Post-fit distribution In experimental particle physics papers, often one reads "pre-fit" and "post-fit" distribution as the caption of some kinematic distribution of data and simulation.
What is meant by "pre-fit" and "post-fit"?
 A: In Bayesian statistics we incorporate the our "initial believe/knowledge" into the analysis. This is done via the so called prior distribution. After the fit our "believe/knowledge" has been changed due to the incorporation of the new data. Therefore, we have obtained the so called posterior distribution. The two names, prior and posterior, are with respect to the gain of new data.
An example:
Suppose you performed an experiment with $N_1=1000$ experimental units and found that the average value of some quantity is $\hat{\mu}_1=1$. Some times later, I perform an analog experiment using $N_2=300$ samples. The average value of my sample is $\hat{\mu}_2=1.5$. How do I incorporate your data into my analysis? There exists simple formulas for doing so. However, if we are not interested in a single quantity (i.e. the average value), but in the "distribution of knowledge", I would use your dataset to define a prior distribution. Fitting my "new" dataset using this prior distribution results in a combination of our datasets. We obtain the posterior distribution, which describes the current state of knowledge.
