CDF Logo Vista/Sleuth Global Search for New Physics in p-pbar Collisions at sqrt(s)=1.96 TeV
Public Page
Exotics Logo

Frequently Asked Questions

  1. What are the conclusions of this analysis?
  2. Is this a proof that there is no new physics in the CDF data?
  3. How are systematic uncertainties handled?
  4. Why are there no model exclusion limits?
  5. What is your sensitivity to new physics?
  6. Could Sleuth have found the top quark in Tevatron Run I?
  7. Is Sleuth the answer to finding the Higgs at the Tevatron?
  8. What are the future plans for this analysis?
  9. If I combine the 5 most interesting Sleuth regions, I get a significant effect! Have I discovered new physics?
  10. Why not use odd numbered events for control, and even numbered events for signal?
  11. What limitations there are in terms of statistical statements made using Vista and Sleuth?
  12. Is it really most likely that new physics will predominantly reveal itself in one Sleuth channel?

Questions and Answers

What are the conclusions of this analysis?

This analysis has not revealed indications of new electroweak scale physics in ~ 1 fb-1 of CDF data.


Is this a proof that there is no new physics in the CDF data?

No.


How are systematic uncertainties handled?

Systematic uncertainties reflect flexibility in the Vista correction model, which is adjusted in the presence of constraints from all available data.


Why are there no model exclusion limits?

The space of possible new physics models is very large. The fraction of these models that can be tested in finite time is very small.


What is your sensitivity to new physics?

For new phenomena satisfying the assumptions on which Sleuth is based -- that new physics will appear as an excess at large summed scalar transverse momentum, predominantly in one appropriately defined exclusive final state -- Sleuth's sensitivity is comparable to a targeted search. Sleuth's sensitivity is reduced (relative to a targeted search) to the extent that one or more of these assumptions are not satisfied, and/or to the extent there is a particular kinemetic feature of the signal not exploited by Sleuth.


Could Sleuth have found the top quark in Tevatron Run I, if we had not known about the top quark's existence?

Running pseudo experiments in which top quark pair production is removed from the Standard Model background, Sleuth is able to find a discovery-level excess in the Wbbjj final state in a Run II integrated luminosity comparable to that accumulated in Tevatron Run I.


Is Sleuth the answer to finding the Higgs at the Tevatron?

Sleuth is an answer to the question: "How do you search for something when you do not know what you are looking for?" For the Standard Model Higgs boson, however, there are very clear predictions as to what to look for, so it is unnecessary to use the machinery of a model-independent search. In addition, dedicated searches for the Higgs boson can take advantage of its predicted properties to create greater sensitivity to a potential signal. So a dedicated search should be expected to discover the Higgs before Sleuth does.


Where are the future plans for this analysis?

We will continue searching for new physics in the remainder of Tevatron Run II data.


If I combine the 5 most interesting Sleuth regions, I get a significant effect! Have I discovered new physics?

No, because you have not accounted for the trials factor when you combined these.


Couldn't you use odd numbered events for a control sample (to develop the correction model), and even numbered events for the statistical tests?

The approach of splitting data into odd numbered events as control and even numbered events as signal needlessly throws away half of the collected data, at a cost of roughly 300 M$. Splitting data certainly does not help address systematic problems, since the systematic problem will appear in the control region just as in the signal region. The only reason splitting data into odd numbered events and even numbered events may be useful is to avoid sculpting of cuts, but Sleuth already solves this problem by having algorithmically defined regions in the single variable SumPt.


What limitations there are in terms of statistical statements made using Vista and Sleuth?

The statistical limitations of Vista+Sleuth are a proper subset of the statistical limitations affecting the traditional targeted search.


Is it really most likely that new physics will predominantly reveal itself in one Sleuth channel?

As a practical matter, a significant excess in a single localized final state will be necessary for a discovery claim -- the field will not believe a story that links together five 2&sigma effects. In addition, quantum mechanics (and our detectors) are such that significant fine-tuning is required for a signal to appear at nearly equal strength in two or more Sleuth final states. Top quark pair production is a good example: although top quark pair production results in a number of different final states, the final state Wbbjj carried most of the statistical weight in the argument for the top discovery. As was done for the top discovery in 1995, Sleuth gathers the Vista final states 1e+1pmiss1b3j, 1e+1pmiss2b2j, 1mu+1pmiss1b3j, and 1e+1pmiss2b2j into its single Sleuth final state Wbbjj.


Back to main page