This is a message to Lyndon Homewood, he answered my case 1156539 but I had further questions. Fogbugz says to "send email to email@example.com mentioning 'Case 1156539' in the subject" but that only created a new Customer Case ¯\_(ツ)_/¯ how do I follow up on a case?? Lyndon, Thank you so much for the explanation. It took me a while to notice the "Top markers from median frame" label in the Top 10 but even then I couldn't figure out why it was different. The information on Top 10 doesn't seem much important to me anymore: it is still probably useful to know the size of the medium frame, but knowing how a particular method ran in the median frame seems pointless; it doesn't tell me anything about how that method is performing, and if you want to look at a method in a single frame, the median frame is as good as any. Does my thought make sense? What's the best way to find a culprit that's making my frame summary mean/medium higher? At first I thought of simply checking the methods with highest mean or medium but I actually have to look the occurrences counter and/or the occurrences in the graph. Difference between mean and medium may help indicate spikes, but there's nothing that tells me how much impact a method has over the total set, is there? Thanks again.