Evidence of Gravitational Waves, or Confirmation Bias?

[Michael Mozina]

Benburch:

High energy physics has a thing called a trigger processor. This is the thing that constantly looks at the detector suite and which decides when something interesting enough to log and analyze has happened. Often these are exquisitely sensitive and will provide a very high percentage of interesting events. But the people running the show are never happy with them. They fear that the trigger, whose function is to throw away data, is going to throw away the events they are trying to find, so they disable or hobble the trigger.

This means they need to waste a lot of computer resources sorting out the things the trigger would have found for them from all the noise.

But when they do find it, the fact of ignoring the trigger does not in any way degrade the observation!

I hear you and I tend to agree with all of that by the way. My beef really isn't with the fact that there was a veto, or the fact that they chose to override that veto. My issue and my criticism relates to the fact that they erroneously and inaccurately claimed that there was no veto within an hour of the event in their published and peer reviewed paper, and they never bothered to explain what actually caused the veto, or why it achieve a "high confidence" figure. All of that information *should* have been included in the published and peer reviewed paper but it wasn't. Instead we got a pure whitewashed and fabricated version of events in the *published paper*. That's simply unethical IMO.

I think they should have made a real effort to fully *explain* the purpose of adding that veto, the cause of the veto, the actual hardware it's based upon, an explanation of why it was vetoed with "high confidence", and a lot more information related to their determination of "safety", specifically some quantified estimate of some sort.

I didn't even get that from LIGO. All I got was a slightly more verbose (not that much more verbose) version of the LIGO magazine account of events which really only verifies my criticism that the published version was a snow job.

Selfsim:

Yes .. and that's almost exactly how the vetos are used in notifying the 'downstream' follow-up teams (eg: such as other observatories).

See my comments above. I don't take issue with the fact that they received a veto or that they chose to override that veto. I have a problem with their fabricated explanation of the veto events around GW150914 in the *published* paper.

LIGO has said that the veto which was over-ridden, was untested and was subsequently determined to be 'unsafe'.

Why was this particular veto added in the first place, and what kind of auxiliary hardware was attached to that particular veto? Was it an EM reading that caused it to fail, or was it something else altogether? How and why did the software decide to reject that specific signal with "high confidence" the first time through? LIGO has been rather frugal with any specific details. Perhaps you would care to enlighten us? You might start by explaining why that particular veto was added in the first place, and what kind of hardware it was associated with?

Exactly how "safe" was their manual override? Was it 80 percent safe to override that veto? Was it 95 percent safe? 5.1 sigma safe? All I've seen thus far is a little bit of internal jargon and a request to simply "trust us, that veto software was wrong" without any *specific* details about anything.

This would be as expected during what was still an Engineering Run phase (and not an Operational Run phase). There was still a chance that testing was being carried out when the veto went active however, investigations conducted at the time, (over ~2 hour period), determined that no manual testing was being conducted.

Like I said, I don't really have a problem with their actions. I simply have a problem with their whitewashed version of events in the published paper which left out all the relevant details about that veto.

The chances are that the veto assumed that manual tests were being carried out when they were not, and so the veto was (justifiably) over-ridden. This all comes from a cursory read of the LIGO newsletter in question.

How *exactly* (be specific) did you get the idea that "the veto assumed that manual tests were being carried out when they were not" from anything that is written in LIGO magazine? I certainly didn't get that from either the LIGO magazine account, or from their response to my first email. I'm kinda waiting with baited breath to see if they'll be a little more forthcoming with the details about the veto and their decision to override it. I still have no idea what *exactly* caused the veto, what specific hardware was involved with that veto, how it decided to veto that particular signal with "high confidence" rather than say "low confidence", or any pertinent details.

I notice that MM has received nothing more than confirmation of what was in that newsletter from LIGO, so nothing has changed, and he has no new information for consideration.

They just confirmed to me that they gave a less than accurate account of data quality events in their published paper. That's all they've done so far. That information just disturbs me, particularly without any real details about the veto. All they did was verify my criticism about their less than accurate account of events in the published paper.

I do not need any additional information as it relates to the content and theme of my paper. There is no empirical evidence that the signal in question was celestial in origin. If they applied their same process of elimination method consistently, they should have put this signal in the "unknown cause" category, not the "gravitational waves surely did it" category. The confirmation bias problem still applies and it still remains the Achilles heal of the their entire paper, and in fact *both* papers. All LIGO has done so far is confirm my criticism about their less than accurate portrayal of data quality veto events in the published papers. That's not comforting.

[Michael Mozina]

I must say that I find it very interesting and rather revealing that none of the "international skeptics' over at ISF have been able to find a single error in my paper.

Even the veto problem that I mentioned in my paper turns out to still be a very serious problem because virtually nothing about the specific details of that particular data quality veto of this signal have ever been made public. The polite but terse response that I've gotten from LIGO so far really doesn't explain any of the necessary details that are mentioned in my paper, so there is still no way to evaluate the merits of that specific decision by anyone outside of LIGO.

Frankly I'm a little frightened for "science' if this so called "discovery" of gravitational waves is rewarded with a Nobel Prize. The confirmation bias that is necessary to believe this claim is absurd. We are told by LIGO that two invisible objects merged in a galaxy far far away and released the energy equivalent of three full solar masses in the form of gravitational waves in about a quarter of a second, without releasing any detectable EM radiation or neutrinos. The signal we presumably "picked up" looks remarkably like an ordinary "blip transient" event in frequency and duration, and the detectors were still in the "engineering run" stage at the time, yet somehow LIGO already knew that it couldn't be a blip transient event. The fact we didn't even get an accurate account of data quality veto events of this signal is simply mind boggling. They even used a skewed process of elimination method to eliminate potential causes of the signal *other than* any and all celestial origin claims. All celestial origin claims were simply excluded from the same process of elimination methods which were used to eliminate every other potential explanation of the signal.

No peer reviewer and nobody in the public has ever seen a detailed and accurate explanation of the data quality veto of this specific signal, not in the "peer reviewed" account, nor in *any* public account to date.

The fact that every *other* potential cause of this signal was eliminated from consideration due to a lack of external corroboration in auxiliary hardware, whereas no celestial origin claims were ever excluded for the same reason says volumes about the nature of their extreme bias. The fact that there isn't even a possible category for "unknown origin" for *any* signal in their methodology also demonstrates the blatant nature of their bias.

This really is a textbook example of very poor scientific methodology, and it definitely should not be rewarded, or I fear for the future of astronomy. We could end up stuck in the dark ages of physics forever while our children and our grandchildren are told that supernatural invisible processes determine our fate. There really will be no empirical cause/effect difference between science and religion at that point.

I was really excited when I first heard about this so called "discovery", but now I'm just worried for science in general, and astronomy in particular. Even the BICEP2 fiasco paper wasn't this bad because nobody gave an inaccurate account of events in the BICEP2 paper.

[Royboy]

I used to live near here and have visited this facility. http://gravitycentre.com.au/observatory/ Open to the public at a cost to recover the $millions of wasted taxpayer's money. Strange that the Yanks found a wave with LIGO yet no wave was announced from Gingin. Only days after in their newsletter was it announced that a wave had been discovered (by LIGO). My opinion is 'If you want to see gravity waves go to a beach.

[Michael Mozina]

Well, LIGO hasn't responded to my last round of specific questions about the VETO, and it looks like it's 'run and hide" time over at ISF.

I must say that I knew that my paper would be difficult to argue against, but I did expect to see some decent arguments made against at least a few of the points I made. I haven't seen any yet, not even from the "hater posse" as I call them.

I think the flagrant confirmation bias problem is pretty much indefensible. I certainly haven't seen any argument agsinst that point, and more than a year later we still have no visual confirmation to support the claim that this signal was necessarily celestial in origin. We do have ample evidence however of them "gaming the system" in terms of the lack of any falsification mechanism ever being applied to celestial origin claims as they were with all other potential environmental factors. The sigma problem remains serious because it in no way connects the signal to anything celestial, and it doesn't even eliminate other environmental factors as the cause of the signal. All it ever did is verify that it probably was a real signal, but it doesn't speak at all to the "cause" of the signal.

LIGO has at least verified that ti's published account of veto events was not accurate, and they still have not been forthcoming with any specific information about that veto yet, certainly not to the public. I have exactly *zero* confidence that either signal was actually related to gravitational waves because there is zero evidence that the signals was celestial in origin.

I remain open to the possibility that that a "real" celestial event could be visually confirmed or confirmed by neutrino measurements, but I simply see no evidence at all from the LIGO account of events that would suggest that the signals in question were celestial in origin.

Extraordinary claims require extraordinary support according to the mainstream, except when it comes to their own claims, in which case anything goes apparently. There isn't even any "ordinary' evidence to support their claim let alone "extraordinary" evidence.

[kell1990]

Michael,

You are wasting your time arguing with the folks over at ISF. They will never admit that you are right and they are wrong, Ever.

So give it up, because we have already won. Their arguments have been decimated because they have been shown to be fraudulent by the evidence provided.

They just can't stand it, but there is still a corps of illiterates who have no idea of how the universe actually works who are clinging to an outdated idea of the process.

[Michael Mozina]

Michael,

You are wasting your time arguing with the folks over at ISF. They will never admit that you are right and they are wrong, Ever.

So give it up, because we have already won. Their arguments have been decimated because they have been shown to be fraudulent by the evidence provided.

They just can't stand it, but there is still a corps of illiterates who have no idea of how the universe actually works who are clinging to an outdated idea of the process.

I hear you. I guess I just expected to hear the folks at ISF put up some type of explanation/rebuttal for at least few of the points that I made, but apparently they can't find any serious problems in my paper.

The more I think about this LIGO claim, the more cheesy it sounds. LIGO literally had to cheat in terms of their process of elimination methodology and they didn't even provide an accurate account of veto events of this very signal. If that's the kind of unethical nonsense that they have to stoop to in order to make their case, it's clearly not a very strong case. Even the sigma problem alone should be enough to blow away their claim, but the confirmation bias problem in their methodology sticks out like a sore thumb. Every *other* potential cause of the signal was *excluded* based upon a lack of external corroboration, but all celestial origin claims were given a free pass, and no exclusion possibility was applied to celestial origin claims.

I keep wondering how long this gravitational wave charade can last if they can never provide any visual or neutrino confirmation of their celestial origin claims. They're already 0 for 2, and it's been really quiet for the past year.

[Michael Mozina]

https://www.reddit.com/r/Astronomy/comm ... idence_of/

FYI, I also started a discussion on my paper over at Reddit.

Still no word from LIGO in terms of providing any any answers to my last round of veto questions by the way......

[Michael Mozina]

https://www.reddit.com/r/Astronomy/comm ... idence_of/

FYI, I also started a discussion on my paper over at Reddit.

Still no word from LIGO in terms of providing any any answers to my last round of veto questions by the way......

FYI that Reddit thread is worth reading IMO. It's quite telling from my perspective that thicktarget (Selfsim?) seems to be focused almost exclusively on the veto problem of the LIGO paper, and he's pretty much been reduced to handwaving at or avoiding all the serious methodology problems listed in my paper, including the fuzzy sigma issue, the fact they can't distinguish between blip transients and gravitational waves, and the blatant confirmation bias problem in the methodology they used.

I'm not quite sure why thicktarget feels compelled to fixate most of his attention on the veto contradictions since it's really just a minor part of my criticisms and only warranted a couple of paragraphs in my paper. After all those posts, he still hasn't answered any of my *specific* questions about the original purpose of that veto, the specific hardware channels that it's connected to, why that veto achieved a "high confidence" figure, etc. He's not able to quantify anything related to "safety" either.

He does seem pretty emotionally attached to trying to defend LIGO's work, although I'd have to give him a pretty low score since he's been reduced to handwaving at their fuzzy sigma problem, and their confirmation bias problem, and he won't touch LIGO's blip transient problem with a ten foot pole.

It's still worth a read through if only to watch which topics he's forced to avoid. He seems to outright avoid any discussion of the blip transient problem for instance. I'd assume that's a rather touchy subject since they have no logical or methodical way to differentiate between a gravitational wave and an ordinary blip transient event which happens to be observed in both detectors. That's a serious scientific fail.

He doesn't really have much to say about their fuzzy sigma problem other than to erroneously claim that it's "impossible" to link the sigma figure to their cause claim, which is simply not true. If they had a visual confirmation of a celestial event during the timeline in question, in the direction predicted by their mass merger models, they very well might be able to come up with a sigma calculation that relates to the probability of the signal being celestial in origin. Since they didn't have a visual confirmation of either of their gravitational wave claims, he's right about the fact that it's impossible to come up with a sigma calculation that is in any way related to the claimed "cause" of the signal *without* a visual confirmation, which is exactly why they shouldn't be calling this a "discovery" in the first place.

Based on the conversation thus far, I'd have to assume that the fuzzy sigma problem, the blip transient problem, and the confirmation bias problem are the most difficult problems for them. The veto issue is just embarrassing as hell so that's where all the attention seems to go even though it's a relatively insignificant issue compared to the other more serious problems with their methodology.

[Zyxzevn]

How do you calculate a Sigma and how does it degrade with systematic non-white noise and after using filters?

Using the same measure, what was the sigma for this phenomenon?



The problems are very similar:
Systematic noise: shape depends on direction of the sun, a lot less than LIGO.
Filters: filters make shapes smoother than they really are, it also forms pyramids.
Resolution: about same amount of signal samples.

I was not joking, when I compared LIGO's find with the face of mars.
But with LIGO we can NOT make a second picture with better resolution.

[Michael Mozina]

How do you calculate a Sigma and how does it degrade with systematic non-white noise and after using filters?

Using the same measure, what was the sigma for this phenomenon?



The problems are very similar:
Systematic noise: shape depends on direction of the sun, a lot less than LIGO.
Filters: filters make shapes smoother than they really are, it also forms pyramids.
Resolution: about same amount of signal samples.

I was not joking, when I compared LIGO's find with the face of mars.
But with LIGO we can NOT make a second picture with better resolution.

Ya, I did think it was very interesting that thicktarget basically conceded that the sigma calculation was essentially entirely unrelated to cause. That alone should preclude LIGO from claiming this was a "discovery" of any sort. Their sigma calculation doesn't have anything at all to do with their claim about the cause of the signal.

They didn't even offer any quantification of the issue of "safety" related to any veto, including and especially all the vetoes that they deemed "unsafe", including the specific one that vetoed that exact signal.

The worst part IMO is that that LIGO cannot even mathematically or otherwise differentiate between ordinary blip transient events that happen to affect both detectors and "gravitational waves". Their method would literally erroneously classify a blip transient event as a gravitational wave discovery. They don't even have a veto method that is capable of filtering out blip transients and they admit that much in the paper! This is just an *absurd* claim from start to finish. It's not like there's *just* the fuzzy sigma problem or *just* the blip transient problem. They also have that *glaring* confirmation bias problem going too. There's at least three major deal breakers in their claim, all of which preclude them from claiming they "discovered" anything.

The signals should have ended up in the "unknown origin" category if they'd followed their process of elimination methodology consistently, but they instead opted to simply "cheat" by not providing a category for "unknown source", and giving all their claims a free pass. They literally have no way in their methodology to falsify any celestial origin claims as to cause.

This is definitely the worst "discovery" paper that I've ever read. Even the BICEP2 'fiasco" paper didn't include *misinformation*. There were the same basic problems present in BICEP2 in terms of trying to claim to have eliminated every *other* potential cause of the signal in their paper too, but at least they didn't flat out misrepresent the facts as LIGO did with that veto. The peer reviewers of this claim of 'discovery" got a *false* story about their being no vetoes present within an hour of the event, when in fact the specific signal in question was vetoed within 18 seconds and it remained vetoed for the next two and a half hours. No quantification related to safety was ever offered. When folks start misrepresenting the facts, I start to wince. I've never even seen such an unethical thing in any astronomy paper, let alone a "discovery" paper. Yikes!

[Cargo]

If LIGO could imagine itself detecting the Face of Mars it would look like this

A LIGO filter of the Face of Mars

The evidence they found is like claiming every DOT on this picture is WRONG, except ONE. Can you find it?

[Michael Mozina]

It was actually very enjoyable to get the chance to debate my paper in a semi-hostile environment. It went better than I imagined actually.

I still have not seen anyone take apart any of the main arguments that are listed in my paper. ThickTarget seemed to be almost exclusively fixated on the veto controversy, but frankly that is a very minor issue in terms of the actual problems with the LIGO claim. Furthermore, LIGO has already confirmed that there was an actual veto of this specific signal within 18 seconds of the event, so the arguments in my paper still absolutely apply, regardless of their rationalizations for their less than truthful account of events in the published papers. It's almost an irrelevant issue in terms of the arguments presented in my paper, other than the fact that it speaks to the ethics (or lack thereof) of LIGO.

I'd say that the sigma problem, the blip transient problem and the confirmation bias problems are right on target. I still haven't seen a valid argument by anyone yet that attempts to deal with any of those core issues.

After all the laboratory failures of dark matter over the past decade, the numerous revelations of problems in the mainstream baryonic mass estimate techniques, the BICEP2 fiasco, and now this LIGO paper, I'm almost tempted to write a book entitled "The Cosmic Con Game". It feels like the mainstream is running a gigantic scam that is specifically designed to bilk taxpayers out of as much money as possible to support the mainstream, and in return they're just making stuff up on a whim while spewing disinformation to the public about EU/PC theory.

That LIGO paper is just *bad science*. I didn't even realize how bad it really was until I sat down to write my paper. I simply cannot "un-see" all the problems in their methodology anymore. LIGO's claim is really bush league science, and it's based on pure nonsense. No doubt they'll award LIGO a Nobel Prize this year to try to cover it all up, and sweep the numerous problems under the rug.

It's pretty sad when 95 percent of someone's beliefs are nothing more than placeholder terms for human ignorance, but when they make up more invisible entities by the day, it's just absurd. The LCDM model is just terrible. It's almost a complete departure from empirical physics, and there's nothing useful about it in terms of it's predictive ability in the lab, or in space. Everything about it is "postdicted" to start with, and it fosters an "anything goes' attitude in physics. I don't think that the mainstream even cares anymore that none of their stuff actually works. All they seem to care about is procuring the next round of funding.

[Michael Mozina]

After my 'trial run" over at Reddit, I thought that I would try my hand at a slightly larger audience, so I started a new thread on the topic of gravitational waves over at astronomyforrums.net. I basically just copied and pasted the opening post of this thread over there, including the link to my paper. I have posted to their forum in the past a few times, but I had not posted there in quite awhile, the last time was maybe a year ago.

http://www.astronomyforum.net/general-astronomy-forum/

Since I'm still considered to be a relative 'newbie" to their forum, my gravitational wave thread had to be "approved' by a moderator who sent me a very nice private message last night saying that they were in the process of reading and reviewing my paper and they'd let me know. I thanked them for considering my new thread and I went to bed.

When I went to login to their forum this morning, I received the following message:

You have been banned for the following reason:
argumentative demeanor

Date the ban will be lifted: Never

So there you have it. My LIGO paper is evidently a bannable offense in astronomy.

Like I said before in the other thread, the mainstream simply cannot handle an open and honest public debate. They live in fear of even being questioned.

[Zyxzevn]

So there you have it. My LIGO paper is evidently a bannable offense in astronomy.

Congratulations. You are now a certified critical thinker.

[Michael Mozina]

Well, some time has passed at this point so I'm guessing that I'm probably not going to get a detailed response from LIGO explaining why the veto in question was originally written and installed, or what it was designed to veto out to start with, or any response to the quantified definition of "safe/unsafe" in terms of this veto. I did by the way ask them a brief question about the confirmation bias problem that is related to having no visual confirmation of the signal. I guess these are questions that will simply go unanswered.

That veto really does tend to undermine the entire credibility of their claim with respect to the cause of this signal, particularly since the sigma calculation they provided is completely unrelated to cause. We, the tax paying public, have no idea why that specific veto was added to their system, or what it was designed to filter out in the first place, so there's no way that anyone outside of LIGO could make any assessment as to the quantified "safety" of that veto.

I must admit that I'm a tad disappointed that LIGO didn't respond to my specific questions, but I'm not particularly surprised.

I would say that based on the responses to my paper that I've revieved around the internet thus far, that the confirmation bias problem in their paper is essentially indefensible.

Thus far most of the attention I've gotten has been focused on the veto of this signal, but that simply leads us right back to the 'safety' issue which tends to be rather problematic, particularly since there is no effective veto for blip transients.

The veto which took place may have vetoed out something that it was designed to filter and and which it should have vetoed out, or it may have vetoed something that it shouldn't have, but there's no way for anyone outside of LIGO to know that from their published paper, or any subsequent materials released by LIGO.