passing percentage for all dowsers to be close to 20% by chance alone. So why are you arguing that a 20% result means something?
Again, from the statement they signed, they agreed that:
“I agree that the rules as outlined in the accompanying Document Number Two (Rules for Test) are fair and proper.”
If someone signs off on this form, then claims an 80% or better success rate when their actual success rate is 20%, then they’ve misrepresented themselves (i.e., lied).
–Patch
I came here to read the stuff about Randi but the Chess thing was interesting too, so sorry that this is a bit OT, but there is a way that anyone can play two chess games blind, one as white and one as black, and (with a couple of conditions placed on the order of moves) definitely win one of them 
But I assume Glee can do just as well, if not better, playing a single game, for which there’s no clever shortcut system.
David
Couldn’t seem to edit my post, but anyway, here’s something I remembered. Some time ago a guy got his application through and went to be tested (preliminarily) by Randi. Randi’s write-up of that test is here:
http://www.randi.org/jr/032902.html
And here’s the thread on the JREF forum with the dowser’s excuses (which include, but are not limited to, “it wasn’t fair”):
David
A fact that has been mentioned, but not given sufficient emphasis, is that Randi first applies a test in which all the psychics pass. For instance, with the dowsing, they show them the pipes, tell them which ones have water flowing through them, and then ask the dowsers to check and see that yes, they can detect that water. Randi not only allows, but demands that the dowser calibrate his instrument. Then Randi covers up the pipes and asks the dowser to repeat the same measurement he just did. And suddenly they can’t detect the water anymore… despite the fact they successfully detected it just moments ago.
So all of Peter’s complaints are utterly baseless. Randi not only restricts his test to what the psychic claims, he restricts his test to what the psychic can actually perform - without any controls. Then Randi applies a simple contol, like for instance not telling them which pipe has water, and they fail, fail, fail. This demonstrates, to any rational person, that the control applied was the crucial variable.