Ahh, right. I don't often use appendpipe so I got confused by the example in the docs. Yes, appendpipe operates at the already aggregated values so it won't be very helpful here. Unfortunately if you have a parameter which has to be calculated over the whole set you can't calculate it partially 😉 Your solution will skew results a bit because with the first appendpipe you change the overall count of the results. And if I'm not mistaken it will not be that much more effective than eventstat so you could probably use that one instead. The other solution could be to "manually" calculate the parameters - but that could be tedious. ... View more