Higgs Boson Live Blog: Analysis of the CERN announcement

Good morning and welcome to what is expected to be an exceptional day for physics as CERN announces important new results in their hunt for the elusive Higgs Boson. Here in one mammoth expanding post I will be reporting on the search for the Higgs Boson in straight forward terms free form silly analogies and patronizing phrases such as “for the layman”. I hope that many interested people with varying degrees of foreknowledge  will find the level helpful. I will explain the basic preliminaries first but if there is anything you don’t understand just Google it or wing it.

The present excitement started to build during the summer when it became clear that the Large Hadron Collider experiment was gathering data at a much higher rate than anticipated, meaning that they would soon be able to tell whether the Higgs boson exists or not and most importantly, what mass it has.

I am a theoretical particle physicist based near London independent of the teams working at CERN,  and I have been following events at the Large Hadron Collider and blogging about them since it started colliding protons in 2009. In a minute I will answer a few basic questions about the Higgs for the uninitiated, including the Paxman question “What does the Higgs boson look like?” Then I will be live-blogging the events from CERN as they happen, so first let’s look at the schedule of today’s events.

  • 14:00 – Fabiola Gianotti, spokesperson for the ATLAS collaboration delivers a 30 minute summary of their latest developments. ATLAS is the largest particle detector ever built and it sits on an intersection point of the Large Hadron Collider rings to observe the trillions of particle collision events taking place.
  • 14:30 – Guido Tonelli will talk about similar observations at the CMS experiment. CMS is another equally sophisticated but different and complementary detector placed diametrically opposite ATLAS on the LHC ring gathering another independent set of collision data.
  • 15:00 – When the talks end, which may not be on time, there will be an hour long technical discussion between the scientists about each others results. Until these talks the two 3000 strong teams of physicists had not officially compared their data so there will be much to talk about.
  • 15:20 – At this time we expect a release of information and pictures to the press as the scientific discussion continues.
  • 16:30 – Press conference. Questions and Answers from the experts
During these events I will be posting news and exclusive analysis right here as it happens. You can refresh this page for updates and post your own views and observations in the comments section. However, please accept that I may delete comments that I consider unhelpful to a general audience. You can continue to post broader material on the previous post about the rumours
Amongst other things I will be attempting to combine the results in real time as soon as the necessary plots become available. The CERN director General has forewarned us that the announcement today will not provide conclusive evidence for the existence, or non-existence of the Higgs boson, but that could be because the two experiments have not had time to combine their results. The official combination will not be ready until next year because the full computational process is long and difficult. However, it is possible to do a quick approximate “bloggers” combination that will allow us to anticipate what the eventual result will look like. In fact the method has been shown to be reasonably accurate in the past. I will be doing more combinations right here today.
Let me just reiterate that again. My combinations are approximate. They assume a flat normal probability distribution. That is a good approximation that improves as more data is added. They also assume that there are no correlations between uncertainties among the different parts of the experiments. This is not the case. Such correlations have a small effect that does not diminish with more data. In order to claim a discovery using a combination the collaborations will have to get together and do an official version the hard way and that will take time. However, my quick combination method is good enough to give a very good idea of what the final result will look like and it is certainly not “Nonsense” as some of the experimenters have tried to claim.

Why is the Higgs Boson so special?

During the 1960s and 1970s theoretical physicists using data from the first generation of particle accelerators assembled a theory of elementary particles known as the Standard Model. It included familiar particles such as the electron, photon and neutrinos as well as unseen quarks that bind to form protons and neutrons inside the atom. All the particles in the standard model are of two types with one exception.

The particles which build up matter are all spin-half fermions which obey an equation formulated by Dirac in 1928. This includes the three generation of quark pairs and the three corresponding pairs of leptons, the electron, muon and tauon with their neutrino partners. Each of these has an antimatter partner so there are 24 distinct fermions in the Standard Model.  The second set of particles are the spin-one bosons. These play the role of binding together the fermions with the electromagnetic force (the photon) the strong nuclear force (the gluons) and the weak nuclear force (the W and Z bosons) Of these only the W is charged and so has a distinct anti-particle, meaning that there are 5 different bosons.

Aside from these it was found that the standard model required one further particle. It was known that a consistent model of spin-half fermions and spin-one bosons free from infinities required gauge symmetry, that is a mechanism that would in theory make the bosons massless. On the other hand, nature had shown that the bosons that mediate the weak nuclear force must have mass. The solution was a mechanism worked out around 1960 by a number of physicists that introduces an unusual field into the theory. The field has an unorthodox energy potential that is minimised away from the central point of symmetry so that the value of the field in the vacuum state of space-time must be shifted away from the central point, thus breaking the underlying symmetry and giving mass to some of the particles.

Peter Higgs, one of the pioneers of this mechanism, pointed out that the remnant of this field in its broken form would have excitations corresponding to a unique elementary particle that might be observed as final confirmation of the theory. Unlike the other particles in the Standard Model, this one would be a spin-zero boson. Observation of this hypothetical particle named the Higgs Boson in his honour is what the Large Hadron Collider has been looking for 50 years later.

What does the Higgs Boson look like?

The Higgs Boson exists only for fleeting moments as a fuzzy quantum wave on scales smaller than the inner workings on the proton. It is therefore impossible both theoretically and practically to “see” it in the normal sense of the word. What we can see are traces of its existence in data gathered from countless collisions between high energy protons in the Large Hadron Collider.

In the LHC at CERN on the Swiss-Franco border near Geneva, physicists have been accelerating protons to unprecedented high energies in a circular underground ring 27 km in circumference. When the protons are brought together in a head-on collision the energy can form new particles, perhaps including some never observed before such as the Higgs boson. Many trillions of collisions have been observed but the processes that form a Higgs boson are so rare that only a few thousand are likely to have been created in the experiment so far.

Once created a Higgs boson should live for a fleeting 10-22 seconds, enough time for it to travel between 10 and a hundred times the width of the protons from which it emerged. Then it decays, usually into other particles, most often a matched pair of bottom/anti-bottom quarks which have a much longer lifetime of 10-12 seconds. As the bottom quarks fly apart a string of gluon flux stretches between them before breaking to form new quarks. These emerge along with the decay products of the bottom quarks as jets of hadrons that reach the detectors. Sometimes the bottom quarks will each decay into another quark plus a lepton (electron or muon) with an accompanying neutrino. The lepton makes tale-tale tracks in the detector while the neutrino flies off without a trail only to be guessed at when they add up the energy of all the other particles and notice that some is missing.  Unfortunately there are many other less remarkable processes that produce similar jets and leptons at the LHC making it very difficult to observe the Higgs Boson when it decays in this way.

If the latest rumours about the measurements at CERN are correct the Higgs Boson could have a mass approximately equal to that of a Caesium atom. If this is correct about one in 500 of the Higgs bosons produced will decay into two high energy photons that fly away in opposite directions. Unlike the bottom quarks these fly away cleanly carrying all their energy and momentum to the inner layers of the detectors where a surrounding vessel of liquid argon has been placed to capture them. There they produce a shower of lower energy particles that are carefully tracked so that their energy and trajectory can be measured to reveal the parameters of the original photon. During all of this years run at the LHC this may have happened only a dozen times in each detector, but it could be enough to reveal the Higgs Boson.

Such photons will be thousands of times more energetic than the harmful gamma rays that emanate from nuclear reactions, but they are still photons identical to those of light which differ only by having less energy. If you want to know what the Higgs boson looks like it is the faint glow of these rare photons that answers the question most directly. In the LHC they shine faintly among the brighter radiation of other processes that produce equally energetic gamma rays. The ones coming from the Higgs Boson can only be noticed when enough have been detected to show up as a slightly brighter peak in the energy spectrum of thousands of observations. It is this that we are hoping to hear news of today.

A typical event with two high energy photons as recorded in CMS

Will the LHC find the Higgs Boson?

The theory of the Higgs Boson has been around a long time and all the other particles of the standard model have been found. Several of them were found after they were predicted by the model, especially the gluons, W and Z bosons and top quark. This means that the theory of the Standard Model is in very good stead experimentally. Indeed, physicists have been hoping for some experimental deviation from its predictions for decades and have come away disappointed. Every experiment just seems to confirm its correctness with ever more accuracy. (There are some exceptions such as measurement of the muon magnetic anomaly and the cosmological observation of dark matter that seem to point to something beyond the standard model at higher energy)

With such success it is no wonder that the theorists are quite confident that the Higgs Boson will be found as the last missing piece of the Standard Model. However, experiment is the ultimate judge of nature and theorists are not always right. A minority of physicists notably including Stephen Hawking and Nobel laureate Martinus Veltman have said that they do not believe the Higgs Boson will be found because according to their theories it cannot exist. They are considered contrarians by other physicists but until the “Goddamned” particle has been found nobody can be certain.

One thing that is sure is that the Large Hadron Collider will either discover the Higgs Boson or rule it out as predicted by the Standard Model. If all goes well this will be achieved before the end of 2012, perhaps much sooner. It has been said that if the Standard Model Higgs Boson is ruled out it will be an even greater discovery than its mere existence. This is not just excuses for what some people may portray as a failure. Such a result would indeed by a breakthrough inevitably leading to a new and better understanding of physics.

It is also possible that the Higgs Boson exists but that its characteristics are different from those of the Standard Model. In particlular, it may decay into other lighter unknown particles making it hard to detect. In that case it might appear not to be there even though it is. That will still count as ruling out the Standard Model Higgs but until further experiments are done it will not be known whether it does not exist at all, or is merely hidden from view by non-standard processes. Another even more exciting possibility is that there is more than one Higgs Boson possibly including some heavier versions that are charged. This is predicted by some grander theories such as supersymmetry

However, results from the LHC so far suggest that whatever happens there will be something positive to report today. It will not be quite a full discovery but it will be a strong signal that something like a Higgs Boson exists. Although we have heard some quite detailed rumours already, it is only by seeing the actual graphs that we can get a good idea of what the possibilities are. All physicists are now eagerly waiting to see them.

What will we learn?

You might think that since the Higgs Boson was predicted 50 years ago its discovery today will not be very exciting news. Indeed, before the LHC started collecting data, many physicists saw its discovery as inevitable and uninteresting. This view has changed, partly because nothing else has been quick to manifest itself at the LHC as hoped. This means that the Higgs Boson is likely to be the leading discovery of any new physics.

The mass of the Higgs Boson is a free parameter in the Standard Model. Once it is known, all other features such as its lifetime and interaction rates can be calculated. However, analysis of the physics of the Standard Model shows that if the mass is not within strict limits the theory will break down at higher energies. In particular, if it is too light the vacuum will not be sufficiently stable, but we know that this cannot be happening in the real world. The mass range left where the Higgs Boson can still be found includes a range where this would be a problem for the theory.

If it is lighter than 126 GeV then that may be an indication of new physics that could be found with more data. The theory of supersymmetry which is very popular with theorists actually favours the lighter Higgs and corrects problems with the stability of the vacuum, but it does not support well a heavier mass.  For these reasons today’s announcement could signal the directions of research for future physics depending on what mass is indicated by the experiments.

What will we be looking out for today?

Despite the rumours, it is not certain exactly what will be shown today, but we are hoping for full reporting of all the results in the Higgs search from the two individual experiments. This would include the analysis of each possible decay mode that the experiments can currently observe plus two combination of results from all channels, one for ATLAS and one for CMS. The amount of data collected this year corresponds to an integrated luminosity of 5 inverse femtobarns (5/fb) in each experiment so anything less than this is not complete.

There are three sets of decay channels that are currently of special relevance to the search,

  • diphoton (a.k.a. digamma) where the Higgs Boson decays directly to two photons
  • WW -> lvlv where the Higgs Boson decays to two oppositely charged W bosons which then decay to electrons or muons and associated neutrinos
  • ZZ -> 4l where it decays to two neutral Z bosons that then each decay to two oppositely charged electons or muons making four leptons in total.

If recent rumours are correct it is the diphoton channel that holds the most interest with a signal of a possible Higgs Boson at a mass of 125 GeV, but we will be very interested in the other channels to see if there is any supporting evidence or signs of anything at other masses. It will be especially interesting to see of the earlier weak signal at 140 GeV has gone away entirely. These and other channels may provide signs of something interesting at higher masses but most likely there will just be a strengthening of the evidence for exclusion above 140 GeV.

What do the plots mean?

During the presentations delivered by the collaborations today we will see a lot of new graphs. If you are not familiar with these they will require some explanation. The ones that everyone will be looking out for are the “Brazil band” plots, named for their distinctive green and yellow bands. These plots are the main way of showing the results from each Higgs Boson decay mode as well as the all important combinations.

Here is the best LHC combination plot for Higgs boson searches made public prior to today. It incorporates about a third as much data as gathered during the whole year and was shown in November at the Hadron Collider Physics conference, but I have redrawn it to add some extra features. (With any plot on this blog you can click on the image to enlarge for a clearer picture)

The horizontal axis is marked with the range of possible masses for the Higgs Boson. The units are Giga electron-Volts as an energy equivalent of mass. This is the standard way to measure mass in an accelerator experiment. If the Higgs Boson has a mass of 125 GeV as rumoured you should be able to see where it would appear on this plot.

The black line is usually called “Observed CLs” and represents the calculated result from all the experiments. Its value for any given mass gives a quantity labelled “95% Confidence Level limit for σ/σSM” on the vertical axis. What does this mean exactly? Take an example; At 200 GeV the observed CLs has a value of about 0.6. What this says is that if the signal cross-section over all the decay modes were just 0.6 times the amount expected if the Standard Model is correct and the Higgs Boson has a mass of 200 GeV, then there would be a 95% probability of seeing more events than they did.   This is a roundabout way of saying that we have seen far too few events, so we can rule out the Higgs Boson at this mass with some confidence.

When the black line descends below the red horizontal line at 1.0 on the vertical axis, people sometimes say that the Higgs Boson has been ruled out at 95% confidence level at this mass. This is not strictly correct because such confidence would depend on our prior assessment of the probability for the existence of the Higgs Boson in this mass range in the first place, and also the “Look Elsewhere Effect” would have to be considered. Such knowledge is subjective and dependent on outside influences, but loosely thinking you can interpret it that way.

In the background of the plot I have shaded areas in various grades of pink. The lightest pink indicates an exclusions at 95% confidence. This is often stated as 2-sigma significance because statistically it corresponds to 2 standard deviations away from the normal expectation. Darker shades of pink indicate 3-sigma and 4-sigma confidence. Until recently it was generally accepted that 2-sigmas was enough to rule out the presence of the Higgs Boson at a given mass, but recently people have said they want 5-sigma significance, the same as for the discovery of a new particle. I think in reality most people will accept 3-sigma for exclusions.

But we are no longer just interested in exclusions. How do we know from this plot if the Higgs Boson has been seen? This is where the yellow and green bands come in. The central blue line indicates the expected value under the condition that no Higgs Boson exists at a given mass. The green and yellow bands are the 1-sigma and 2-sigma deviations from that expectation. This means that if there is no Higgs Boson the observed CLs line should wander within these two bands. Statistically it is likely to go outside the yellow bands for about 5% of its range. When we look at the plot we see that this is indeed the case. Despite the excess exceeding 2-sigmas around the 140 GeV region we can only say that the result is consistent with the lack of a Higgs Boson over the whole range. That is not a very encouraging way to put it. Notice that mass ranges where there are excesses will be background shaded in grades of green.

Can we at least say that the plot is also consistent with the hypothesis that there is a Higgs Boson somewhere in the mass range? We can see that it is excluded over the range from 140 GeV to 480 GeV at 2-sigma significance but we can still accept the possibility that it is in the low or high mass region. there are theoretical reasons to strongly doubt that it is at the high mass end so the range 115 GeV to 140 GeV is the best bet.

It is possible to display the same results in a different way that handles the existence and exclusion of the Higgs Boson in a more symmetrical way. This is sometimes called the “best fit” plot or “signal”  plot and for the combination above it would look like this.

The experimenters don’t often display their results this way, but as  theorist I find it the best plot to give a feel for where we stand. If I can get the data from the talks today I may show some of these plots.

The black line varies around a range of signal values where a signal of zero would indicate just the Standard Model background with no Higgs Boson and a signal of one is just the right strength for its existence. The blue and cyan bands are error bands (mostly statistical) around the observed data. When the blue and cyan error bands extend over the whole range between the red line at zero and the green line at one we really have no indication either way for a Higgs Boson or its exclusion in the mass range. However, when it starts to settle on one of either the red or green line and moves clear of the other, then we know that we have the right signal strength for the presence or absence of the Higgs Boson.

What will happen after today?

Whatever comes out today there will still be a lot more work to be done. At the moment the LHC is shutdown for the Winter to allow for maintenance and to save electricity at a time when domestic demand is highest. It will startup again in February next year. Meanwhile the physicists will be using the time to continue the analysis of the data already collected during 2011 and that will include preparing the official combination of today’s results from ATLAS and CMS.

Next year the LHC will run again, probably at a slightly higher energy of 8 TeV rather than the 7 TeV used this year. It is expected to collect three times as much data in 2012 as it did in 2011 so by the end of the year they will have a total of at least 20/fb on tape for each of ATLAS and CMS. If they don’t already have enough data to know whether the Higgs Boson exists they almost certainly will by then.

More importantly, they will start to study the properties of the Higgs Boson to check that it matches the standard model by decaying into all types of lighter particle at the predicted rates. If it doesn’t then they will know that there is new physics outside the Standard Model to be understood.

That assumes that the standard Higgs Boson will show up. If it doesn’t they will have the job of looking for what replaces it . That can be done by looking at interactions between W bosons which should get stronger with increasing energy if there is no Higgs Boson until something gives. Present rumours suggest that the Higgs does exist but these WW scattering experiments will still be interesting.

After 2012 the LHC will shutdown for about 18 months to prepare it for running at higher energies, probably 13 TeV during 2015 and 14 TeV later. They will be searching for more new particles but they will also checking the parameters of the Standard Model including the Higgs Boson in more detail to eek out any signs of dark matter or anything else not seen before. The LHC will continue to run at higher luminosity and possibly even higher energy for perhaps another 30 years. This is just the beginning of what it has to do.

Live Blog starts Here

09:00 (times are Central European)

This morning ATLAS have released an update to the Higgs search in the WW -> lvlv channel. They are using 2.05/fb in place of the previous 1.66/fb so it is only a small advance. This had been around for some time unofficially but was not shown at the HCP2011 conference, Hopefully it will be obsolete in a matter of hours but here is the plot anyway. It provides 95% exclusion from 145 GeV to 200 GeV.


Just to remind everyone, the official build-up for this event is as follows:  “These results will be based on the analysis of considerably more data than those presented at the summer conferences, sufficient to make significant progress in the search for the Higgs boson, but not enough to make any conclusive statement on the existence or non-existence of the Higgs.”

If you come here expecting a life-changing discovery to be announced you will be disappointed, but if you want to see some science in action taking a small step forwards you may enjoy.


With two hours to go the auditaurium was already full.


Here in the UK the BBC are already running reports on the network news. They are saying that each experiment is finding a blip in the same place giving a strong hint of the Higgs.


Speakers introduced, talks getting underway


ATLAS have updated the three most sensitive channels diphoton to 4.9/fb ZZ->4l to 4.8/fb and WW->lvlv to 2.1 (as above)


I have the CMS Combo, here it is with exclusion from 130 GeV up. Excess seen at about 123 GeV of 2.5 Sigma


Here is the CMS diphoton plot shwoing where the excess comes from, but there are other excesses nearly as big


Here is the ATLAS version from the talk. Updated from conference notes.


The CMS ZZ->4l clearly rules out the 140 GeV possibility, but has an excess at lower mass.


ATLAS ZZ->4l and full combo from the talk. Updated from conference notes.

ATLAS full combo from the talk. Updated from conference notes.


First talk is over, now over to CMS

CMS have two versions of the WW channel, cutbased and BDT


Here is the first of my unofficial combinations as the discussion time ends. This is the diphoton channels combined for ATLAS+CMS. Remember that this is approximate and you should not try to read the number of sigmas from this. I may revise it later when better version of the plots become available.


ATLAS have now released 3 new conference notes so I will update the pixtures


I have now digitised the CMS combined plot and produced this signal plot. It gives a clean indication for no Higgs about 130 GeV and the right size signal for a Higgs at about 125 GeV, but there is still noise at lower mass so chance that it could be moved.


Here is the same thing for the ATLAS data


Here is the fully combined exclusion plot. The signal fits best at 124 GeV and just makes 3-sigma. Remember the official version is likely to be a little different. This is just a quick approximation.


Here is the fully combined signal plot. It looks very convincing but the region below 120 GeV is not resolved yet. Until it is there will be a little room for doubt.


But of course we can clean up the lower region by including LEP and Tevatron too. An official combination with Tevatron data included is also planned

A zoomed version


Finally here is one last combination for diphoton + ZZ in CMS and ATLAS. These are the high-resolution channels so they give a cleaner signal, but without WW the significance is less.

Conclusions: The result is very convincing if you start from the assumption that there should be a Higgs Boson somewhere in the range. Everywhere is ruled out except 115 GeV to 130 GeV and within that window there is a signal with the right strength at around 125 GeV with 3 sigma significance. They will have to wait for that to reach 5 sigma to claim discovery and next years data should be enough to get there or almost. I calculate that they will need 25/fb per experiment at 7 TeV to make the discovery. A big congratulations to everyone from the LHC, ATLAS and CMS who found the Higgs when it hid in the hardest place.

I was lucky enough to meet Peter Higgs many years ago when I was a postdoc at Edinburgh and I have a big smile knowing that this has been achieved in his lifetime. Congratulations to him and the other physicists involved in discovering the mechanism of symmetry breaking. Finally, in case they are forgotten, well done also to all the phenomenologists who did the calculations to work out how the Higgs Boson could be found, not least John Ellis.

From here there is much more work to do in order to check that this particle seen today has exactly the characteristics of the Higgs, if indeed it is confirmed with more data. That will take many more years of runs at the LHC. It will also be exciting to see how this mass affects our understanding of what other physics could be in reach. I hope there are some Campaign corks popping at CERN this evening. They have had a remarkable year.

130 Responses to Higgs Boson Live Blog: Analysis of the CERN announcement

  1. mfrasca says:

    Hi Phil,

    Thank you for your very helpful blogging. Please, note that the spokeperson of ATLAS is Fabiola Gianotti and not Fabio. She is a woman.



  2. Nessuno says:

    Hi Philip,
    will be this another “quasi-discovery” seminar, in which CERN seems having specialized in the last years?

    • Philip Gibbs says:

      I’m not sure what you mean by quasi-discovery, examples?

      This is billed as a report of significant but inconclusive progress. I dont know if it will merit the media circus but CERN has tried to play it down so they can’t be blamed. I am excited to see the latest data whatever it shows. You may not be.

      • Nessuno says:

        Dear Philip,
        of course it is interesting to look at the latest data. But what I find wrong is this over-excitation in discussing for days if a signal is 2.4 or 2.6 sigma (i.e. talking about a 0.2 sigma). 4 or 5 hundred persons occupied the auditorium several hours before the seminar; luckily the DG had pre-announced “inconclusive” results, otherwise those people would have bivouacked there for days! Another point: I may be wrong, but the signal with more sigma seems to be based on three “golden” events; this is exactly the same scenario we had at the end of Lep, with the Aleph and L3 “fundons” (thank you Jester) around 114 GeV. If we excluded one of the three events, we would go down to an almost invisible signal, like in the other channels. Finally, forgive my ignorance, but could you point me to a good explanation of the “Look Elsewhere” effect? How pertinent is this in this case? Because its effect, if I understood correctly, is to remove any significance to the level of confidence of the signal “observed”, at least until more statistics ais collected and more events are found.
        Best regards.

      • Philip Gibbs says:

        Even just three events can be significant when the background is low, and this is just one channel in one detector. The main part of the signal comes from a different channel. Overall there are three channels in two detectors that show a signal at the same place. No one of these on its own would be significant but they add up to a clean three sigma signal. The fact that they come from different sources reduces the chance that there is a systematic error.

        The look elsewhere effect has to be taken into account but there is only a small mass range of about 25 GeV where the Higgs is not strongly excluded. Given that the width of the signal is about 5 GeV the look elsewhere effect is only a factor of about five. I think it is being overstated.

        This is basically a half-full/half-empty result. You can state it optimistically or pessimistically according to your political requirements. Another twelve months will be needed to settle it, but it is much more probable that it will be settled with a positive outcome.

        The excitement of the physicists who have waited so long and worked so hard for this is understandable I think

  3. Karin says:

    You have an excellent blog. It is very helpful. Thanks for that! But please take away the moving white dots. They are very annoying and disturbs the concentration while reading. This stuff is hard enough without this kind of challenges for concentration.

  4. Luboš Motl says:

    Dear Phil, won’t you join this chat box


    during the webcast (or before it)? (Via motls.blogspot.com)

  5. Simplicity says:

    Thanks for this great live “broadcast”. There is a storm outside my door, but sunny and warm indoors. This is exciting.

  6. Sunny says:

    So the stream isn’t working so well from CERN this morning. Too bad. But we’ll be giving a live interactive chat with the General Public at 12:30 EST at Perimeter Institute for Theoretical Physics so if you have any questions, tweet to us at #piLIVE. Our physicists who have been working on the experiments will be happy to answer questions.

  7. AnonymousCoward says:

    can we have a ATLAS+CMS combo before the end of the CMS talk?

  8. chris says:

    If you do one of your famous combinations, it would be nice to have it down to 100GeV. To me it looks like Atlas has a strong gamma-gamma signal there, too.

  9. Great concept for a blog post, thanks.

  10. Luboš Motl says:

    Dear Phil, good work! I hope you will eventually gift us a ZZ ATLAS+CMS combination, and ZZ+gammagamma CMS+ATLAS quadruple combination. 😉

    Because of the mismatch of the center of the peak, adding CMS to ATLAS in diphotons doesn’t seem to improve the peak much. However, ATLAS seems to agree in the ZZ channel that the Higgs is below 125 GeV, so the ATLAS digamma is the only source that prefers above-125 GeV values.

  11. Tony H says:

    Just to be clear.. your CMS certified unofficial combo “signal” plot starts to show a signal at the low 120s GeV because the yellow line goes to one there AND the cyan band starts to be centered around the green line?

  12. Kea Parody says:

    When will you people finally get it through your heads, there are no fairy fields! You can show all of the fancy data and plots you want, but none of you no better than me, the maverick supergenius of the age!

  13. DC Whitworth says:

    Thanks for the really interesting posts. Can anyone clarify to a relative layman why the acceptable percentage certainties for finding and excluding the Higgs are so different ? i.e. 95% exclusion seems to be cheerfully acceptable without demur yet a much higher percentage is required for discovery.

    • Nigel says:

      Because the number of “interesting” or “candidate” events is small, but it’s guaranteed that some noise will occur several sigma or standard deviations beyond what is “expected”. There is an expected fluctuation around what is expected. If you throw a coin 10 times and get 10 heads, that does not prove that both sides of the coin are heads. It may make it seem “likely”, but that’s your prejudice. It may just be random luck.

      All these interactions are described statistically in terms of cross-sections, which are directly proportional to probability. If you collide particles and see two gamma rays going in opposite directions, that “could” be a Higgs or it could just be a random fluke. There is no solid signal that is definitely going to a Higgs, it’s thus a statistics game. You look for two gamma rays or four leptons coming off. When you find this, is it a Higgs, or just a random pattern due to your own prejudice, like seeing the face of Jesus in a tomato?

      • DC Whitworth says:

        OK I get that, but why is 5 tails in a row = Higgs Excluded, but 5 heads in a row only = Hint of a Higgs ?

      • Leo Vuyk says:

        However, something interesting is happening there right?
        If two protons collide, then we may guess only two quarks will collide.
        If we think of these particles as real knotty complex things, able to unwind and regroup to different particles then no Higgs is needed at all.

      • anna v says:

        DC Whitworth :

        “OK I get that, but why is 5 tails in a row = Higgs Excluded, but 5 heads in a row only = Hint of a Higgs ?”

        Please note “Higgs excluded at 95% confidence level”. The reason we accept 95% for exclusion but require 0.000x% for verification is because :
        a) “extraordinary claims need extra ordinary evidence”: It is not extraordinary for the background to have no signal. A claim of signal though is by definition extraordinary.
        b) experience has shown that often there exist in experiments insidious systematic errors that have not been taken into account and even 4sigma statistically peaks have become insignificant ( I know of a mu pi resonance in neutrino physics long ago) once the error is computed correctly. This sort of errors goes in the direction of extending the insignificance regions of the plots so are not guarded against.

  14. JollyJoker says:

    The three sigma total was completely unexpected. Apparently Atlas was fairly weak where CMS had its peak (124?) and CMS really dragged down Atlas’ peak at 126.

  15. JollyJoker says:

    Btw, I really like your red/green background color scheme for exclusion of SM Higgs / background only. Although I’d rather hope for the data to make it break down 😉

    I guess shades of yellow or brown would work.

  16. Luboš Motl says:

    Your tricollider Yes/No chart is just cute! The 125 GeV “Yes” Higgs over there is really sharp and everything else up to 500 GeV is noise around “No”.

    • Philip Gibbs says:

      I did not expect it to come out so clean, especially with the WW included. I should still do the combination of high res channels to see what it looks like.

    • Tony H says:

      Yes agreed. I’m thinking of printing it out and putting it on the corridor noticeboard. You happy for that sort of exposure Phil?

    • Philip Gibbs says:

      Just make sure nobody sees you putting it up 🙂

    • Luboš Motl says:

      I think it’s OK to post it. You must just add a disclaimer:

      “Kids, don’t try to cook such combos at home. You could end up outdoing CERN.” 🙂

  17. SD says:

    Amazing job, congratulations. Of course much bigger congratulations to the experimental collaborations, but it’s really nice and useful to see this approximate combinations delivered in such a timely manner. Thank you very much!

  18. Kea says:

    Fairies really exist! Wow!

    I’m actually really happy to KNOW at last. Well, almost, anyway. Congratulations and thanks for all the amazing work.

  19. PSTJ Editor says:

    Thanks for keeping us informed! Beautiful job!

  20. JollyJoker says:

    Thanks from me as well. I waited for the combination with more interest than the official releases 🙂

  21. Ulla says:

    Ye, Kea, go to a library and look. Plenty of them 🙂 Ancient ones are called myths.

    Look at Mitchells picture

    How much does todays picture mean?

    Of course i don’t know anything.

  22. That combined plot at 18:11 should be the headline for the media folk

  23. Paul Hoiland says:

    Might remend everyone of some already recorded words on the SM: “The success of the Standard Model (SM) offers very few experimental clues how it may break
    down, and at what scale. One clue is provided by the discovery of neutrino masses, which
    suggest the appearance of new physics at a mass scale of a TeV or more, probably at least
    1010 GeV in the simplest versions of seesaw models. Another clue might be offered by the
    measurement of the anomalous magnetic moment of the muon, if one could be sure of the
    value within the SM.

    The baryon asymmetry
    of the Universe can also be explained only by physics beyond the SM, which could appear
    anywhere between the electroweak and inflation scales.” FromThe Probable Fate of the Standard Model J. Ellisa, J.R. Espinosaa,b, G.F. Giudicea, A. Hoeckera and A. Riottoa

    • Paul Hoiland says:

      Anyway you go the story on the SM is not over. The light Higg’s mass is already known to effect vacuum stability. So get ready for some sort of extension to the SM out of all this.

  24. Luboš Motl says:

    Dear Phil, you broke the link under “Skip to live updates” because you changed your date. The new URL of this post has date 12/13 in it but the “Skip to live updates” points to 12/10.

  25. Paul Hoiland says:

    Requiring that the SM cannot develop a minimum deeper than the electroweak vacuum for any scale the lower bound is 128.6 GeV. The range here is below that point whcih in general favors Finite-T metastability. That in itself translates to physics beyond the SM. So in some sence the Faries as far as extension beyond the SM has come out of the closet. This is where further and deeper looks at the Higg’s will tell us what exactly is out there beyond the SM. So even if the Faries are out we do not know the nature of these faries yet.

  26. Vladimir Kalitvianski says:

    Philip, how about the angular momentum in gamma-gamma channel? Does it correspond to the spin-0 boson? What does experiment say?

    • Philip Gibbs says:

      photons have spin states +1 or -1 so anything that decays to two photons must have spin 0 or spin 2. If they can get a positive signal in one of the fermion channels such as bb or tautau that will tell them it is spin 0 or spin 1, so it would leave spin 0 as the only consistent option. It will be a while before they get there.

  27. someone says:

    Thanks for doing this! Can you explain the meaning of the curves and the color coding in your combination plots?

  28. ottonieri says:

    Thanks a lot. Your plots are highly helpful for the “quasi-layman”, and I am much more excited now than by reading the summaries of Cern announcement. It really looks like it’s hard to expect from more statistics anything else than a confirmation.

  29. EDR says:

    As many have said, this is great work. I am, however somewhat unsettled by an aspect of the analyses that you’re building on.

    Formulated as a question:

    When does a broad deviation from an expected background constitute substantial evidence that the actual background differs from what was expected?

    In your LHC + Tevatron + LEPS plot, for example, the background is more than one sigma above than expected value across (very nearly) the entire interval between 135 and 170 GeV. (Note that this omits the region where your plot suggests a substantial Higgs contribution; values in the region immediately below are also high.)

    If an excess were found everywhere, I’d seriously question the background calculation. As it is, the width of the region is large enough to seem troublesome.


    One approach to producing an empirically adjusted estimate of the background would work along the following lines:

    Calculate an expected background, then calculate, for each energy, a function that fits (in some tasteful way) all of the data that would not be affected by a (hypothetical) peak at that point. The adjusted estimate is then a curve in which the value at each point is the value of a global fit to the total data set minus the data from a range around that same point.

    There are surely more-or-less standard methods that implement this principle. Applied here, they would yield a notable, but substantially smaller signal.

    • EDR says:

      (Basically, the result would look like what you’d get by ignoring the putative peak region, drawing (by eye) a smooth curve through everything else, and then measuring the peak height relative to the curve.)

    • Philip Gibbs says:

      The signal as seen is what should be expected given the energy resolution of the various channels. The broader excess can be removed by taking out the WW channel as I did in the last plot. The remaining width of the signal corresponds to the resolution of the diphoton and ZZ-> 4l channels.

  30. Ulla says:

    Kea talks of triplets XXX⊕YYY⊕ZZZ in massivation. Also there may be many Higgs? So this is just one possibility?

    Kea says: It is heartening to see the low mass range gradually take shape in the 0 to 1 SM band,

    We simply do not yet know what the data are saying. There is an excess running across a wide range. What does it mean?

    Higgs slowly eaten? Not at one but at different bandwiths or channels? How come they to think there would only be one way?

    There also can be oscillations in this process? What forces are involved/are created? Spin? L-shell, parity???? I guess they tells a lot? The first, strongest, is the force holding quarks together? So this process goes from strongest to weakest (decay?)

    But Higgs is no decay.

    • Philip Gibbs says:

      At this point we can only say that there is a signal of something that is consitent with the properties of the Higgs Boson. It has spin 0 or 2. It has even parity. These follow from the decay to two phiotons or two Z or two W. The cross section in each channel where the signal can be seen is consistent with the Higgs cross-section but error bars are still large. The broad excess is understood as poor mass resolution in the WW channel and is not seen in diphoton or ZZ->4l.

      There is still room for non-standard models but already the picture looks good as a SM Higgs. There is some room for a second signal at about 120 GeV but no real evidence to favour it. Any other signal would have to be above 500 GeV

      • Ulla says:

        So, how is the Planck length as hbar-length created in the Higgs spin-0 boson with a parity?

        How long must we wait? Until the money is secured for LHC/Atlas?

  31. JollyJoker says:

    “I calculate that they will need 25/fb per experiment at 7 TeV to make the discovery.”

    Why on earth so much? The original assumptions were that 17/fb would be enough for a 115 GeV discovery and we already have more than 10 combined.

    Even if you assume all those 10/fb are in use in the current plots and calculate from 3 sigma to 5, shouldn’t that only need 18/fb total more, or 9/fb/experiment?

    • Philip Gibbs says:

      In normal circumstances the significance goes up with the square root of the amount of data. But in these plots they read from a logarithmic scale and progress is slower. They may be able to interpret it differently, we shall see.

  32. Kea says:

    OK, I’m not as convinced now as I was this morning, even though I can see it is probably real. According to Strassler, Gianotti was talking only a 2 sigma ATLAS signal when the ‘look elsewhere’ effect was taken seriously. Combine that with the ATLAS/CMS mismatch and we go down under 3 sigma. Could go either way.

    • Philip Gibbs says:

      Some caution is merited. The signal is only 3 sigma combined and the possibility of systematic contributions is there. However, look elsewhere effect is very small given that most regions are strongly excluded. Systematic effects look less likely because of consistency across channels. I agree with Dorigo’s more optimistic assessment but until they have 5 sigma it is not a discovery and collapse of the signal is not out of the question.

      The DG has told everyone to express caution and to doubt any unofficial combinations. They are all very loyal and are dutifully complying, except Dorigo of course.

      • D R Lunsford says:

        1) Any examples from history of signals like this disappearing?

        2) Any room for a spin-0 even parity particle that is NOT the SM Higgs?

        Thanks in advance


      • mfrasca says:

        Dorigo has a bet on. Less supersymmetry evidence less danger to pay in a near future.

      • Luboš Motl says:

        OK, I, for one, disagree with this reticence.


        The actual significance of the signal near 125 GeV is about 4 sigma from all known experiments, and this does include look-elsewhere punishment. One must be careful not to include the look-elsewhere punishments multiply times. If the loci of the signal agree, and they arguably agree within the error margin (although there are discussions about whether CMS is able to nearly exclude the best ATLAS figure of 126 GeV: but not quite haha), only one of the locations of the signal – for one experiment – is variable and produces a look-elsewhere effect. The other one is already confirming a specific hypothesis which doesn’t allow us to “look elsewhere” and no punishment should be counted.

        For me, the question is whether it’s the SM Higgs boson or whether it deviates and whether there’s something else as well. Betting that all this business around 125 GeV will go away completely is just completely bold.

        One must carefully distinguish the claim “we can be really sure and bet our lives” from the claim “the existence of the Higgs near 125 GeV is more likely than the absence”. The former is at least technically untrue at this point but the evidence suggests that the latter statement is true. It’s “extremely likely” that the/a Higgs is there.

      • JollyJoker says:

        Lubos, you seem not to take into account that the peaks are on slightly different masses? The effect of that is very noticable in Phil’s combination.

        On the other hand, one could try to assign a bonus value for having roughly (well, almost excactly) the expected cross section. I mean, all sigma counts I’ve seen only take into account the odds of getting at least as large a deviation by chance, without subtracting the chance of getting a larger one.

        In addition, if there’s an SM HIggs and this bump is pure noise, random chance has also hidden the real one. There’s a negative look-elsewhere effect.

  33. BJ says:

    We should start a campaign for Peter to be knighted this summer.

  34. wl59 says:

    “1) Any examples from history of signals like this disappearing?” Look only how fast disappeared the ‘signals’ around 140 ! My interpretation of the new data published yesterday by CERN I gave already in the previous thread, or more detailed here: http://netphysik.de/forum/index.php?page=Thread&postID=46075#post46075 . I think a health interpretation of these data like of experimental data in general, means there simply is no Higgs, at least no significant indication. But it should be verified very carefully.

    • Luboš Motl says:

      Dear wl59, there have been many 3-sigma signals in “randomly spotted anomalies” of a 6368th most important hypothetical particle that went away.

      But as far as I can say, there has never been a 4-sigma overall signal in the search for the 1st most eagerly awaited particle in particle physics.

      The examples that went away were things like “pentaquarks” looked for in a particular way (among many other ways), and so on, things that should be punished by extra “interdiscipline” look-elsewhere correction because these possible signals were not a priori expected and were random and ad hoc.

      However, there is only one lightest Higgs boson, and it is the holy grail of experimental particle physics in 2011. And the signal isn’t 3 sigma but, in total, 4 sigma. It’s just a different situation. One must be careful to appreciate that claimed 3-sigma anomalies from random places often/usually go away. But one must be equally careful about appreciating that this isn’t a single 3-sigma signal and it isn’t a signal in a random unexpected locus for a potential excess, either. It’s the search for the God particle, stupid, one privileged combined experiment, and the no-Higgs hypothesis is excluded at 4 sigma near 125 GeV.

      The 140 GeV bumps never came anywhere close to this level of significance. Phil has been obsessed by the 140 GeV path and I’ve been telling him for months that it was dead. Now, when he combined his excellent new graphs, I think he will surely agree that there’s nothing interesting happening specifically around 140 GeV.


      • anna v says:

        One should also read the CERN courier 1982? article on the discovery of the W and Z:


        “On 4 May, when analysing the collisions recorded in the UA1 detector a few days earlier, on 30 April, the characteristic signal of two opposite high energy tracks was seen. Herwig Schopper reported the event at the Science for Peace meeting in San Remo on 5 May. However the event was not a clean example of a particle-antiparticle pair and it was only after three more events had turned up in the course of the month that CERN ‘went public’, announcing the discovery of the Z to the Press on 1 June 1983. Again the mass (near 90 GeV) looked bang in line with theory. Just after the run, Pierre Darriulat was able to announce in July that UA2 had also seen at least four good Z decays”

        No brazil plots.

        and remind everybody that the square root of four is two, and statistically both experiments combined did not give 3 sigma for the discovery of the Z.

        It was a) the two independent experiments and b) agreement with the minimal SM theory that substantiated the claim then and it is similar now: two experiments and theoretical projections that we depend on to say that the glass is half full.

      • Philip Gibbs says:

        Yes I agree the 140 GeV signal has gone away as I always said it might. It was the best signal in the earlier data because the error bands were tighter there. In fact the 119 GeV signal that you and Dorigo liked has faded too. It is now just a small blip to the left of the main signal. It would only require a small fluctuation for it to come back but I don’t expect it will. 125 GeV is sufficiently far away to be a new signal where previously there was just a weak excess.

        I am glad that the 140 GeV did not survive because it would have been a disappointing mass for BSM theory.

        Thanks for your excellent assessment on your blog and of course I can confirm that the combination uses the correct weighting, otherwise it would not agree with the official combinations as well as it has in the past.

      • Luboš Motl says:

        Dear Anna, well, caution at that point was still justifiable. After all, Rubbia later “discovered” supersymmetry and top quarks 25+ years before they’re discovered, too. 😉 At any rate, you mentioned an example of a 3-sigma bump that did *not* go away. One shouldn’t forget that those exist, too. And their survivor frequency is orders of magnitude higher if it’s 4-sigma bumps. 🙂

        Phil, don’t get me wrong, I surely do think that 119 GeV is most likely also eliminated in favor of 125+-2 GeV Higgs. Still, the situation remains more inconclusive when it comes to the 119 GeV questions. There could be two Higgses there etc. (I find it unlikely both because of theoretical expectations and the detailed shape of the LHC data.) Needless to say, this asymmetry of conclusiveness between 119 and 144 GeV isn’t due to the particular observed data. It could have been expected a priori as it follows from the theory.

        The 144 GeV Higgs should have brought much faster positive or negative evidence and given this expected “speed of proof”, the proof at the given stage just wasn’t great, and 144 GeV pure-SM got eliminated many months ago, I think. The 119 GeV Higgs Standard Model was still alive – free from 95% CL exclusions – although it’s arguably no longer alive today.

        At any rate, I am confident that the degree of risk that this 125 GeV alert is fake is orders of magnitude lower than it would have been with the previous bumps. The confidence is just at a totally different level. About four collider/channel combinations really agree about the existence of excess and its rough place withis a 3-GeV window.


  35. I do not see the results as a proof for the existence of Higgs. Also Matt Strassler seems to share my view.

    Probably there exists a new particle perhaps several of them: there are bumps around 140 GeV region and at both sides of 300 GeV. If these bumps are real, the standard model Higgs and maybe also standard SUSY Higgses are in difficulties.

    My own proposal about scaled up variant of hadron physics is consistent with the findings too and explains Higgs like signatures in terms of the mesons of this physics. Only a detailed study of Higgs signatures will show whether either of the options survives. For details see my blog.

    See also this and this.

  36. Ervin Goldfain says:


    Here is a crazy thought: what if the 124 GeV bump in the di-photon channel is actually a spin-2 resonance? Could it be the long-thought graviton that is seen here and not the SM Higgs?



    • Luboš Motl says:

      Dear Ervin, ordinary graviton is massless (M=0) – that’s because the force it mediates, force called gravity, has an infinite range and drops as a power law.

      Models with extra dimensions predict “new modes of graviton” that are massive. But their coupling to matter such as photons is much weaker than the coupling of the Higgs boson to photons.

      At any rate, the signatures for extra-dimensional ADD/RS gravitons are different and they have been looked for separately (in other papers from CERN) and could have been excluded pretty much up to 1 TeV or so.

      • Ervin Goldfain says:

        Yes Lubos, I know that gravitons are massless. And this is why I called this idea “crazy”. But what if our understanding of GR is incomplete (GR is a low-energy framework after all) and there is a massive spin-2 resonance showing up in the TeV sector that does not come from extra-dimensions?

  37. Guybrush says:

    Great great job Phil! Thanks for that!

    But one question remains: What happened to the peak at about 100GeV in the ATLAS gamma-gamma plot? I see you excluded the region m_H<114GeV in your plots, but is this legitime? I mean, this obvious "fake"-peak in the ATLAS gamma-gamma plot at 100GeV could be a hint for underestimated systematics, which could be then compared to the peak at 126 GeV.

  38. Guybrush says:

    Oh, yes of course. Here it is:

    At 102GeV there is an upward fluctuation in one bin. Ok, at 126GeV there are two promising bins, but nevertheless, it would have been interesting to compare this fluctuation at 102GeV with the “maybe-signal” at 126GeV.

    • Luboš Motl says:

      Dear Guybrush, there seems to be a 2-sigma bump near 102 GeV which is smaller. But moreover, a 2-sigma bump doesn’t mean that it’s actually enough to account for a 102 GeV Higgs decaying to 2 photons. The 102 GeV SM-like Higgs is excluded by many empirical facts, and the (absolute) smallness of the excess of the diphoton branching ratio is one of them.

      Even if the exclusion of a 102 GeV Higgs by LEP were the “only” counter-argument that makes this point weaker than 125 GeV, it would still be a very powerful one. 😉

      • Guybrush says:

        Dear Lubos,

        I did not declare that there may be a 102GeV Higgs…. I never would say such things 🙂

        My idea is the following:
        At 102GeV we have a fluctuation in the diphoton channel, which is definitely not a real physics signal (LEP excluded etc..).
        While at 126GeV we have a compareable bump, which may be real phyics. This 126GeV bump is furthermore supported by the three events in the ZZ-channel in this mass-region.
        I think this constellation would be a good opportunity to compare how trustworthy the given pure statistical errors are, by directly comparing a fake-bump with a maybe-real one of a similar size.

        Therefore I just wondered, how the here cut-off fake-bump at 102GeV would look like in the combined higgs-plot, compared to the 126GeV-bump. Maybe we could then learn something about underestimating errors.

  39. confused says:

    Sorry. I have a hard time to interpret the graphs here. Could someone explain it for me? It is the first graph in the “what do the plots mean?” section. According to this blog, “1” in the vertical axis means “the amount expected if the Standard Model is correct and the Higgs Boson has the given mass in horizontal axis.” And now the central blue line in the yellow and green bands indicates “the expected value under the condition that no Higgs Boson exists at a given mass”. How this two lines can meet at 125Gev and 520Gev? This alone makes me not to proceeed further. Am I missing something?

  40. wl59 says:

    The doublegamma of CMS shows a signal of about 0.5 for a whole interval 113 till 126, but just there it drops deeply down, and exclude where ATLAS goes high; vice versa ATLAS shows for a part of the quoted CMS range even a negative value (fig. 4 and 7 in the CMS / ATLAS combination). Thus, each of these experiments exclude positively a part of the interval where the other affirms it, and thus makes very questionalbe these signals in its whole. Without more calculations, this means, that the results are not conclusive. Perhaps there is something – but it don’t looks like a Higgs.

    Although less sensitive, H –> WW should show something like a peak (even if less mass-resolution-sensible), but dont’ show nothing. H –> ZZ has a peak at 124, on the other hand it’s beside of a deficience at 123, which ‘cancels’ it. Such a peak at the side of a valey we have in many examples in physics, and this means simple a little energy shift, by any effect, but not a net additional intensity – or with other words, there aren’t produced no additional particles (as we would expect in case of a Higgs), but just the background particles of a certain energy are a little shifted in energy or in something else. What, in contrary, one would expect, is something like a peak agreeing in the different experiments (even a weak peak in the less sensible experiments) with an effective additional number of events.

    So, consisering all experiments and dates which I see in the last decades, but also that all previous ‘signals’ disappeared, I think there is simply no Higgs.

    • Luboš Motl says:

      Dear wl59, there’s no law that both experiments will report exactly the same results. And they don’t. Chance influences their differences, too.

      If you wanted to show a real discrepancy, you would have to show a statistically significant one, too. You haven’t done so. You’re deliberately sloppy about your statements. But there’s no contradiction. CMS doesn’t even exclude 126 GeV at 95% confidence level. The cross section for CMS goes sharply down if you go from 125 to 129 GeV or so but 126 GeV would still survive.

      Moreover, the most likely ATLAS value of the mass is really 125 GeV because the ZZ channel also agrees that the mass is likely below 126 GeV, only ATLAS diphoton would prefer 126 GeV but that may easily be a fluke.

      Also, some of these differences may be not just flukes but systematic differences in energies measured by CMS and ATLAS. It’s not impossible that energy reported as 126 GeV by ATLAS is really the same thing as 125 GeV reported by CMS.

      Phil didn’t assume any adjustments of the energies CMS vs ATLAS whatsoever, and he still got the powerful graph where the bump at 125 GeV just jumps up and returns down. So I think that your sloppy comments about a contradiction – which doesn’t exist – combined with an equally sloppy implication that the Higgs doesn’t exist – which wouldn’t follow from the first, even if the first thing were right – combine to an example of totally irrational reasoning.

      • wl59 says:

        Pls check, using earlier results, how many mass-values one could defend with your argumentation … inclusive f.ex. at 119 and 140 few months ago we could. And observe that the highest part of ATLAS, by CMS is ‘almost’ ruled out – much more anyway as it should be if it would real.

        Fig. 4 of HIG-11-032-pas.pdf of CMS gives the ‘date’ (not the ‘weight’/exclusion probability) for the most part of 117-125 as 0,8 , for 126 0,5 , for 127 0,2 . ATLAS-CONF-2011-163 fig 7 has for 125-7 about 1.5 and for 128 still 0,9, what at least for 127 and 128 is proved significantly as wrong by CMS. Thus, it’s questionable the value of this peak of ATLAS in it’s whole. On the other hand, ATLAS makes 120-122 less significant, although not ‘exclude’ it , in not so sharp but enough difference to CMS. So I think, in the mildest way one can say that there is ‘nothing significant’.

        It’s right that pure formally, just at 125-6 where these two rather different results change, there ‘can be something’, and it’s really not impossible that Higgs hides just there, but considering all data/experimental results what I see in my live, I would judge this as completely insignificant. If there would be something, we’ld have there something clearer, better agreeing, not just a point where one curve goes up and the other down.

        Moreover, just to produce two photons, there may exist plenty other sources. But on the other hand one can argue, if there is no net excess, then also couldn’t be there any Higgs what would produce such an excess. Pls look at fig 3 of the ATLAS paper. The integral under the ‘observed’ diphoton plot (or the total number of higgs-independent events and photons corresponding to that interval) is the same as under the ‘expected’, so that there may be any effects which change slightly the energy of the background photons and what results in the peaks and valleys, but there is no net excess of photons, i.e. no additional photons are produced there by Higgs

        If we inspect fig. 3 of ATLAS and fig. 3 etc. of CMS, then neither H –> WW nor H –> ZZ nor H –> ditau has any indication of a peak at 123-6 . They aren’t so sensible like diphoton, but anything should be indicated there, even if weak !! Thus, what enters in the combined formal results, is most diphoton and the general excess of WW from 100-180 (thus, any model error) — this general WW excess one should perhaps let apart for any combination. For ZZ there is, in opposite to the predictions, a peak near 124 in ATLAS and 119 or so in CMS, but at the side of a valley, again integrated both canceling or no net production; thus this also seems to be any effect which changes slightly the energy of the background there.

        Thus, there’s nothing significant !!

        The total result of the CERN results is, that Higgs is better excluded as before; outside 115-130 already by the first look, and within this interval as explained, even if formally in the range 123-6 it may be ‘less improbable’ than in others.

      • Luboš Motl says:

        “…is almost ruled out…”

        This is a very bizarre term, analogous to “almost pregnant” or something like that. You know very well that it means “not ruled out” but you try to pretend that it means exactly the opposite. Rational reasoning is just an inaccessible goal if you can’t distinguish “ruled out” and “not ruled out” and you even deliberately engineer nonsensical adjectives that help you confuse P and non P.

        The more accurately the Higgs mass – or any mass – is known, the closer values to the right value may be ruled out. But ruling out a nearby value doesn’t mean ruling out the original value. The 125 GeV Higgs boson is not ruled out by any experiment.

        “…but considering all data/experimental results what I see in my live, I would judge this as completely insignificant.” – Well, in that case, Spring 2012 data will show that all your life was one giant mistake.

    • anna v says:

      To set the experimental record straight as far as the energy scales between the two experiments goes, have a look at the slides presented. http://indico.cern.ch/conferenceDisplay.py?confId=164890 .

      in Atlas slide 21, it is clearly stated that the mass resolution is 14%. 14% of 125 is 17.5GeV ! in slide 33 of CMS something smaller is shown for the Z and W peaks, FWHM less than 10GeV..

      The callibration error for Atlas ,which might systematically shift the x scale, is given as 6% at the same slide. Suppose that both gamma were at the pt of 62.5, that might give a shift of 1.4 Gev. In slide 33 CMS is claiming 0.99+/-.01 GeV for instrumental contribution to the mass resolution.

      The work of the group that will combine the two experiments involves working within these constraints and resolving discrepancies.

      • Philip Gibbs says:

        Slide 11 of the CMS talk gives a good indication of the overall mass resolution in each channel. (This would differ a bit between detectors of course). diphoton is 1%-2%, ZZ -> 4l is 1%-3%. The others are all much bigger.

        These are all factored into the individual channel plots so when I do my combinations these uncertainties are included. I don’t have to do anything more to account for them. The combination group go back some steps to do the full combination which is why it is a much bigger job.

      • anna v says:

        On second look I think ATLAS is calling resolution the full width at half maximum, which is not the same as the error on the mass value. With enough statistics one can have good mass accuracy even with large FWHM.

      • Luboš Motl says:

        Great that you found the right interpretation, Anna. 17.5 GeV error in the diphoton channel would really devalue almost all information from that channel about the mass.

      • H says:

        This is ridiculous! 14% is the uncertainty of the mass resolution, not the resolution itself. The resolution is ~ 1.7GeV and the FWHM is ~ 4.0GeV.

      • anna v says:


        Good to have more reasonable numbers. This means that on slide 21 where it says: “main systematic uncertainties” it should say “main systematic uncertainties on the value o :f”. A comment like that under a mass plot implies uncertainties on the mass.

        It would be good to have a link for the more reasonable numbers you are stating, which also defuse the worry of the difference in absolute mass scale between the two experiments.

  41. Makes sense Albert Z,

    But I am already taking a yoga-breath (hey poetry helps when we get knots in our stomach over the number crunching.

    The standard theory has many more problems than those discussed here, the nature of such a particle if it is there- the idea of something unseen happening beneath the wave length of the photon so is unseen, the need or not for faerie particles with the weights of sparticles implied. (but why would this be so if only within the standard theory?) What is theory anyway but some depth of abstract thinking and cognition?

    and motil

    “So I think that your sloppy comments about a contradiction – which doesn’t exist – combined with an equally sloppy implication that the Higgs doesn’t exist – which wouldn’t follow from the first, even if the first thing were right – combine to an example of totally irrational reasoning.”

    Does the term irrational reasoning really make sense and are there sloppy implications- well maybe, that sort of logic may perfectly well describe some sort of paradox or contradiction for a particle that does or does not exist, totally.

    Just because you may be right that we can only apply chance to the outcomes of experiment ultimately, we do have to distinguish from such experiments what is or is not the results that are science or like ESP may be there but not always a repeatable experiment. In any case logic is more a branch of philosophy than science and when our expectations are not fulfilled or our faith in a system in danger do we not lapse in our reductionism and appeal to the metaphysics- or should we just take a deep breath between the dialog and get back to honest work?

    Our minds are like such a Higgs particle and mechanism when we tarnished open enquiry and needed objectivity with such rhetoric as may be on someones particular habits of research.

    That said, the standard theory is quite beautiful even if limited and is a great achievement- but alas, to be sensitive to beauty is to raise the old philosophic issues and interpretations of “Why?”

    Thank you Gibbs and all the other beautiful people.

    The PeSla

  42. wl59 says:

    The SM is essentially correct, however it’s important to see its limits. One can ‘unify’ or explain as constituted by something more basical, only what is exactly this, i.e. what’s secondary. Anywhere, however, one reaches elementary things (which ‘shifted’ from molecules, atoms, particles smaller and smaller, but anywhere may be an end, at least in a certain sense). To my opinion, already explained, are basical, primary natural forces / fundamental interactions each ‘connected’ with an own dimension, and for their existence and function aren’t necessary ‘particles’/objects (or more general, accumulations of the dynamical variable such like energy, impulse …). This don’t make impossible that long time after the origination of the dimensions/interactions such structures arose and will happen often ‘near’ interactions.

    Thus, I think, the SM is right where it ‘unifies’ SECUNDARY interactions (such like the strong and weak interaction, electrodynamics, etc), most optimized by represent their constitution by primary ones, removing redundances; however it has to observe its limits, and never will be possible a substitution/representation of the Time by the Space (or vice-versa), thus also not of Gravitation with Inertia etc, which are (or at least include different) PRIMARY dimensions/natural forces. There are formal connections among them, which results in equations of motion, conservation laws, observer-dependent variances/transformation etc etc (all physical rules), but never it will be possible to substitute one of them by another ! According to my opinion, on the occasion or near any inertia, gravitation interaction may perhaps appear something like a Higgs or Graviton, however, not as the causer of the interaction, but only as a by-product or phenomen of it.

  43. wl59 says:

    ” “…but considering all data/experimental results what I see in my live, I would judge this as completely insignificant.” – Well, in that case, Spring 2012 data will show that all your life was one giant mistake. ” —

    Not only on every student praktikum but on plenty scientific experiments you see steady-steady ‘measurements’ more ‘significant’ than these of Higgs, for each of which we would have to change the physics completely, but they are simply — nothing. As explained, in this case, for diphoton we have rather different (in the most mass ranges one excluding the other) results from ATLAS and CMS, what indicates specially that the ATLAS results on their highest peak are not reliable, but to call ‘significant’ just the few intervals in which they are agreeing, is not an avaliation which corresponds to the above mentioned experience with data and health avaliation as significant.

  44. Paul Hoiland says:

    One could have guessed that the debate will continue and that some would never be satisfied either with any finding of the Higgs, or with it being in any range except a complete SM agreement with no sidelines, so to speak.

  45. wl59 says:

    Iself don’t evaluate that, primarily, from any ‘politics’ in favor of or against any theory. But just from the experience and with a health scepsis which I gained during the years with ‘significant’ (but wrong) data. Just in this sense: careful, here is NOTHING really significant.

    It’s right that I have the (personal) opinion, there is NO NEED for a Higgs — the Inertia and also the Gravitation can be explained perfectly from the geometry, forces/interactions should exist since the origination of the dimensions (time, space …) as two aspects of the same, whilst structures like particles arose much later. However, this por si wouldn’t forbit but still let POSSIBLE a Higgs, Graviton etc as a by-product/-effect on the ocasion of interactions. For this opinion, the question is not critical – a Higgs could exist, but (in opposite to the SM) it would be also possible, that it simply don’t exist; criterions for check my explained opinion are others.

    But, as said, looking at the data, one don’t see nothing really significant.

  46. gunn says:

    It is not Higgs ( http://arxiv.org/abs/physics/0302013v3 ) because every physics event is interpretted by particles which similar well-known elementary particles – leptons, quarks and gauge bozons. Therefore, if anybody will claim that he had found Higgs then not believe – this is not Higgs.

    • Ulla says:

      Is there something wrong with the GUT-theory? If we compare it to how masses were born in baryonygenesis they didn’t go that route. These were nuclear processes, without leptogenesis? This means we begins with the strong force, as seen in pions, or a pair of quarks. There are no W or Z bosons yet?

      A diphoton-pion transition?
      there exists two asymptotic regimes for the pion transition form factor. Photons in spacelike form (gamma x gamma), no electromagneticity, nor W or Z?
      One regime with asymptotics F 1/Q^2 corresponds to the result of the standard QCD factorization approach (weak suppression that comes in late), while other violates the standard factorization and leads to asymptotic behavior as F
      ln(Q^2)/Q^2 (persists over the whole range, in particular in the BABAR region).
      for the photon-pion transition form factor in very wide kinematical region up to large photon virtualities Q^2 about 40 GeV^2.

      An earlier preprint http://arxiv.org/abs/hep-ph/0203064

      How differ a diphoton from a neutral pion?

      • wl59 says:

        The GUT theory has the problem, that it would be one ‘primary’ interaction or natural force, however not formulated in and representing only one own unified dimension, but in time and space, and thus in several dimensions supposed to pre-exist which however should represent the same number of primary forces instead of one unified.

        To my opinion, instead, the primary, first dimension and natural force are simple Events or occured Facts and produced Informations, and their Action. That’s one basical dimension (just a world line of world-points) with one natural force, both discrete, and the action is just the physical action, so that the product of the units of its static (event number) and dynamic (actions) aspect and terms of the line-element is the action too.

        As I explained before, this primary and most basical force is just the Action Principle that ‘everything exists exactly how and for whom it acts (and produce something new)’; any existing fact efectuates steady-steady next facts. Facts somewhere acting and thus ‘valid’ for someone, always stay valid and can’t be made unhappened, on the other hand their re-affirmation would not be anything effectively new, so that this action by the already existing facts includes a restriction of the freedom what can happen and the production of News linear independent of Old, ascertaining the development of the nature always to new aspects, and that the future isn’t contained in the present and past. Subsequent dimensions as collective actions from the subsequent ranks of evolution thus are linear independent on each other, too, and they all are ‘divisions’ of the action into two new mutual complementary static/dynamic aspects such as n/S , t/E , x/P , potential/speed-of-the-surface-increasing . The statical and dynamical form of the line element for real existing objects, with intervals of the static and relative values of the dynamic variables refered to any observer or other reference, are 0 = (dn/1)² – (dt/tpl)² + (dl/lpl)² … and 0 = (S/h)² – (E/Epl)² + (P/Ppl)² – … The relation of these terms to the neighboured and/or to the other of the two aspects, gives conservation laws and all basical physical laws, where the 1st dimension is discrete so that the line element also gives a condition for discrete values which the other variables can assume (depending on the dimensions of the world of the observer, geometry, experiment). For example, the relation between the second and third terms gives c=lpl/tpl as the limit speed and the laws of inertia. The static/dynamic terms of the two aspects are (in a certain sense) interchangeable; specially for the 4th and last-for-matter-relevant terms the potential is either sometimes added to the dynamical form (what results in the classical total energy) or sometimes to the statical form (what results in the second and third spatial direction equivalent to the first).

        Advice: that is only my own opinion, not part of the mainstream physics.

  47. wl59 says:

    According to my opinion, Strings, Branes etc doesn’t exist anyway, Higgs probably don’t exist according the recent data, however if so it isn’t the cause of the Inertia but in the worst case a by-product. I also think that only the effectuated Events/Facts/Informations and Actions are discretized, but not the other dimensions, which get only discrete values according to the geometrical conditions with the action, and the elementary units reflects the average interval or lengh density of events and thus of the realization of the empty space of the dimensions. For clear all these questions, we currently have no significant observations.

  48. Luboš Motl says:

    Comic Sans rules, Phil! 🙂

  49. Soap_Bubbles says:

    In the H –> WW mode;

    If the Higgs boson does exist between 115 GeV and 125 GeV and two W bosons weigh in at 80 GeV, then where does the extra 35 – 45 GeV mass come from?

    Do you think the Heisenberg uncertainty principle is enough to explains the imbalance?

    Could the extra 35-45 GeV indicate that the bosons have structure and hence are composed of even smaller particles?

    • Philip Gibbs says:

      I see what you mean now. Yes the lifetime of the W is short so it has a large decay width. This allows the Higgs to decay via two W bosons even though there is not sufficient energy to produce them on shell. In this mass range most of the Higgs decay to two b quarks but a few still decay via WW and ZZ, enough to be seen. The branching ratios are all carefully worked out and used in the analysis.

      • Soap_Bubbles says:

        Thank you, that helped clarified the math nicely.

        Another point that caught my eye, this is most likely dribble but here goes anyway.

        First, all the mass energy given are center of mass energies, yes?

        Second, when you say in this mass range most of the Higgs decay to two b quarks. I assume you mean via the Z^0 decay?

        Third, can (are) they measuring the ratio of the number of three–jet events to the number of two–jet events, i.e. the strong coupling constant? If so what is the SM prediction of the coupling constant to the existence of the Higgs?

        Hence, if this discovery pans out and there is a 125 GeV Higgs, its existence so very close to the Z^0 resonance could be of major significance. I guess what I am asking is if there are more three-jet than two-jet events and with a fairly large cross-sections both particles could have a lot in common, enough maybe to confuse them as two separate particles instead of one very unstable particle?

  50. Soap_Bubbles says:

    Thank you, indeed, for a super job in keeping us abreast in the historic conference. Very interesting.

  51. wl59 says:

    I looked to another aspect of the data, even trying goodwill to see there a Higgs.

    For a real signal, rather than a peak because of a simple lack (or statistical low weight) of the data, the ‘value’ or amount of the necessary contrary data for an EXCLUSION should not diminuish with increasing luminousity (whilst in contrary, at a valey or already-/nearly-exclusion the additional data for a better exclusion are decreasing with more luminousity).

    Take we diphoton what’s the main reason for the claimed indication of Higgs. On ATLAS, with 1.08 and 4.9 luminousity. With 1.08 , there was a peak at 128 with the value 6 . At 126 was no peak, but the value was 4 there. Now, with 4.9 , the peak at 129 gone; now there’s a peak at 126 with 4 . Between 123 and 126 the value increased slightly. For CMS, a peak was at 121 with the value 4 ; now it is at 123-4 with 3 where before it was only 2. Thus, doublegama would indicate, although not significantly, there is something.

    However, look we to the other channels near 124. doubleW was there almost 4, doubleTau almost 10. Now they are 1.5 and 3.5 . Also, they show no peak there.

    Thus, there is no significant indication. Perhaps two photons, anything else may manage to produce them or to shift a little bit their energy. But we can expect that with a luminousity of appr. 7-8 , each of doubleW and doubleTau may exclude a Higgs with 95% .

  52. Leo Vuyk says:

    How certain are particle physicists that the 125- 127 GeV bump can not be a so called TOP PRIME Quark?

    • Philip Gibbs says:

      Very certain. A Top prime quark would have spin 1/2 but this particle must have spin zero or spin two if it decays to two photons. They would search for a top prime in different places.

      • Tony Smith says:

        As to Tprime quark around 125 GeV
        what about a pion-type meson made up of
        a Tprime quark and an Up-type antiquark ?

        Would it have 125 GeV mass and spin zero
        and preferred diphoton decay mode ?

        If it were much like a conventional T0 meson
        made up of Tquark and Up antiquark
        the short Tquark lifetime might make somewhat improbable but maybe the LHC has enough luminosity to make some of them.


  53. JollyJoker says:

    Sorry if this was addressed earlier, but I just noticed I see no difference between the LHC only combination and the LHC + LEP + Tevatron one above 120 GeV, except perhaps a small widening of the 153-187 red band.

    Is the Tevatron contribution really completely insignificant already?

    • Philip Gibbs says:

      It has not really been mentioned, but the Tevatron and LHC have equal significance at about 110 GeV. At 120 GeV the Tevatron contributes only a quarter to the combination and above that it drops steeply. In that sense the Tevatron is virtually insignificant. It is also inferior because it cannot compete with the mass resolution that the LHC has in the diphoton and zz channels.

      However, it will still be interesting to see what their final results say. Will they get a hint of the Higgs at 125 geV if that is where it is? They claim that with the full data set they will see or exclude the Higgs. It sounds very optimistic given that they have only about 25% more data to add, but it is impressive that they can get so many channels into the combination.

      • JollyJoker says:

        Ok, it’s far less than I thought. I realize now I should have been able to make a rough guess based on the 5 sigma exclusion area being about what the Tevatron had excluded at two sigma earlier.

        The official LHC + Tevatron combination you mentioned will be quite interesting, especially if they publish a LEE-corrected significance of the peak. I’d like to see the press ask them why a 99.5% certainty is called a vague hint 😉

  54. keep me up to speed in February

  55. Vijay Gupta says:

    It is pleasure to participate in this live discussion. Purpose of participation is bring forward the point-of-view of Pico Physicist

    From PicoPhysics viewpoint is Energy is confined in space. Such contiguous space in which energy is confined can be termed ‘Particle’. Elementary particles are once in which the geometrical energy distribution is simple – single core surrounding by field region. In the core, motion of energy is geometrically circular. Bosons are particles in which the circulation is not closed. A Boson can be considered to be superposition of two particle classes. In Bosons circulation is closed and ‘Higgs Bosons’ circulation is open and matches the curvature of larger particle. The Boson for which circulation is open may be as Higgs Boson. Higgs Boson resembles photons in free space. (They are still defined by plank’s constant ‘h’, but associated wavelength ‘=hc’ is different). Before Gamma rays are emitted in nuclear decay, they are present in nucleus at least for a short while as Higgs Bosons.

    Picophysics believe this is only a particular case of refraction, and shall be handled as such.

    Thanks and Regards,
    Vijay Gupta
    Pico Physicist
    Proponent Unary Law (Space Contains Energy)

%d bloggers like this: