Tolkien, on 08 January 2013 - 08:13 AM, said:
Thanks Skyscream.
I won't say it's dead, but I can say it is on life support... The only argument against the data that I view as having much merit at this point is that me carrying an ECM with my fearsome 50.6% win/loss ratio is skewing the data. That 4.2x win ratio just looks way too high to me, even as someone who thinks ECM is too beneficial for 1.5 tons, I have trouble accepting it at face value and I was there collecting the data.
Still, I would prefer to collect using a true observer mode so I could just count ECMs and see what happens, as some critics will never be satisfied by anything else. Granted there are some who would debate any result since they don't like to hear that ECM is at least highly correlated with skewed match outcomes.
I'm going to give data collection a rest for a little bit and hope that I can get my hands on some raw data - hopefully Garth can round some up for me so I can find out if the data is fully representative.
Great effort. Were you able to test ECM ability for shutdown mechs?
Edited by StalaggtIKE, 08 January 2013 - 11:59 AM.