Banner
    Another Betting Offer
    By Tommaso Dorigo | June 6th 2011 08:59 AM | 51 comments | Print | E-mail | Track Comments
    About Tommaso

    I am an experimental particle physicist working with the CMS experiment at CERN. In my spare time I play chess, abuse the piano, and aim my dobson...

    View Tommaso's Profile
    I see people around very, very interested in what the CDF experiment has recently unearthed. I am talking, of course, of the jet-jet resonance candidate that they observe in their W+jets sample. A recent update of the previous result shows that the significance of the bump is just short of the coveted five-sigma: that is to say, for non-insiders, there is now a chance in two or three millions that the effect is due to a statistical fluke.

    If not a fluke, surely it must be a signal of a new, unpredicted particle, right ? Well, in fact many of my colleagues keep asking me about this new find. Some ask if the spin can be measured (no), others if the signal appears together with both electron and in muon decays of the W (yes), still others if it has been seen also associated with a leptonic Z decay (no). Few ask if DZERO sees anything (not yet clear), or if CMS sees anything (no comment), or if ATLAS sees anything (apparently not but their stats is still poor). Fewer still wonder whether this is finally the Higgs boson by now (no, it cannot be). On the theoretical side, papers flourish of course.

    Now, with a five-sigma result out there, you must be willing to bet this is the discovery of the century, right ? If there is a chance of one in 2 million that the particle is not there, the Standard Model is disproven! There is a one-in-two-million chance that the SM still holds. Ah, reverse inference, we love you.

    I think you (yes, you) might be willing to bet money on this. If they told you there's a one-in-two-million chance that Nadal wins the next open, would you not bet with one-to-one odds on that ? Sure. But what fool would then be offering you an even bet ?

    Here I am. I realized I have offered bets on other new signals in the past, with alternating fortune (in terms of people taking or ignoring the offer). Now, I feel compelled to do the same here, since if I did not, it might look like I believe this indeed IS a new particle, while I want it to be clear that I believe it is not.

    Here's the bet: $100 (a more popular sum than previous bets, to entice you) that the new dijet bump in the CDF W+jj analysis is not coming from the production of a yet-unseen new particle. Even odds. Payable as soon as it is either proven to be a new particle (that is, if it is confirmed by the LHC by the end of next year), or if it is unconfirmed by then. I might even wait one more year to be paid, if you were willing for more security that the bump is not a particle.

    You are eligible if you either A) are a known physicist (ie if you have some reputation at stake), or B) find a known physicist who guarantees you will pay the bet in case you lose. Please line up tidily in the thread below - I accept up to 20 takers. But hurry! Once DZERO confirms it, I will retract the bet faster than you can say "bump".

    About the resonance: if you want more commentary, see Peter Woit's wondrous Not Even Wrong (you might in particular be interested to read the comments thread). And I should of course remind all that for once Jester did not oversleep on this one (he actually mentioned Punzi's Blois slides before I had a chance to see them, shame on me).

    Comments

    lumidek
    Even odds is ludicrous. From a perspective of decades, people believing that the Standard Model has fallen make an extraordinary claim that deserves an extraordinary bounty....

    Moreover, you're likely to know some secret data about the CMS - no comment, right - so this is a case of insider trading, isn't it?


    Concerning the adjective you used for the world's parasitic crackpot headquarters, you're surely joking, aren't you?
    dorigo
    Ok, the SM is the SM, but five sigmas are five sigmas Lubos. Please list a few five-sigma signals from the past thirty years that have been canceled by further investigations. I know some, but these are real oddities from dubious experiments, not a signal confirmed with added data by a 25-years-old famous experiment.

    Cheers,
    T.
    So aside from DAMA 8-sigma below, let me list the 4.2-4.8 sigma 150 Wjj bump claimed by the 25-year-old famous CDF experiment that was dismissed by an equally old and famous D0 in 2011:

    http://motls.blogspot.com/2011/06/d0-denounces-cdf-for-4-5-sigma-claim-o...

    lumidek
    Dear Tommaso, according to Jester,
    http://resonaances.blogspot.com/2011/06/more-details-about-cdf-bump.html

    the current relevance is 4.1 sigma rather than 5 sigma. If you want another example in which such as 4.something sigma claim has evaporated within a month, be my guest. Wu's claim of an enhanced diphoton 115 GeV Higgs a month ago did exactly that. While I am sure you feel the urge to scream that it's not the same thing because Wu is a woman, I feel that political correctness will strangle your voice so that you won't be able to raise this objection. ;-)
    you may have missed that the Wu thing was not a published result. Not even a categorized internal paper.

    lumidek
    I haven't missed it but fine. If you care about these formalities, I may pick another woman, DAMA/LIBRA that announced 
    http://arxiv.org/abs/0804.2741

    about 8.2-sigma observations of dark matter (sorry if I confused it with PAMELA in a previous version) and published it in European Physics Journal, a statement that's been claimed impossible by XENON10, XENON100, CDMS, and other bastards in the dark-matter-is-not-seen axis of evil.
    you're surely joking Mr (Stand-In) Feynman ;-)

    lumidek
    Not now! First, I didn't miss that the Wu's note wasn't published as an official document of the collaboration but only rumored on crackpot websites and later on more serious physics blogs as well.
    Second, I am not joking that DAMA/LIBRA claimed a 8.2 sigma signal - just read the abstract I linked to - and they're swearing that there can't be any systemic error of any known type to current physics. And I am not joking that these statements are being refuted by XENON and CDMS who have claimed to probe exactly this regime and claim to have found that there is nothing there, refuting DAMA/LIBRA at 90 percent confidence level, e.g.

    http://arxiv.org/abs/1104.3121

    If there's something else where you think I could be joking, you will have to be more specific. ;-)
    dorigo
    Funny, Lubos - DAMA-LIBRA occurred to me too as I wrote the comment above. However, that is not a new particle signal, nor is it necessarily a refutation of the SM.

    Cheers,
    T.
    lumidek
    Interesting, DAMA/LIBRA would probably disagree. They think it is both a detection of a new particle as well - which follows from it - a refutation of the Standard Model.
    Let me post the 2008 abstract so that we're on the same frequency and I don't have to expect that you will actually click and remind yourself. ;-)

    The highly radiopure $\simeq$ 250 kg NaI(Tl) DAMA/LIBRA set-up is running at the Gran Sasso National Laboratory of the I.N.F.N.. In this paper the first result obtained by exploiting the model independent annual modulation signature for Dark Matter (DM) particles is presented. It refers to an exposure of 0.53 ton$\times$yr. The collected DAMA/LIBRA data satisfy all the many peculiarities of the DM annual modulation signature. Neither systematic effects nor side reactions can account for the observed modulation amplitude and contemporaneously satisfy all the several requirements of this DM signature. Thus, the presence of Dark Matter particles in the galactic halo is supported also by DAMA/LIBRA and, considering the former DAMA/NaI and the present DAMA/LIBRA data all together (total exposure 0.82 ton$\times$yr), the presence of Dark Matter particles in the galactic halo is supported at 8.2 $\sigma$ C.L..


    Just if you didn't know, Tommaso, Gran Sasso is in a crazy country called "Italy" where they will put 7 top seismologists to the prison for 12 years because they caused an earthquake in L'Aquila by laughing at God that it wouldn't happen:

    http://motls.blogspot.com/2011/06/italian-seismologists-on-trial-as.html

    What a bunch of heretical witches who are secretly working with dark matters, too! Burn them at stake and back to the Middle Ages! Let's hope that God of quakes will forgive us by sacrificing these 7 scapegoats and the land in Southern Europe will never vibrate again. And the climate and weather will stop changing, too.
    JoAnne Hewett is said by Iamtheeye.com (5 June 2011) to have said that word on the street is that Dzero is not seeing the CDF signal.

    Giovanni Punzi in his slide 33 said that the CDF signal has no significant tagged component, and so could not be a Tquark signal.

    Even so, I am willing to bet now $100 (on the possibility that Giovanni Punzi's analysis might be problematic) that the CDF peak represents a low-mass Tquark state of 120-160 GeV (see the link to Truth Quark on the index.html page of my web site).

    As to timing, the issue might be clarified soon
    Tomorrow (7 June 2011 at 11:00 CEST) CERN will webcast about jets at ATLAS
    Resonaances said Dzero is supposed to announce its update on 10 June 2011
    and
    should indeed be resolved by the LHC by the end of next year (or maybe even by the end of summer conferences), assuming that the LHC continues to operate as well as it is operating now.

    Tony

    PS - Since I am not a "known physicist" and my reputation is such that I have been blacklisted from the Cornell arXiv, I am willing to send you a check for USA $100 plus the price of a Strega, say USA $150, and let you be both stakeholder and bettor.

    PPS - To be clear, my bet is not that my model is valid or invalid, but only whether or not this particular CDF signal represents my model.

    lumidek
    Right, Hewett said that a word on the street says that there is no bump in D0:
    http://iamtheeye.com/2011/06/05/fermilabs-bump-hunters-see-hints-of-new-particle/

    I am afraid that a Dorigo on the street may know even more than that. ;-)

    Did you hijack the name "truth quark" from the "top quark" in order to create a new "truthism quark" of a lower mass? :-)

    I am kind of confident that the LHC has already accumulated the data - 0.7/fb now - to resolve this issue, too.
    Vladimir Kalitvianski
    Luboš, tell me please if one day they will observe bare particles you so believe in. They must be around.
    dorigo
    Yeah yeah we want them bare, ready, willing and aplenty! ;-9
    T.
    Vladimir Kalitvianski
    So where are they? 
    lumidek
    Higgs is probably between 115 and 130 GeV, superpartners are probably between 1 and 5 TeV, string excitations are somewhere between 3 TeV and 10^{17} GeV, and similar for other types, and no one in the world knows anything with a higher certainty. If you want particles and crazy new physics to be promised, you will have to ask one of the crackpots such as yourself.
    Quite generally, you're fighting a straw man because I am among the first people who always loudly claim that it is very unlikely that we detect new physics - or even test remote profound phenomena - experimentally. This is true even on this very thread where I have claimed that the conditions of Tommaso's bet are unfair because he only offers 1:1 to someone who claims that the Standard Model fell in a particular (almost) random experiment. It may happen, and I kind of secretly do hope it will, but it is still unlikely. Less likely than Yes, so to say.
    Lubos,

    Stop the same old deception and ideological propaganda on Strings and SUSY. Both are dead ends, my dear friend, you are certainly delusional…

    Jason

    lumidek
    Hank had to fight very hard against the spam and waste filter to override the realization of his computer that a piece of spam was just posted by an aggressive overflooded imbecile.
    String theory is the most important and most accurate thing that the mankind has ever discovered, supersymmetry at some scale is its inevitable component, and all creatures who fail to recognize these things are just worthless and dull piles of fat and proteins.
    How can the worth of a human being depend on its belief in string theory?

    This attitude implies that you will never experience or find love.

    dorigo
    Hi Tony,

    you are well known and I can certainly accept the bet, but please send no check. Your signal has waited for a long time to be proven or disproven, and we need not be hasty... Let us put the resolution of this at the end of 2012. As for the Strega, I think $10 will suffice ;-) If I win, I will post a picture of myself drinking it from the bottle !

    Cheers,
    T.

    PS to be clear, I understand what you bet on, but I will pay whatever new particle this turns out to be, not just your pet particle !
    PPPS - I should also say that I do not consider my bet to be against the Standard Model, as I regard my 3-state Higgs-Tquark condensate system as an elaborationn of the Standard Model using its known particles, similar to a massive neutrino sector, not a replacement for the Standard Model and not a radical extension like conventional supersymmetry.

    Tony

    Lubos, the term "Truth Quark" goes back a long way (only old codgers like me actually remember when the term was used about as much as the term "Top Quark"). Here is a short history:

    The first generation quarks were named Up and Down.
    When one of the second generation quarks became known,
    it was called Strange because it was different from Up and Down.
    Then it became clear that the second generation also needed a pair of quarks
    instead of a single one,and when it became knownit was named Charm
    because it was the Charm that made Strange fit theoretically.
    Then the third generation pair of quarks became known, and there were two proposals to name them, both based on the letters B-quark and T-quark:

    The proposal that I liked back then and still use a lot was to continue the Strange and Charm line of thought, and name them Beauty and Truth.

    The proposal that eventually became the consensus popular use was to go back to the Up and Down terminology and name them Bottom and Top.

    Tony

    lumidek
    Very intelligent. If another physicist will join you in your belief in the particle, will she be considered to be a truther?
    I guess I should provide some reference for my statements about the term "Truth Quark",
    so here is a quote from the book "Gauge Theory of Elementary Particle Physics" by Cheng and Li (Oxford 1984, reprinted several times including 2000) in section 4.4 at page 122 :
    "... In 1977 yet another set of narrow-resonance Upsilons ... were successfully interpreted as bound states of yet another heavy quark, b (for 'beauty' or 'bottom') ...
    one anticipates at least one more superheavy flavour of quark ... t (for 'truth' or 'top') ...".
    Note that Cheng and Li listed the terms 'beauty' and 'truth' first.

    For a bit more history, Murray Gell-Mann was responsible for the term "quark", based on a passage from Finnegan's Wake ("Three quarks for Muster Mark."). According to his biography "Strange Beauty" by George Johnson: "... Quarks, the mischievous young scientist [Gell-Mann] proposed, came in three different varieties, which he dubbed "up", "down", and "strange".
    When other scientists later found hints of three additional "flavors" of quarks in the universal ice cream shop, what else could they call them but "charm", "beauty" and "truth".
    (The latter two were also called ... "bottom" and "top.") ...".

    Quarks were also independently invented substantially simultaneously with Gell-Mann by George Zweig, who called them "Aces",
    and
    even earlier (around 1960) by Liu Yao-Yang who was working at the University of Science and Technology of China, which was then located at Beijing, and who was actually the first to invent the quark model, in which his term for quarks was "Ceng Zi". He wrote a paper and submitted it to a Chinese journal, but it was rejected. After the quark model had been independently re-invented around 1962-1964, with most of the credit going to Murray Gell-Mann, the editors apologized for rejecting the paper. The last that I heard was that Liu Yao-Yang was still working at the University of Science and Technology of China, which is now at Anhui, in the fields of atomic and molecular physics, quantum field theory, and quantization of gravity.

    However, like the term "Truth Quark", the terms "Aces" and "Ceng Zi" are not the most popular in usage.

    Tony

    dorigo
    Hi Tony,
    sometimes the interface fails to realize yours are honest comments and puts them on hold. I do not know what triggers the decision... When this happens, your comment is not lost and you do not need to repost it (although feel free to do that if you so wish): I see it and can publish it at the touch of a key.
    If you want to be sure it appears, upon seeing it is not published immediately the best thing is to mail it to me (I cannot offer this to all, but the rate of comments on hold is smallish anyway).

    About truth and bottom, I have nothing to add to your nice recollection. Thanks in particular for the story about the rejected paper.

    Best,
    T.
    Hank
    I've put him on the whitelist a dozen times and it does no good - we use Akismet and they have a filter of their own so if he uses revolving IP or anything that may have been used by spammers it flags it.  We only flag here if it has 3 links but akismet is the biggest in the world at that so we let them handle the ordinary stuff.   Obviously he can create an account here and then it would never happen but I know people have low patience for having to login to places by 2011.
    lumidek
    Dear Hank, when you're fighting your computer this vigorously, by white lists and other weapons, haven't you ever considered the idea that your computer could be right and you could be wrong?
    Hank
    I'm not qualified to make that determination so I leave it up to the members to moderate their own columns that way - I just want to keep out malicious entities so all of you can talk about science and/or make fun of each other.
    dorigo
    Oh Lubos, this made me laugh out loud! Kudos for the best witty remark of the week.

    To Caesar what is Caesar's... I needed to let you know.

    T.
    lumidek
    Gratias tibi ago! Such a laughter is always appreciated, especially from reliable supporters such as you, Brute! ;-)
    I am just happy that in our era humans can still overrule computers, even if it now somewhat difficult to do so,
    and
    that we are not quite yet into the world described by Lubos in which computers are always right and people who disagree are always wrong.

    My favorite movie reference for such an era goes back to 1970:
    Colossus, the Forbin Project ( www.imdb.com/title/tt0064177 )
    According to the Wikipedia article about the movie,
    it is now being remade with Will Smith in the lead (human) role.
    It will be interesting to see whether the remake has the same ending
    described in Wikipedia as:
    "... in the final scene, the machine ... declares that Man has a choice between the peace of a human millennium under its authority or one of "unburied dead". ...".

    Tony

    Vladimir Kalitvianski
    And my favorite movie reference is that.

    It is in Russian, though ;-)
    Surely the W+jet authors on CDF or the CDF spokespeople would accept this bet? Or maybe they are too busy with interviews to notice... :-)

    lumidek
    Well, I agree! If the CDF authors wouldn't accept Tommaso's bet offer, they would be proved to be sloppy hypocrites. ;-)
    Lubos,

    Speaking of hypocrites: are you going to openly admit the demise of SUSY and String Theory when findings will become crystal clear?

    lumidek
    They have already become crystal clear. String theory is right.
    If you think that we will directly test string-theory effects at its characteristic energy scale, then you're naive because it's just far.

    If you think that the LHC has the capacity to answer any important question about the world or find God, you're naive as well.

    Let me send you to my two extensive answers to the questions "What if the LHC finds SUSY" and "What if the LHC does not find SUSY":

    http://physics.stackexchange.com/questions/6438/what-if-the-lhc-doesnt-see-susy

    http://physics.stackexchange.com/questions/6482/what-if-lhc-finds-susy
    String theory is right ... wow - and we used to think it wasn't even wrong ;-)

    Well, you cannot hold Lubos responsible for what you used to think.

    Tommaso, the D0 paper at hep-ex 1106.1457 entitled
    "Measurements of inclusive W+jets production rates as a function of jet transverse momentum in ppbar collisions at sqrt{s}=1.96 TeV"
    did not refer (as far as I saw) to the CDF bump at all.

    Do you expect further D0 papers that do explicitly discuss the CDF bump,
    or do you think that 1106.1457 is all that D0 will say (by official publication)?

    Tony

    dorigo
    No, Tony, I expect a D0 paper on the Wjj signal to appear in the matter of a week or so...

    Cheers,
    T.
    Presumably the D0 paper on Wjj "in a week or so" will discuss issues such as reweighting, which is discussed on pages 101-104 of fermilab-thesis-210-48 by Shannon Maura Zelitch to which reference was made over on Peter Woit's blog. When I Iook at figure 39 of the thesis, it seems to me that it shows how reweighting makes a bump on the falling part of a distribution go away.

    An uninformed naive question that I have is:
    When you adjust the weight parameters during the reweighting procedure,
    how do you know whether you are
    modeling the background more accurately
    or
    erasing a signal ?

    Is the next thing that LHC runs for another 10 days or so getting about 1/fb
    followed by some time for analysis
    and then EPS-HEP 2011 in Grenoble 21-27 July 2011
    at which LHC might shed a lot of luminosity on the matter (and maybe on the Higgs as well)?

    Do ATLAS and CMS use the same techniques for reweighting etc as D0 ?

    Will 1/fb be enough for ATLAS and CMS to resolve the matter?
    If not, how much more will be needed and when will it be obtained?

    Tony

    dorigo
    Hi Tony,

    LHC should run for three more weeks, collecting something like 1.4/fb per experiment if all goes reasonably well. I predicted 1.5/fb a few weeks ago, and I think I'm still good.

    Techniques are still being developed for this kind of business, and I do not know what the experiments will come up with. In principle a signal could be erased, yes, but typically one works in control samples where a signal should be absent. There are many ways to do so.

    I think 1.4/fb of LHC data is more than enough to draw a conclusion.

    Best,
    T.
    Tommaso, over on viXra log Philip Gibbs mentions problems being encountered by LHC in its current run (which problems seem NOT to be seriously impeding progress toward 1.5/fb or so by late June) such as:
    1 - beam dumps caused by "software errors that are being rapidly fixed"
    and
    2 - "UFOs have been another major headache. These are thought to be dust particles that fall into the beam."

    Software errors that are being fixed are to be expected, so that is really a good thing (part of the experiment maturing, so to speak),
    but
    as to the dust Philip Gibbs said "There is so far no sign of their numbers dropping as the beam zaps them up ... the dump thresholds can be changed in the worst hit areas ...".

    Why are there so many that the beam itself does not seem to clean them up as it runs?

    What are the "worst hit areas" and why do they have so much dust?

    Will the dust probably be cleared out during the maintenance break at the end of June?

    Tony

    dorigo
    I think these are quite minor issues... The machine has been performing well recently. I have no more information on this however...

    Cheers,
    T.
    I cannot afford to take your bet, even though I think the CDF result is probably real.

    Kea, Tommaso said
    "You are eligible if you ... find a known physicist who guarantees you will pay the bet in case you lose ...".

    I am on the same side of the bet as you would take,
    and
    I am willing to guarantee to pay the bet if you lose,
    so
    (subject to Tommaso's approval and if it is OK with you)
    you will get the $100 if you are right.

    Tony Smith

    Deal, Tony! We can split it 50/50.

    dorigo
    Hey guys, you are my friends, and I hate to win money from my friends!

    Furthermore, I should warn you about the information that has been published in Peter Woit's blog, which talks about DZERO excluding the CDF signal with a significance of over four sigma.

    If you still want to take the bet, though, I cannot stop you. It defaults to a  yes, but you can wait until tonight before confirming... The bet remains open as much as people wants. DZERO will publish their results at a seminar today.

    Cheers,
    T.
    dorigo
    Thanks Anon! I just published a post on this.
    Cheers,
    T.
    Hah, hah! Well, I believe D0 now, because no fairy fields sounds right.