Guess The Plot 4: Two Hypotheses
    By Tommaso Dorigo | June 9th 2011 12:38 AM | 32 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
    Today's guess of the plot will be very poor of comments on my side, for reasons I will disclose in due time. Also, it is unfortunately a rather expert-only one, since it represents something that experts will surely recognize for its generalities, while only the details will probably remain mysterious to them.

    I will be glad to read your guesses in the comments thread, which I hope will be plenty and insightful. Hopefully in a few days I will disclose what the plot represents.

    UPDATE: In place of the plot, events force me to put in its place something else, for just a few hours. See below - the change is not dramatic, but what is added is probably giving a stronger hint than what is taken off.


    It looks like the output of a boosted decision tree, or something like that, which should give a negative value in the case of a new physics signal and a positive one in the case of just SM background (the expected output for the two hypothesis are the dashed lines). Obviously the experiment confirms clearly the "background only" hypothesis.

    But I have no idea what the x axis may be, nor what the "injected" log likelihood ratio (LLR) means...

    So .... where did you get this plot from ? You should not know abut this, and for sure making it public before it is public is not kosher.

    So given the above two comments and a recent post about rumours on "not even wrong" I can guess this plot may be refuting a recent interesting result published by a certain collider experiment?

    btw I hope I am on the wrong track...if not probably I agree with poster number 2 this is not "very sporting"...

    since you're not part of D0, I guess it's not what's going to be presented tomorrow at Fermilab :-)

    I guess it might actually be CMS' take on one of the CDF bumps, perhaps the 150 GeV almost-5-sigma one. They're both "your" collaborations after all so I guess you might be one the analysts that are currently looking into this. (and the graph colours clearly recall LHC official result plots.) that I think about it: it might still be presented tomorrow, but in Perugia ;-)

    Are you working anything at all if you have time to all this?
    What for are they paying you I am wondering ...

    Are they paying you for reading this blog?

    LOL, not to mention that I am paid to write it :)
    you do have a lot of nerves to change the plot to this new one !!!

    I consider it is outrageous what you do! This is unethical and not fair to the community!

    Anonymous commenters have no privilege to discuss named people's nerves or other behaviour. What I find unethical is that you hide your face and throw a stone.

    Please let me know in what way am I damaging whom, or in what way is this unethical or unfair (to whom ??).

    Tommaso, I don't no how you do all this, but you are a genius...
    Thanks for bringing so much fun in the particle community! ;-)

    Bonny Bonobo alias Brat
    Actually Tommaso you are very human and we all love you.
    My article about researchers identifying a potential blue green algae cause & L-Serine treatment for Lou Gehrig's ALS, MND, Parkinsons & Alzheimers is at
    Much appreciated. In fact, I do try to enjoy my job as much as possible. It is a pity that there's a sizable number of colleagues who lack any sense of humor, and who consider themselves and what they do so important and serious that anything off the path is to be condemned as unethical.

    ha ha - I think I get it! If so very funny ;)

    hmm...aparrently I was wrong to change my mind, and right in the first place.

    so you've stolen this from D0 before publication.
    Surely you must be proud of that!
    Sunt pueri ...

    Hi Tommaso

    Surely as a physicist you understand the that leaking other peoples work is wrong? To say nothing else of the scientific processes it is not fair to the people who spend the time doing the work for others to release it ahead of them.

    If you honestly believe that this is acceptable behaviour then please send me the CMS plots for Higgs exclusion limits, m_gamma,gamma and any other interesting plots. I'd like to have a look at them before the summer conferences and possibly release them to the world early. Im sure you dont mind.

    I wouldn't want to be accused of being unethical so I've left my name and email.



    I agree with this - I would be quite upset if I had worked hard on an analysis and then found one of the main plots appeared on leaked on someone's blog. I don't see this can be good for science either - I think there are good reasons analyses go through internal reviews so we should all just be patient and wait for the official release. Presumably you do not fully agree with me? Well anyway thats my 2 cents. I generally really like your blog, but am quite disappointed with this post if the original plot really was the one from the d0 paper....btw when I changed my mind I thought for a moment the plot was a hoax made (to encourage a debate and be a bit mischeivous) to look like what d0 would release - maybe I remember wrong, but the one in the d0 paper seems identical to me.


    Hi Mark,

    the plot was from the DZERO analysis, but it was amended such that nobody would be able to tell from which experiment it was (unless he or she had already seen it), nor whether it was showing GeV or seconds or horse powers on the x axis. The plot as I pasted it here was useless for any practical purpose. So nobody is harmed in my humble opinion.


    I did nothing like what you describe. I placed in my post a internal figure after removing axis labels, experiment name, and any information which could conclusively tie it to a particular experiment or a particular search.

    As for the others leaking stuff to me - well: you must be living somewhere else than in a Physics Department, where internal documents are shared freely among colleagues since a couple of centuries ago, and nobody ever really complained.

    Hi Tommaso

    Your modified figure indeed has had the labels removed, although in the original post you had the axis labelled.

    Im a theorist here at Fermilab, and whilst I do enjoy looking at results from colleagues and discussing them with them I would never release their results to the wider world before them. That is their privilege I feel.


    That is false Ciaran, the original picture (pasted below) did not have any label on the x axis, thus making it impossible to figure out what unit was being displayed, or in what range.

    I insist I did not do what you claim. I did not release any result from colleagues, period.

    In my opinion what Tommaso did here is a breach of copyright. The figure in any shape and form is the intelectual property of D0. Presenting it to the public while under media embargo is not acceptable. Also messing with it in any way (erasing labels, units, logo's and so on) without the agreement of the owner is not right at all.

    You know that your blog is read by people in the community and in the two plots you gave them all they wanted to know.

    Tommaso as also other people said please make public any CDF or CMS plots that you find interesting. And while you mess with them photoshop them in a way where you erase/add peaks of new/old physics. If in the future we'll hear all the official news from our experiments from official lab press releases and people will be kept out of the collaboration reviews you'll know who to blame.

    In my opinion what Tommaso did here is a breach of copyright.
    One of us does not know what a copyright is - and it isn't me.
    Dear Htpft,

    there is no copyright in internal collaboration plots - it is simply impossible to copyright anything we produce with our desktops.

    As to the fact that "people in the community" read this and may understand what that is, you are again wrong. I showed it to people in the community and they were guessing it was coming from ATLAS, and nobody understood what the injected signal was. Only a DZERO member could figure that out. You could, however, "guess" what the plot might be, with small probability - and that is what the game was.

    In the future, as is already happening and as I predicted a long time ago, most of the interesting physics results will be out and public before the collaborations have the time to approve them. This is just the necessary consequence of having 3000-strong collaborations: somebody will always disagree with keeping secret stuff that more people want to know.

    You should know better, Tommaso. This is not about keeping secet stuff but rather avoiding to spread half-baked results.

    I do know better than what you quote. Half-baked is the H->gamma gamma "result" by ATLAS; but a draft of a paper v9, two days before an already scheduled seminar, is not half baked in any way.


    any CDF or CMS v9 paper you wanna share with us ? =))

    Or I'll say anything with a version greater than 5 will be almost ready to eat.
    You should not "keep secret stuff" from your readers ! How about put you actions where you're mouth is and publish what is secret and what you hate being secret.

    Dear Anon,

    I do not distribute internal stuff from my collaborations. In doing this I comply to the rules of these experiments, as I consider it a privilege to be part of them. There is a misconception about this: in fact, I have been attacked in the past for talking about published results in a way that the collaborations considered not appropriate, or by explaining how there existed lots of internal material from those experiments that people could access with google searches, or by reporting humorous discussions occurred within internal meetings of those collaborations. Never did I report on unpublished material from CDF or CMS (although once I did put out a W mass measurement by CDF which I had been told was public but was only going to be so in two days, and in a few hours was notified and deleted it -that was a mistake done in good faith and it did not cause any criticism by the experiment).

    What I believe about the free diffusion of information, and on how things might work better if the rules were different, is another matter.

    Now, there are people in other experiments who sometimes diffuse internal stuff, for whatever reason. If I get that material, I usually do not distribute it either -I do not particularly like to be the recipient of hatred by large groups of people. However, I have no obligation in this case, and at times I may decide to act differently.

    What happened this time was that I distributed NO INFORMATION, but nevertheless I showed to the dzero folks that the document had been leaked. Not a big deal in my opinion - the stuff was going to be public in less than 24 hours anyways. In such a case, the matter is just formal and one cannot really speak of "damage".