Optimize Pileup Protection Window

[jrquirk]

Thus far the pileup protection window has not been optimized and is currently +/-15 microseconds in some analysis and may become +/-10 microseconds in others. At an analysis teleconference elog:273 we decided to raise this issue. The following points were made

• If a mu- stops in plastic, it's lifetime of ~2 microseconds may merit this long a window
• For comparison and consistency with mu+ analysis, a lifetime of ~2 microseconds may merit this long a window
• The cut is symmetric, but an asymmetric window should be examined (lifetime effects, rate effect, background contamination, etc.)
• The effect on a variable time window has been studied and shown to induce lifetime biases in certain situations
• A shorter window improves the maximum theoretical data rate