<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html;
      charset=windows-1252">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    Hi Miriam, <br>
    <br>
    I think I put that cut, because I feel it is dangerous just looking
    into pass1 data, 0.00001 was corresponding to chi2/NDF = around 6, <br>
    and put 1.7e-16 corresponds to chi2/NDF=20 (if I am not wrong).<br>
    <br>
    This was discussed in the analysis meetings, before we put this in
    the steering file.<br>
    <br>
    Rafo<br>
    <br>
    <br>
    <div class="moz-cite-prefix">On 1/31/19 2:35 PM, Solt, Matthew
      Reagan wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:MW2PR07MB3913CD4CD6B2511F8457C6189F910@MW2PR07MB3913.namprd07.prod.outlook.com">
      <meta http-equiv="Content-Type" content="text/html;
        charset=windows-1252">
      <style type="text/css" style="display:none;"><!-- P {margin-top:0;margin-bottom:0;} --></style>
      <div id="divtagdefaultwrapper"
style="font-size:12pt;color:#000000;font-family:Calibri,Helvetica,sans-serif;"
        dir="ltr">
        <p style="margin-top:0;margin-bottom:0">Miriam,</p>
        <p style="margin-top:0;margin-bottom:0"><br>
        </p>
        <p style="margin-top:0;margin-bottom:0">For some reason, it's
          set at 1.7e-16 in the steering file:
          <a
href="https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FJeffersonLab%2Fhps-java%2Fblob%2Fmaster%2Fsteering-files%2Fsrc%2Fmain%2Fresources%2Forg%2Fhps%2Fsteering%2Frecon%2FPhysicsRun2016FullRecon.lcsim&data=02%7C01%7Crafopar%40jlab.org%7Cb18072903ac543c4cc8608d687b33d45%7Cb4d7ee1f4fb34f0690372b5b522042ab%7C1%7C0%7C636845601240365635&sdata=zowqSQvWApwhIPQYZmnOji9z3ZKVVRA7MXDZHQ%2FTZhw%3D&reserved=0"
originalsrc="https://github.com/JeffersonLab/hps-java/blob/master/steering-files/src/main/resources/org/hps/steering/recon/PhysicsRun2016FullRecon.lcsim"
shash="A4c+/RC+TpRtZGptZug04FlSwNpOt7W6difMELcG3+y7jPK8aqnz3Mt6bhg9N7hp2CUMrWl9JFtPzsAArKt1Pg5US3XBPF0dPgTi4es2UKqUwGMEsR1eGwXvupsB/6IbC5eJqALByqqFoQ10mWQnl7znhb1576iRl9pHhGjbq5s="
            class="OWAAutoLink" id="LPlnk813197" previewremoved="true"
            moz-do-not-send="true">
https://github.com/JeffersonLab/hps-java/blob/master/steering-files/src/main/resources/org/hps/steering/recon/PhysicsRun2016FullRecon.lcsim</a></p>
        <p style="margin-top:0;margin-bottom:0"><br>
        </p>
        <p style="margin-top:0;margin-bottom:0">No idea why, but I think
          this value helped us a bit.</p>
        <p style="margin-top:0;margin-bottom:0"><br>
        </p>
        <p style="margin-top:0;margin-bottom:0">Matt Solt<br>
        </p>
      </div>
      <hr style="display:inline-block;width:98%" tabindex="-1">
      <div id="divRplyFwdMsg" dir="ltr"><font style="font-size:11pt"
          face="Calibri, sans-serif" color="#000000"><b>From:</b>
          Diamond, Miriam <a class="moz-txt-link-rfc2396E" href="mailto:mdiamond@slac.stanford.edu"><mdiamond@slac.stanford.edu></a><br>
          <b>Sent:</b> Thursday, January 31, 2019 11:22:26 AM<br>
          <b>To:</b> Solt, Matthew Reagan; Graf, Norman A.; Graham,
          Mathew Thomas; Maurik Holtrop<br>
          <b>Cc:</b> <a class="moz-txt-link-abbreviated" href="mailto:hps-analysis@jlab.org">hps-analysis@jlab.org</a><br>
          <b>Subject:</b> Re: [Hps-analysis] Vertex Chisq Probability
          Cut?</font>
        <div> </div>
      </div>
      <style type="text/css" style="display:none">
<!--
p
        {margin-top:0;
        margin-bottom:0}
-->
</style>
      <div dir="ltr">
        <div id="x_divtagdefaultwrapper" dir="ltr"
          style="font-size:12pt; color:#000000;
          font-family:Calibri,Helvetica,sans-serif">
          Wait, how did the cut value get to be <span
            style="font-family:Calibri,Helvetica,sans-serif,serif,EmojiFont;
            font-size:16px">1.7e-16?  The default is <span style="color:rgb(0,92,197); font-family:SFMono-Regular,Consolas,"Liberation Mono",Menlo,Courier,monospace; font-size:12px; white-space:pre; background-color:rgb(255,251,221)">0.00001</span>
            (<a
href="https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FJeffersonLab%2Fhps-java%2Fblob%2Fmaster%2Frecord-util%2Fsrc%2Fmain%2Fjava%2Forg%2Fhps%2Frecord%2FStandardCuts.java%23L144&data=02%7C01%7Crafopar%40jlab.org%7Cb18072903ac543c4cc8608d687b33d45%7Cb4d7ee1f4fb34f0690372b5b522042ab%7C1%7C0%7C636845601240375640&sdata=A0Kcpgc8Tt9Gn9ZTY5q3CkUNOE%2F47dsywq46DW5ozjc%3D&reserved=0"
originalsrc="https://github.com/JeffersonLab/hps-java/blob/master/record-util/src/main/java/org/hps/record/StandardCuts.java#L144"
shash="d5gQRFidsbkfuZpyZSi/nXiDhifLI/i1AyXDkgPpZC3shFnnPE9qRIj0TNMCr0Wd9zWQJrioPEB4UkUddonA9vz0zTBSC+zDAkXbqeO05NN/R4aEbudmhuGm6qH2D94MCZu05UEaCyjfkncPCA941oKQLHXMOoeWeUJhPbk5MqA="
              class="x_OWAAutoLink" id="LPlnk572592"
              moz-do-not-send="true">https://github.com/JeffersonLab/hps-java/blob/master/record-util/src/main/java/org/hps/record/StandardCuts.java#L144</a>)</span><br>
        </div>
        <hr tabindex="-1" style="display:inline-block; width:98%">
        <div id="x_divRplyFwdMsg" dir="ltr"><font style="font-size:11pt"
            face="Calibri, sans-serif" color="#000000"><b>From:</b>
            Solt, Matthew Reagan <a class="moz-txt-link-rfc2396E" href="mailto:mrsolt@slac.stanford.edu"><mrsolt@slac.stanford.edu></a><br>
            <b>Sent:</b> Thursday, January 31, 2019 10:49:36 AM<br>
            <b>To:</b> Graf, Norman A.; Graham, Mathew Thomas; Maurik
            Holtrop<br>
            <b>Cc:</b> Diamond, Miriam; <a class="moz-txt-link-abbreviated" href="mailto:hps-analysis@jlab.org">hps-analysis@jlab.org</a><br>
            <b>Subject:</b> Re: [Hps-analysis] Vertex Chisq Probability
            Cut?</font>
          <div> </div>
        </div>
        <style type="text/css" style="display:none">
<!--
p
        {margin-top:0;
        margin-bottom:0}
-->
</style>
        <div dir="ltr">
          <div id="x_x_divtagdefaultwrapper" dir="ltr"
            style="font-size:12pt; color:#000000;
            font-family:Calibri,Helvetica,sans-serif">
            <p style="margin-top:0; margin-bottom:0">I think I agree
              with Norman here, that I don't really trust the
              probabilities now even if it's a simple calculation. Has
              anyone studied the value in detail? I'm not sure how much
              data this gets rid of right now, my guess is not a lot.
              The cut itself is already pretty loose at 1.7e-16, but I
              don't really know what vertices this gets rid of. I would
              feel more comfortable if this cut were removed. But if
              someone shows me that it has been studied and doesn't cut
              into displaced signal much, I would be happy to keep it
              in.</p>
            <p style="margin-top:0; margin-bottom:0"><br>
            </p>
            <p style="margin-top:0; margin-bottom:0">I do, however, like
              Maurik's change that that cut only should apply to
              unconstrained so that every vertex is guaranteed one type
              of constraint (whether we apply that cut now or in the
              future).</p>
            <p style="margin-top:0; margin-bottom:0"><br>
            </p>
            <p style="margin-top:0; margin-bottom:0">Matt Solt<br>
            </p>
          </div>
          <hr tabindex="-1" style="display:inline-block; width:98%">
          <div id="x_x_divRplyFwdMsg" dir="ltr"><font
              style="font-size:11pt" face="Calibri, sans-serif"
              color="#000000"><b>From:</b> Graf, Norman A.
              <a class="moz-txt-link-rfc2396E" href="mailto:ngraf@slac.stanford.edu"><ngraf@slac.stanford.edu></a><br>
              <b>Sent:</b> Thursday, January 31, 2019 10:32:42 AM<br>
              <b>To:</b> Solt, Matthew Reagan; Graham, Mathew Thomas;
              Maurik Holtrop<br>
              <b>Cc:</b> Diamond, Miriam; <a class="moz-txt-link-abbreviated" href="mailto:hps-analysis@jlab.org">hps-analysis@jlab.org</a><br>
              <b>Subject:</b> Re: [Hps-analysis] Vertex Chisq
              Probability Cut?</font>
            <div> </div>
          </div>
          <style type="text/css" style="display:none">
<!--
p
        {margin-top:0;
        margin-bottom:0}
-->
</style>
          <div dir="ltr">
            <div id="x_x_x_divtagdefaultwrapper" dir="ltr"
              style="font-size:12pt; color:#000000;
              font-family:Calibri,Helvetica,sans-serif">
              <p style="margin-top:0; margin-bottom:0">I, for one, don't
                trust the current chi-squared values for the vertices
                (nor for the track. And to be honest I'm suspicious
                about the hit uncertainties, too.) So unless it's a yuge
                number of vertices that this cut is eliminating I would
                prefer to loosen/eliminate this cut and try to
                understand what those vertices are. </p>
              <p style="margin-top:0; margin-bottom:0"><br>
              </p>
              <p style="margin-top:0; margin-bottom:0">Norman</p>
              <br>
              <br>
              <div style="color:rgb(0,0,0)">
                <hr tabindex="-1" style="display:inline-block;
                  width:98%">
                <div id="x_x_x_divRplyFwdMsg" dir="ltr"><font
                    style="font-size:11pt" face="Calibri, sans-serif"
                    color="#000000"><b>From:</b> Solt, Matthew Reagan
                    <a class="moz-txt-link-rfc2396E" href="mailto:mrsolt@slac.stanford.edu"><mrsolt@slac.stanford.edu></a><br>
                    <b>Sent:</b> Thursday, January 31, 2019 10:11 AM<br>
                    <b>To:</b> Graham, Mathew Thomas; Maurik Holtrop<br>
                    <b>Cc:</b> Diamond, Miriam; <a class="moz-txt-link-abbreviated" href="mailto:hps-analysis@jlab.org">hps-analysis@jlab.org</a>;
                    Graf, Norman A.<br>
                    <b>Subject:</b> Re: [Hps-analysis] Vertex Chisq
                    Probability Cut?</font>
                  <div> </div>
                </div>
                <div dir="ltr">
                  <div id="x_x_x_x_divtagdefaultwrapper" dir="ltr"
                    style="font-size:12pt; color:#000000;
                    font-family:Calibri,Helvetica,sans-serif">
                    <p style="margin-top:0; margin-bottom:0">Thanks
                      Maurik and Miriam,</p>
                    <p style="margin-top:0; margin-bottom:0"><br>
                    </p>
                    <p style="margin-top:0; margin-bottom:0">I think
                      only having the cut on unconstrained, and writing
                      out other collections based on that, is a viable
                      solution. As long as we trust the unconstrained
                      chisq probabilities enough to do this, which I
                      have never looked at before yesterday. But, the
                      question is do we?</p>
                    <p style="margin-top:0; margin-bottom:0"><br>
                    </p>
                    <p style="margin-top:0; margin-bottom:0">Matt Solt<br>
                    </p>
                  </div>
                  <hr tabindex="-1" style="display:inline-block;
                    width:98%">
                  <div id="x_x_x_x_divRplyFwdMsg" dir="ltr"><font
                      style="font-size:11pt" face="Calibri, sans-serif"
                      color="#000000"><b>From:</b> Graham, Mathew Thomas
                      <a class="moz-txt-link-rfc2396E" href="mailto:mgraham@slac.stanford.edu"><mgraham@slac.stanford.edu></a><br>
                      <b>Sent:</b> Thursday, January 31, 2019 9:24:03 AM<br>
                      <b>To:</b> Maurik Holtrop<br>
                      <b>Cc:</b> Solt, Matthew Reagan; Diamond, Miriam;
                      <a class="moz-txt-link-abbreviated" href="mailto:hps-analysis@jlab.org">hps-analysis@jlab.org</a>; Graf, Norman A.<br>
                      <b>Subject:</b> Re: [Hps-analysis] Vertex Chisq
                      Probability Cut?</font>
                    <div> </div>
                  </div>
                  <div style="word-wrap:break-word;
                    line-break:after-white-space">Great!  Thanks Maurik.
                     <br>
                    <div><br>
                      <blockquote type="cite">
                        <div>On Jan 31, 2019, at 10:20 AM, maurik <<a
                            class="x_x_x_OWAAutoLink" id="LPlnk966409"
                            href="mailto:maurik@physics.unh.edu"
                            moz-do-not-send="true">maurik@physics.unh.edu</a>>
                          wrote:</div>
                        <br class="x_x_x_x_x_Apple-interchange-newline">
                        <div>
                          <div style="word-wrap:break-word;
                            line-break:after-white-space">Matt, Norman,
                            <div><br>
                            </div>
                            <div>I started iss398 branch and am testing
                              the code right now. It seems to be passing
                              the Integration tests.</div>
                            <div><br>
                            </div>
                            <div>I will process an EVIO file locally and
                              send it to Jlab or SLAC. Let me know where
                              to put that at SLAC.</div>
                            <div>Should have it available at the end of
                              the day for testing.</div>
                            <div><br>
                            </div>
                            <div>Best,</div>
                            <div><span class="x_x_x_x_x_Apple-tab-span" style="white-space:pre"></span>Maurik<br>
                              <div><br>
                                <blockquote type="cite">
                                  <div>On Jan 31, 2019, at 12:06 PM,
                                    maurik <<a
                                      class="x_x_x_OWAAutoLink"
                                      id="LPlnk907847"
                                      href="mailto:maurik@physics.unh.edu"
                                      moz-do-not-send="true">maurik@physics.unh.edu</a>>
                                    wrote:</div>
                                  <br
                                    class="x_x_x_x_x_Apple-interchange-newline">
                                  <div>
                                    <div style="word-wrap:break-word;
                                      line-break:after-white-space">Thanks
                                      Miriam! That is really helpful.
                                      <div><br>
                                      </div>
                                      <div>I am going to start an issue
                                        with a new branch, just to
                                        suggest how we could modify this
                                        part of the code so the
                                        chisquared check is ONLY on the
                                        UNCONSTRAINED vertexes. If it
                                        passes the unconstrained vertex,
                                        then all versions of constrained
                                        are written out.</div>
                                      <div><br>
                                      </div>
                                      <div>- Maurik</div>
                                      <div><br>
                                        <div><br>
                                          <blockquote type="cite">
                                            <div>On Jan 31, 2019, at
                                              11:49 AM, Diamond, Miriam
                                              <<a
                                                class="x_x_x_OWAAutoLink"
                                                id="LPlnk263094"
                                                href="mailto:mdiamond@slac.stanford.edu"
                                                moz-do-not-send="true">mdiamond@slac.stanford.edu</a>>
                                              wrote:</div>
                                            <br
                                              class="x_x_x_x_x_Apple-interchange-newline">
                                            <div>
                                              <div
                                                id="x_x_x_x_x_divtagdefaultwrapper"
                                                dir="ltr"
                                                style="font-style:normal;
font-variant-caps:normal; font-weight:normal; letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none;
                                                font-size:12pt;
                                                font-family:Calibri,Helvetica,sans-serif">
                                                <div
                                                  style="margin-top:0px;
                                                  margin-bottom:0px">It
                                                  is indeed part of the
                                                  MOUSE cuts: <span style="color:rgb(36,41,46); font-family:SFMono-Regular,Consolas,"Liberation Mono",Menlo,Courier,monospace; font-size:12px; white-space:pre; background-color:rgb(255,251,221)">MinVertexChisqProb</span>. 
                                                  The cut value can be
                                                  set via </div>
                                                <a
                                                  class="x_x_x_x_x_OWAAutoLink"
                                                  id="LPlnk920184"
href="https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FJeffersonLab%2Fhps-java%2Fblob%2Fmaster%2Frecon%2Fsrc%2Fmain%2Fjava%2Forg%2Fhps%2Frecon%2Fparticle%2FHpsReconParticleDriver.java%23L281&data=02%7C01%7Crafopar%40jlab.org%7Cb18072903ac543c4cc8608d687b33d45%7Cb4d7ee1f4fb34f0690372b5b522042ab%7C1%7C0%7C636845601240385649&sdata=xNvdKxBG8zDXIzTcPXN8kREe6eTitJyUWX3%2FVVMgr5o%3D&reserved=0"
originalsrc="https://github.com/JeffersonLab/hps-java/blob/master/recon/src/main/java/org/hps/recon/particle/HpsReconParticleDriver.java#L281"
shash="ZaRqDjHBcUeVH7SwwnkhIujEA1PJzp9V/m32DmbeA1qdWILfarNqjQrBF5D4qKhw5MwlOtJSlfmE5KWodLFXeWh30a0VpK/VYzrsDJPn/xgrnIOFENiDH161DlPfXZig2HZeweLALqLtK/EOdTJOKDiikUwAmY9jOVlzvCaPjqo="
                                                  moz-do-not-send="true">https://github.com/JeffersonLab/hps-java/blob/master/recon/src/main/java/org/hps/recon/particle/HpsReconParticleDriver.java#L281</a> .;
                                                you could essentially
                                                disable it by setting it
                                                to 0.
                                                <div><br>
                                                  <div>The cut is
                                                    applied here: <a
                                                      class="x_x_x_x_x_OWAAutoLink"
                                                      id="LPlnk866444"
href="https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FJeffersonLab%2Fhps-java%2Fblob%2Fmaster%2Frecon%2Fsrc%2Fmain%2Fjava%2Forg%2Fhps%2Frecon%2Fparticle%2FHpsReconParticleDriver.java%23L626&data=02%7C01%7Crafopar%40jlab.org%7Cb18072903ac543c4cc8608d687b33d45%7Cb4d7ee1f4fb34f0690372b5b522042ab%7C1%7C0%7C636845601240395654&sdata=xnA3cEGf90O9dv7KCYorknQ6ZU8hOWhdmWf8DDFlHDM%3D&reserved=0"
originalsrc="https://github.com/JeffersonLab/hps-java/blob/master/recon/src/main/java/org/hps/recon/particle/HpsReconParticleDriver.java#L626"
shash="IhJ6mKMMInaeyNolG7HVuZy9Y8a2kxvMr2MzV/eiVRqXP1sjbwuyx0f8jl5o++4I5AkqbLrlBJ1tNPqDhzNr58AxW9Z4z+CsSxNRQTr5C2xiimXso/JBkoLPSTAFA01zkBVt8keyJQMnCxkzebSjDBWonSznDbMOmvMdIcd1uiE="
moz-do-not-send="true">https://github.com/JeffersonLab/hps-java/blob/master/recon/src/main/java/org/hps/recon/particle/HpsReconParticleDriver.java#L626</a> </div>
                                                  <div><br>
                                                  </div>
                                                  <div>There are no
                                                    grave consequences
                                                    to removing this
                                                    cut, you'll just get
                                                    more junk vertices
                                                    in the lcio.<br>
                                                    <div
                                                      style="margin-top:0px;
                                                      margin-bottom:0px"><br>
                                                    </div>
                                                    <div
                                                      style="margin-top:0px;
                                                      margin-bottom:0px">--
                                                      Miriam</div>
                                                    <br>
                                                  </div>
                                                </div>
                                              </div>
                                              <hr tabindex="-1"
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none;
                                                display:inline-block;
                                                width:918.25px">
                                              <span
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none;
                                                float:none;
                                                display:inline!important"></span>
                                              <div
                                                id="x_x_x_x_x_divRplyFwdMsg"
                                                dir="ltr"
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none">
                                                <font
                                                  style="font-size:11pt"
                                                  face="Calibri,
                                                  sans-serif"><b>From:</b><span
class="x_x_x_x_x_Apple-converted-space"> </span>Solt, Matthew Reagan
                                                  <<a
                                                    class="x_x_x_OWAAutoLink"
                                                    id="LPlnk973086"
                                                    href="mailto:mrsolt@slac.stanford.edu"
moz-do-not-send="true">mrsolt@slac.stanford.edu</a>><br>
                                                  <b>Sent:</b><span
                                                    class="x_x_x_x_x_Apple-converted-space"> </span>Thursday,
                                                  January 31, 2019
                                                  12:15:15 AM<br>
                                                  <b>To:</b><span
                                                    class="x_x_x_x_x_Apple-converted-space"> </span><a
class="x_x_x_OWAAutoLink" id="LPlnk795165"
                                                    href="mailto:hps-analysis@jlab.org"
moz-do-not-send="true">hps-analysis@jlab.org</a><br>
                                                  <b>Cc:</b><span
                                                    class="x_x_x_x_x_Apple-converted-space"> </span>Diamond,
                                                  Miriam; Nelson,
                                                  Timothy Knight;
                                                  Graham, Mathew Thomas;
                                                  Graf, Norman A.<br>
                                                  <b>Subject:</b><span
                                                    class="x_x_x_x_x_Apple-converted-space"> </span>Vertex
                                                  Chisq Probability Cut?</font>
                                                <div> </div>
                                              </div>
                                              <div dir="ltr"
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none">
                                                <div
                                                  id="x_x_x_x_x_x_divtagdefaultwrapper"
                                                  dir="ltr"
                                                  style="font-size:12pt;
font-family:Calibri,Helvetica,sans-serif">
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">Hi
                                                    all,</div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">Long
                                                    email, executive
                                                    summary: we should
                                                    remove the vertex
                                                    chisq probability
                                                    cut in the
                                                    reconstruction.
                                                    Details below.<br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">There
                                                    is a cut in the
                                                    reconstruction on
                                                    the vertex chisq
                                                    probability, which I
                                                    think is part of the
                                                    MOUSE cuts, that is
                                                    causing issues for
                                                    displaced A's. Where
                                                    has this cut been
                                                    verified
                                                    (specifically for
                                                    displaced A's)?
                                                    Plots are attached
                                                    that show the
                                                    following story.</div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">There
                                                    is currently a cut
                                                    at 1.7e-16 for the
                                                    current
                                                    reconstruction which
                                                    seems fine, except
                                                    the distributions
                                                    for the beamspot
                                                    constrained and
                                                    unconstrained
                                                    probabilities look
                                                    crazy, with bsc
                                                    having a significant
                                                    number events at a
                                                    probability of 0 and
                                                    the unconstrained
                                                    distributions have
                                                    events closer to 1
                                                    (I didn't check
                                                    target constrained).
                                                    For the same vertex,
                                                    the reconstruction
                                                    can write an
                                                    unconstrained vertex
                                                    that passes the cut,
                                                    but not a bsc that
                                                    doesn't pass the cut
                                                    (which is definitely
                                                    not what we want).
                                                    This particularly
                                                    effects high z
                                                    events, including
                                                    clean signal events,
                                                    as bsc probability
                                                    is highly
                                                    anti-correlated with
                                                    z. This is actually
                                                    how I just noticed
                                                    it today, while
                                                    looking at displaced
                                                    A's which have very
                                                    few reconstructed
                                                    high z events. This
                                                    effect is not
                                                    noticeable in either
                                                    data or MC because
                                                    the it really only
                                                    effects high z
                                                    events (both good
                                                    signal and
                                                    background).</div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">What's
                                                    worse is that
                                                    apparently the tuple
                                                    maker requires a
                                                    vertex to have both
                                                    a bsc and
                                                    unconstrained
                                                    collection, so if it
                                                    doesn't have a bsc
                                                    vertex it throws
                                                    them both away even
                                                    if the unconstrained
                                                    passes the cut (not
                                                    true in the lcio). I
                                                    first noticed it in
                                                    the displaced A'
                                                    tuples (which have
                                                    the default cut of
                                                    0.00001, so even
                                                    worse than the data
                                                    reconstruction).</div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">What
                                                    affect does this
                                                    have on current
                                                    pass4? It doesn't
                                                    effect the core
                                                    distributions much
                                                    at all, which is why
                                                    I didn't catch it
                                                    yet. It only really
                                                    affects high z
                                                    signal and
                                                    background (which is
                                                    obviously
                                                    important). The
                                                    unconstrained
                                                    collection is
                                                    probably ok in the
                                                    lcio files, it's
                                                    more just the
                                                    beamspot constrained
                                                    collection and the
                                                    ntuples.</div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">So,
                                                    why do the vertex
                                                    chisq probability
                                                    distributions look
                                                    weird? I don't know,
                                                    I haven't had time
                                                    to dig that far. But
                                                    since I don't think
                                                    we understand this
                                                    effect, and we are
                                                    trying to compute
                                                    our data ASAP, we
                                                    should remove the
                                                    cut in the
                                                    reconstruction for
                                                    now. If there are
                                                    consequences to
                                                    removing this cut,
                                                    someone should speak
                                                    up (Tim, Miriam,
                                                    Matt, Norman?).</div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">I
                                                    hope I was clear,
                                                    but I'm sure some of
                                                    you will have
                                                    questions. Also, if
                                                    someone gets a
                                                    chance, can someone
                                                    verify these
                                                    probability plots?
                                                    Thanks.</div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px"><br>
                                                  </div>
                                                  <div
                                                    style="margin-top:0px;
                                                    margin-bottom:0px">Matt
                                                    Solt<br>
                                                  </div>
                                                </div>
                                              </div>
                                              <span
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none;
                                                float:none;
                                                display:inline!important">_______________________________________________</span><br
style="font-family:Helvetica; font-size:14px; font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none">
                                              <span
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none;
                                                float:none;
                                                display:inline!important">Hps-analysis
                                                mailing list</span><br
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none">
                                              <span
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none;
                                                float:none;
                                                display:inline!important"><a
class="x_x_x_OWAAutoLink" id="LPlnk527808"
                                                  href="mailto:Hps-analysis@jlab.org"
                                                  moz-do-not-send="true">Hps-analysis@jlab.org</a></span><br
style="font-family:Helvetica; font-size:14px; font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none">
                                              <span
                                                style="font-family:Helvetica;
                                                font-size:14px;
                                                font-style:normal;
                                                font-variant-caps:normal;
                                                font-weight:normal;
                                                letter-spacing:normal;
                                                text-align:start;
                                                text-indent:0px;
                                                text-transform:none;
                                                white-space:normal;
                                                word-spacing:0px;
                                                text-decoration:none;
                                                float:none;
                                                display:inline!important"><a
class="x_x_x_OWAAutoLink" id="LPlnk369654"
                                                  href="https://mailman.jlab.org/mailman/listinfo/hps-analysis"
                                                  moz-do-not-send="true">https://mailman.jlab.org/mailman/listinfo/hps-analysis</a></span></div>
                                          </blockquote>
                                        </div>
                                        <br>
                                      </div>
                                    </div>
_______________________________________________<br>
                                    Hps-analysis mailing list<br>
                                    <a class="x_x_x_OWAAutoLink"
                                      id="LPlnk319238"
                                      href="mailto:Hps-analysis@jlab.org"
                                      moz-do-not-send="true">Hps-analysis@jlab.org</a><br>
                                    <a class="x_x_x_OWAAutoLink"
                                      id="LPlnk649932"
                                      href="https://mailman.jlab.org/mailman/listinfo/hps-analysis"
                                      moz-do-not-send="true">https://mailman.jlab.org/mailman/listinfo/hps-analysis</a><br>
                                  </div>
                                </blockquote>
                              </div>
                              <br>
                            </div>
                          </div>
_______________________________________________<br>
                          Hps-analysis mailing list<br>
                          <a class="x_x_x_OWAAutoLink" id="LPlnk353836"
                            href="mailto:Hps-analysis@jlab.org"
                            moz-do-not-send="true">Hps-analysis@jlab.org</a><br>
<a class="moz-txt-link-freetext" href="https://mailman.jlab.org/mailman/listinfo/hps-analysis">https://mailman.jlab.org/mailman/listinfo/hps-analysis</a><br>
                        </div>
                      </blockquote>
                    </div>
                    <br>
                  </div>
                </div>
              </div>
            </div>
          </div>
        </div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <pre class="moz-quote-pre" wrap="">_______________________________________________
Hps-analysis mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Hps-analysis@jlab.org">Hps-analysis@jlab.org</a>
<a class="moz-txt-link-freetext" href="https://mailman.jlab.org/mailman/listinfo/hps-analysis">https://mailman.jlab.org/mailman/listinfo/hps-analysis</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>