Tevatron posts excess with final Higgs results

Today is another big day for the Higgs boson and this time it is the turn of Fermilab to give us some new information. The venue for the latest offerings is the conference in Moriond. The talks are still ongoing but the main plot has already been shown. (see QDS)

Later I will give a new combination with the LHC results which may also be updated today, but for now here is a useful variation on the plot showing it as a Signal curve. In this plot the zero line represents no Higgs boson while the line at one is expected for the standard model Higgs boson. What we see is a signal perfectly consistent with a boson in the mass region of about 115 GeV to 140 GeV. The mass resolution is not as good as the LHC results and the significance of the excess is less but consistency with the other results is what we were hoping for, so well done to CDF and D0 for this nice final Higgs result from them.

Update: Here is the combination of the new Tevatron data with the latest ATLAS and CMS. On the left is just ATLAS+CMS on their own, and on the right the Tevatron is included. (Remember these are just my unofficial approximate combinations) The result is a small improvement in the overall level of the peak excess.

21 Responses to Tevatron posts excess with final Higgs results

  1. Frank says:

    So they see a 2.2 sigma excess even in regions – say 130 GeV – that the LHC has alreadz excluded. This just shows that a 2.2 sigma from Tevatron is worthless.

    • Lawrence B. Crowell says:

      There were 120GeV signals from the tevatron going back before the LHC. The data was very sub-par (triple bogey in fact), and 2-sigma data on its own is at best suggestive of something that might requires much closer examination.

  2. Vladimir Kalitvianski says:

    Tevatron excess is as big in the white window as in the neighboring (excluded already) windows, for example, in the 130-140 GeV./c^2 region. Can we use this excess in favour of Higgs at 126 GeV/c^2 if it is certainly wrong (fluke, systematics, etc.) for 140 GeV/c^2?

  3. Vladimir Kalitvianski says:

    Are you telling that the signal is at the right place but the mass resolution is poor? Can’t be other reasons?

    • Lubos Motl says:

      The resolution is inevitably poor because the dominant channel is composed of Higgses decaying to the bottom quark-antiquark pair. Quarks produce jets which have to be separated from other products (does this slow hadron a bit away belong to the jet or now?) so the resulting precision of the jet energy is always worse than if we measure photons in a diphoton channel or 4 leptons in the golden ZZ channel.

      It would be a contradiction if they could measure a sharper signal.

  4. Philip Gibbs says:

    The signal covers a wide range because the detector cannot measure the mass precisely enough to narrow it down as well as the LHC can. Everything comes out as you would expect t to if the Higgs boson is there. The centre of the excess is around 125GeV which is comforting but it does not mean that the Higgs is exactly there. Even the LHC results with their better resolution still allow it to be anywhere from about 115GeV to 128 GeV. The case is not yet settled.

  5. There is definitely a signal but again I am wondering why the signal is interpreted by certain bloggers as Higgs without even asking whether alternative interpretations might be possible. There is also signal around 200 GeV and Higgs or whatever at 125 GeV requires new physics. But since it does not fit the naive expectations we just forget it!;-)

    In TGD framework, where Higgs is possible but does not correspond to the simplest and most elegant option, the Tevatron signal around 125 GeV could correspond to decays of pionlike state to bbar pair via two gluon intermediate state.

  6. wl59 says:

    From the CMS results published today, I updated my analyse somewhere else:

    Thus, it’s rather clear now, that also the 125 peak will vanish soon, and there don’t exist any Higgs

  7. publicfundig says:

    I must add a corollary to my theorem.

    THEOREM: “every experiment paid with public money shows anomalies when turned off, even if for a short uphrade”. The interpretation of this seems to be related to a principle of survival of fittest to possible change of mind of politicians due to financial crises or austerity programs and to attract public money. All in all it is similar to the Higgs mechanism but applied to money rather than to masses.

    COROLLARY: if an experiment shows anomalies, the rival experiment, looking more carefully at its data, actually has discovered that anamalies first.

    The reason is always to attract more public money.

  8. wl59 says:

    The ATLAS people seems to don’t got through yet ‘definitive’ results. However, they published already that they get also a fermiofobic Higgs at 125 (excluded by CMS) with gammagamma – a next indication that gammagamma at that mass value suffers any problem by something else.

    Resumed, in CMS (picture 6 of their combined results published today), gammagamma shows peaks at 115, 125, 135, giving earlier false positives there. Meanwhile, that at 115 is excluded by tautau (and a weaker signal in gammagamma itself), that at 135 by doubledoubleWeh and doubleZet, however that at 125 isn’t yet excluded just because there the other channels aren’t yet significant enough. But tautau is coming, already down at 2.5 times the data need for a 95% exclusion (and tautau isn’t showing neither systematic excesses like WW or ZZ, nor wavelike excesses like gammagamma), so that soon we’ll get rid of this 50-year old higgs zumby

    • publicfunding says:

      Hopefully physicists will start to face forgotten problems of theoretical physics in a natural way instead of postponing them with less predictive models more similar to alchemistic recipes.

  9. wl59 says:

    addition: The biggest argument against the relevance of these ‘peaks’ in gammagamma is, however, that beside them are ‘lacks’ of background gammas. Such effects appears often in physics, f.ex. at the side of stellar spectral lines. This means, that simply the background photons are shifted a little. In the gammagamma Higgs case, any effect ‘focus’ backgrund photons in a way corresponding to a 115, 125, 135 higgs mass result (what even can be any instrumental effect). One see in the poisson-green plot in fig. 6 of the today CMS combined result, that the integral/space under these ‘peaks’ fit good inside the ‘lacks’ at them side — so that in the best case are influenced background gammas but ANYWAY AREN’T PRODUCED ADDITIONAL GAMMAS AS IT SHOULD BE IF THERE IS ANY HIGGS.

  10. Soap_Bubbles says:

    If it is the Higgs, then how can it be lighter than the top quark?

    If true, could it indicate that the top quark has an internal structure?

  11. wl59 says:

    Today also ATLAS published them combined results. And that also don’t looks good for Higgs. See figure 1 in them publication.

    The gammagamma results are the same like before. However, meanwhile the exclusion probability by H > WW > lvlv isn’t longer so much higher than its expectation, and it reaches almost 95% near 127 GeV . Also, the TauTau and bb are of similar exclusion weight like gammagamma, but don’t show no excess over the expectation.

    Thus: byebye Higgs … After almost 50 years

  12. PaoloChe says:

    Do you plan to “unofficially combine” the sigma/sigma_SM
    plot that is the most straightforward thing to do?
    In addition: I’m surprised that at the point with the
    largest deviation on the LHC combination the addition of
    Tevatron results seems not to add significance.

    Thank you

  13. We reach your goals in enterprises which demand the positive qualities we possess, but we excel in those that may also utilize our defects.
    Although legislation of competition could be sometimes hard for your individual, it is best to the race, as it ensures the survival of the fittest in each and every department.

  14. publicfundig says:

    TEVATRON is loosing credibility by putting around rumors every time LHC is on the point to present new results.

  15. A Higgs hunter says:

    The problem with your plots is that when adding a channel the expected goes down, so the bands become thinner, so even if you see the same upper limit, its significance is a bit better….

    • Philip Gibbs says:

      I agree with your statement about how the plots work but I am confused about the part that says “The problem with your plots is”.

      Firstly I don’t regard this as a problem, it is the way the plots work. As more data is added the expected line moves down, but if any excess is a genuine signal, the observed line remians above the base line at 1.0. On a linear scale the bands get thinner. If they are plotted on a log scale as they sometimes are the bands actaully stay about the same size on the plot but move down further. The result is that the significance seen on the plots goes up in line with expectations.

      Secondly I am confused at why you say this is a feature of my plots. All the Brazil band plots work this way. I would agree that to calculate the correct level of significance requires the p-value plots. I have not done any of those because they seem to require more information than is provided in the Brazil band plots and cannot be reliably combined without more information. I mentioned this in a post at

      The same applies if you are referring to the blue signal plots. These are the same as the “best fit” plots that the experiments are producing except that I show one and two sigma bands where they show just one. Here the significance gets better as the bands get thinner.

      It could be that I have missed the real point of your comment in which case I would welcome some clarification of what you meant.

%d bloggers like this: