Variances in T-Stats Exchange Measure to Actual: Is it pulling ticket status codes or other?

Hello!

I'm trying to identify where T-Stats is pulling exchange data from in Tessitura when using Measures in the Package cube.

In looking at transactional data (pulled everything out of Tessitura and are reviewing) and trying to fully identify exchanges there are wild variances in the amounts.  If we look at a price type that indicates it’s sub or flex (contains keywords like sub, flex, CYO, etc.) but the package name is empty it's assumed that is an exchange.

Perhaps this is the wrong way to go about looking at the transactional data?

We are trying to identify and confirm (with T-Stats being the confirming report) how many subscription exchanges are occurring each season.

Does anyone have any insight into the deeper bits of T-Stats or how best to look at this?  Perhaps I should be pulling all exchange price types rather using the exchange measure?

Thank you!

Michelle