5 out of 5 scientists are statistical idiots

Status
Not open for further replies.
M

mrmorris

Guest
In the SDC article, it's stated that 5 out of 5 scientists agree that our solar system is 'special' <b>because</b> all of the planets found to date indicate solar systems incredibly different than ours (namely a preponderance of huge planets very close to their sun or huge planets with highly elliptical orbits).<br /><br />It's hard for me to believe that presumably trained minds could even consider making such an assumption. The two methods used to locate extra-solar planets (The Spitzer method wasn't used to 'find' a planet, so it doesn't count... yet) <b>would not find</b> a solar system like ours. Even if I were to include Spitzer -- it wouldn't work either without a huge planet close to the Sun.<br /><br />- So we've found 150ish planets in solar systems completely unlike our own. <br />- The methods being used wouldn't *find* a solar system like ours. <br />- The 'unusual' solar systems almost assuredly make up less than 1% of the total number in the sphere of space around our solar system that would contain them.<br /><br />The upshot is that they can't use the specs of currently located planets to make *any* statistical statements because their data is completely skewed towards the conclusion they're proposing. This would be like using only residents in Beverly Hills to make a study about the average annual income for residents of the entitre planet, and concluding from the results that the average household income worldwide was $10,000,000.
 
Y

yevaud

Guest
That's a really sad statistic.<br /><br />I always took the point of view that since we have exactly *zero* data points for any statistical analysis, it's a moot question anyways.<br /><br />And to derive the probability of Earthlike planets existing using the data that they *do* have, is rather like counting a pile of Grapes...and then determining therefore how many Apples exist! <div class="Discussion_UserSignature"> <p><em>Differential Diagnosis:  </em>"<strong><em>I am both amused and annoyed that you think I should be less stubborn than you are</em></strong>."<br /> </p> </div>
 
A

aaron38

Guest
Aren't the researchers also a bit biased? Of course our solar system is "special", we're in it!<br /><br />The town I was born in is "special" cause it has a park by the river and a Happy Joe's Pizza and my uncle's house. But in reality it's just a crumbling factory town like a thousand others, nothing really special about it.<br />
 
Y

yevaud

Guest
I don't know. <br /><br />I will say that any Time/Space continuum that has a "Happy Joe's Pizza" in it is a very good continuum indeed. <div class="Discussion_UserSignature"> <p><em>Differential Diagnosis:  </em>"<strong><em>I am both amused and annoyed that you think I should be less stubborn than you are</em></strong>."<br /> </p> </div>
 
T

tap_sa

Guest
Creationist News headlines:<br /><br />DARWINISM DOOMED! ALL SCIENTISTS AGREE THE SOLAR SYSTEM IS INTELLIGENTLY DESIGNED!
 
N

nexium

Guest
Since none of the extra solar planet searchs are capable of detecting an Earth like planet; we should not be surprised that no Earth like planets have been found to date in distant solar systems.<br /> Important parts of Darwinism can be approximately true even if thereis intelegent design/God. Neil
 
E

earthseed

Guest
These scientists are well aware that current planet detection methods can only find large planets. Their point is not planet size, but rather the shape of their orbits. Circular orbits may not be essential for life, but they sure make it a lot easier.<blockquote><em>With the vast majority of the alien planets found in eccentric orbits, Butler has a different view. I think with the data at hand, we can say that our solar system is rare. Eccentricity dominates, said Butler. It's just a matter of how rare we are, he added.</em></blockquote>There remains the possibility that there is an entirely different planet formation process that results in small rocky planets with circular orbits. But this is pure speculation, the actually observed data suggests otherwise.<br /><br />I am sorry folks, but you have to score one for the Rare Earth hypothesis this time. Maybe the next discovery will go the other way. It seems to go back and forth.
 
S

Saiph

Guest
even if they meant "eccentric" orbits are the rule, we still don't have enough data that isn't prone to such a huge selection bias to make such a statement. <div class="Discussion_UserSignature"> <p align="center"><font color="#c0c0c0"><br /></font></p><p align="center"><font color="#999999"><em><font size="1">--------</font></em></font><font color="#999999"><em><font size="1">--------</font></em></font><font color="#999999"><em><font size="1">----</font></em></font><font color="#666699">SaiphMOD@gmail.com </font><font color="#999999"><em><font size="1">-------------------</font></em></font></p><p><font color="#999999"><em><font size="1">"This is my Timey Wimey Detector.  Goes "bing" when there's stuff.  It also fries eggs at 30 paces, wether you want it to or not actually.  I've learned to stay away from hens: It's not pretty when they blow" -- </font></em></font><font size="1" color="#999999">The Tenth Doctor, "Blink"</font></p> </div>
 
Y

yevaud

Guest
Not sure how you reach your conclusion, as there is zero data to show evidence - one way or another - of Earthlike planets in the appropriate orbits.<br /><br />To conclude that we're "rare" is premature, and just plain guesswork. <div class="Discussion_UserSignature"> <p><em>Differential Diagnosis:  </em>"<strong><em>I am both amused and annoyed that you think I should be less stubborn than you are</em></strong>."<br /> </p> </div>
 
T

thalion

Guest
Of all the planets that have been found to date, only one (the outermost planet of 55 Cancri) has an orbital period greater than or equal to Jupiter's. If our Solar System is going to be the gold standard for others, we should wait until we have more data on planets that take a good 12-24+ years to orbit their star, and see if we can't find more systems with tidy orbits before we come to certain conclusions--all IMHO, of course.
 
M

mrmorris

Guest
<font color="yellow">"Their point is not planet size, but rather the shape of their orbits. "</font><br /><br />Again... NOOOO! The 'Wobble' detection method (which I believe has generated the vast majority of 'found' planets so far) works best for solar systems with planets in highly eccentric orbits. Again -- you can't make an intelligent prediction using results that are predisposed towards a given conclusion.<br /><br /><font color="yellow">"I am sorry folks, but you have to score one for the Rare Earth hypothesis this time."</font><br /><br />No -- I don't. <br /><br />Actually, while the article talks about the hypothesis before any extra-solar planets were found being that solar systems like ours might be the norm -- what *I* remember is that there was significant speculation that it might be very rare or even unique for *any* planets to form around a star. What I recall from the news when the first extra-solar planet was discovered was a set of article stating essentially 'Woo Hoo -- the solar system is not unique in having planets around it'. Of course, given the number of solar systems in the universe -- anyone who truly believed that was <b>also</b> a statistical moron.<br /><br />Considering that they have found the number they have in a few years of these methods being developed -- that notion is pretty well defunct. Carrying that further -- since theey have found a significant number of <b>the specific type</b> of systems that we are currently able to locate, it stands to reason that once methods are available which expand the range of planetary types we are able to detect, we'll locate a good number of them as well. Once we have improved methods that are capable of locating smaller planets in more regular orbits, I believe that they'll stack them up like cordwood. Until we *have* that technology and have used it for some time -- making predictions such as are contained in that article is moronic. It's equivalent to the famous quot
 
E

earthseed

Guest
I never suggested we are anywhere near able to decide the great question of Rare Earth vs. Life Everywhere. Rather, what is interesting is each new discovery seems to reverse the implications of the previous one. We thought planets might be the exception, but discovered they are common - score one for Life Everywhere. But the large planets are in awkward places, and the orbits are not nice - score one for Rare Earth, as the five scientists argue.<br /><br />Yes, the data is selective, biased toward finding large planets with fast orbits. What will the next set of observations find? Not what you, or I, or the five scientists expect. It seems nature always has a surprise, although I imagine it is Life Everywhere's turn to score a point.<br /><br />My point is these scientists are not idiots. Scientists work from the data that exists. That data is ellipical orbits, not good for life as we know it. Do other solar systems we can't see yet produce the right sized planets with circular orbits? That is pure speculation, except we have an example of one - ours. So it is possible, but clearly not universal. Fortunately progress in this field is rapid, so we may have a better idea relatively soon.<br /><br />I can't wait for the next surprise.
 
A

arobie

Guest
<font color="yellow">Scientists work from the data that exists.</font><br /><br />Yes, but they should be able to see that their data does not and cannot represent an average for the universe because of the methods used to gain this data. They should also know not to try to make a definite conclusion with incomplete data. That is their mistake.
 
T

toymaker

Guest
I posted this already in another discussion:<br />http://www.astro.columbia.edu/~dave/papers/earthplanetsstability.pdf <br />"We have shown that Earth-mass planets could survive in variously restricted regions of the habitable zones (HZs) of most of a sample of nine of the 93 main-sequence exoplanetary systems confirmed by May 2003. In a preliminary extrapolation of our results to the other systems, we estimate that roughly a third of the 93 systems might be able to have Earth-mass planets in stable, confined orbits somewhere in their HZs. Clearly, these systems should be high on the target list for exploration for terrestrial planets. We have reached this conclusion by launching putative Earth-mass planets in various orbits and following their fate with a mixed-variable symplectic integrator."
 
E

earthseed

Guest
Thanks, Toymaker, it is interesting that Earth-like planets could find stable orbits in the habitable zones of some of these stars. But I think our five maligned scientists are aware of that. Remember that it is hard to tell what they really think after the information has passed through the mind of a reporter. But they mention more problems than elliptical orbits, such as<blockquote><em>Thus far, 90% of all detected alien planets have host stars that can flare and sterilize the surface of the planet. Furthermore, planets, which are that close to their host star, would be in a synchronous orbit. This means that only one side of the planet would face the host star and all potential water on that side would evaporate and go to its “dark” side.</em></blockquote><br />The article makes it clear that none of these people are claiming certainty.<blockquote><em> “I think with the data at hand, we can say that our solar system is rare. Eccentricity dominates,” said Butler. “It’s just a matter of how rare we are,” he added.</em></blockquote>That is the way it seems today. New information in the future may change that view.
 
G

g_riff

Guest
Our methods of planetary detection may result in skewed data, but we can still garner some information from that. Since we cannot see planets in solar systems like our own, we should expect a lot of systems to appear to have no planets if our system is not rare. By the same token, if most of the systems we observe appear to have planets (which we can see due to their eccentric orbits/massive planet size), then we can say that our type of system appears to be rare. <br /><br />It doesn't matter how many solar systems are out there, it only matters what we've seen from the ones we've observed. I don't know the percentages on how many of the systems we've observed appear to have planets around them, but if we aren't detecting planets from a lot of these systems, then my entire argument is invalid, and those scientists really don't know what they're talking about. However, if this is not the case, then there is good reason to argue that our solar system is rare.
 
E

earthseed

Guest
Assuming there are plenty of suitable stars for which we cannot yet detect planets, the "rare solar system" hypothesis of the astronomers is still reasonable. The more unknown stars, the weaker the hypothesis. At this point we do not know if these stars have planets with elliptical orbits, or no planets at all. On the other hand, a new set of conditions may kick in and they have solar systems like ours.<br /><br />The point people seem to miss is the elliptical orbits are observed, while circular ones (except for here) are pure speculation at this point. Scientists like to give more weight to observation than speculation.<br /><br />But (how many times do I have to say it) the question is still open.
 
M

mrmorris

Guest
<font color="yellow">"...we should expect a lot of systems to appear to have no planets if our system is not rare. By the same token, if most of the systems we observe appear to have planets (which we can see due to their eccentric orbits/massive planet size), then we can say that our type of system appears to be rare. "</font><br /><br />There have been ~150 extra solar planets reported found to date -- some of which are in the same solar system (i.e. there have been less than 150 *stars* found with planets). I can't think of a reliable means of determining exatcly how many stars are within the spherical area that contains all of the 150 planets, but I would be suprised if it did not contain between 100 and 1000 times that number of stars (i.e. 15,000-150,000).<br /><br />I am positive (albeit without definitive evidence) that they have not detected planets on a sufficiently large percentage of the available stars to make a reasonable statement about the remainder.
 
M

mrmorris

Guest
<font color="yellow">"...more problems than elliptical orbits...<br /><br />90% of all detected alien planets have host stars that can flare..."</font><br /><br />We're talking primarily about systems with planets that are mega-Jupiters in close orbits. It may well be that either:<br /><br />A -- stars which flare tend to form planetary systems with huge planets in close orbits.<br />or<br />B -- stars that have huge planets in close orbit about them tend to be induced to flare more often.<br /><br />Again -- the data is skewed and *known* to be skewed. The subject at hand (namely 'foreign' solar systems) is also one about which we have *very* little data. It's therefore impossible to point to any aspect of the data and make a sweeping statement about the remainder of the dataset (i.e. other solar systems for which we have no data).<br /><br /><font color="yellow">"...planets, which are that close to their host star..."</font><br /><br />Again -- the data is <b>specifically</b> skewed towards locating planets with very small (and fast) orbits. The shorter the planetary orbit, the faster the 'wobble', the easier it is to detect. A solar system like our own with a Jupiter in a ~12 year orbit will not show a verifiable 'wobble' for (I believe) two complete revolutions. Most likely, planetary scientists watch planetary candidates for several orbits before making the announcement that they've found a planet. Since many of the found planets have 'years' that are a fraction of an Earth year -- this is reasonable to do. However, this very factor means the planets are close to the sun and are hot and tidally locked.<br /><br /><font color="yellow">"...none of these people are claiming certainty..."</font><br /><br />They shouldn't be claiming <b>anything</b>! The only solar systems detectable by the current methods are solar systems that are completely unlike ours. To then take that set of data and make ***ANY*** kind of statement about systems like
 
G

g_riff

Guest
<blockquote><font class="small">In reply to:</font><hr /><p>I have a bag of poker chips -- 15,000 of them. I've been pulling them out one by one and checking their color. Unfortunately, I'm color blind, and anything except 'blue' simply looks grey to me. I've pulled out 600 chips so far. 150 of those chips were blue and the remaining 450 'grey'. Is it reasonable **AT ALL** for me to then make a statement that red chips must be rare?<p><hr /></p></p></blockquote><br /><br />I like your analogy. Continuing with that analogy, though, I have to ask, do you know how many of the chips that really were pulled out were grey? How many systems have they looked at, and been unable to find any planets? I'm not talking about systems they haven't looked at for long, but ones they've looked at for long enough to safely say there are no hot jupiters orbiting around them. Mabye all 600 of the chips they pulled out were blue after all. Or at least the vast majority.
 
M

mrmorris

Guest
<font color="yellow">"Mabye all 600 of the chips they pulled out were blue after all. Or at least the vast majority. "</font><br /><br />Even 1-in-4 (i.e. 150-in-600) is stretching the bounds of possibility to the breaking point. If <b>every</b> star or nearly every star they pointed a telescope to was wobbling and generating 'extra-solar' planets -- that would be <b>**huge**</b> news. Since the actual ratio of 'stars viewed' to 'planets found' has not been plastered about in media releases, it's fairly safe to assume that it's nothing remarkably high.
 
I

iron_sun_254

Guest
Sigh...I think you're missing the irony in this article. The reason "5 out of 5" researchers agree is because only 5 were chosen and it's likely they were chosen specifically because they held this viewpoint. I could start a panel of 5 scientists who believed global warming is not caused by humans and put out a similar headline.
 
Status
Not open for further replies.

Latest posts