Proceedings for my WWND2016 talk on global Lambda polarization.

There are a few things that make the results I showed at WWND quantitatively different from my "current" results

1) There are no systematic error bars here. I had none at the time of the talk.

2) At the time of the talk, when I performed the mass purity correction to the polarization results I did so including the residual effect in the mass background:

Currently I simply scale the results by (S+B)/S. This assumes that there is no polarization signal in the combinatoric mass background - which makes sense, purely combinatoric Lambda candidates shouldn't have any polarization signal. However we see a small signal looking right and left of the mass peak (that is pairs of protons and pions that pass all Lambda topology cuts but have m_inv < (m_Lambda - LambdaMassWindow) ||  m_inv > (m_Lambda + LambdaMassWindow)). We think this signal comes from protons which are really from Lambdas paired with a bath of pions that come from elsewhere. Since protons are more massive they carry most of the polarization signal so a pion that looks a lot like the pion that the proton should be paired with may make a good Lambda candidate. This candidate may carry a reduced residual signal. At the time of this talk I was accounting for this residual signal. This is not such a large effect (as demonstrated by systematic errors in my current analysis).

3) The feed-down correction at the time 
a) Only included Sigma0s and assumed all other Lambdas where primary
b) Corrected Lambda and AntiLambda PH results, assuming the entire PH comes from vortical effects -- which is to say ignoring magnetic effects.