User name D.J. Mack
Log entry time 10:56:18 on June 24,1999
Entry number 296
keyword=ssshsum vs ssshtrk at Q2=1.6Very similar to the situation at Q2=0.6. Figure 1 is low epsilon, Figure 2 is high epsilon. Here we lose about 3% good events at low epsilon using ssshtrk. Again I checked the MM spectrum for these ssshtrk events near zero and they are mostly good events. I won't swear they are all good events; that needs more study. But we make a larger mistake by retaining ssshtrk. Let's dump it and use ssshsum.
Unfortunately, this will increase our low epsilon cross sections and temporarily worsen our separated cross sections.That's life. When we find all the problems it will all click into place.
Of course, this begs the question of how fucked up our tracked beta is. We know we used the wrong hodo slop parameters in replay_3 (ie, 2,2,2,2 instead of 2,2,4,4). I'll look at that next.
FIGURE 1
FIGURE 2
