Did anyone get 2012 right for the June SIPN? excluding those who always go low?
Why exclude those who always go low? At this time of year when there is no reason not to go low is it such a crime to say so? IMO we are in an era where until the end of June, even mid-July at least, a) anything is possible... and b), the combination of unpredictable circumstances which will lead to a record low grows increasingly more likely every year. So my strategy is to assume the worst. I'm no spring chicken but I think it's very likely that I'll live long enough to be right it the end
It's not a crime to go low anymore than it is a crime for WTFWT to put in a high estimate every year. But I am interested in sea ice prediction more as a test of the underlying theory involved in the prediction method. So the information gained from a consistent high or low voter tells us about the bias in their judgment, not about ice conditions or the merits of any underlying theory.
The closest thing I have to a theory is that the collapse will be sudden, that it will take many of the models by surprise, because small uncertainties become much more significant around a tipping point, and that it's timing will be entirely at the whim of short-term weather conditions, for the same reason.
I've estimated consistently low in June since 2013 because I thought that conditions to put this to the test had been favorable up until that point. Does that reveal a bias in my judgement, or a problem with my theory, or both, or neither?
Well I can't predict the weather more than 5 days ahead, and neither can ECMWF or NOAA or anyone else, so if I judge that the early-season weather has been favorable and I have to make a prediction and my theory says that the outcome is a coin-flip...
- if I predict high and the outcome is low and the weather remains favorable then not only am I wrong, but I didn't actually hold my theory up to test - I bet against it instead.
- If I predict high and the outcome is high, then that tells us nothing about my judgement or my theory, regardless of the weather.
- If I predict low and the weather remains favorable and the outcome is high, then my judgement and/or theory are wrong.
- If I predict low and the weather becomes unfavorable and the outcome is high, then then that again tells us nothing about my judgement or my theory.
- If I predict low and the weather remains favorable and the outcome is low then my judgement is ok, or at least not called into question. If the low outcome is a surprise to the models then my theory is looking pretty shiny too.
So given that I think that weather conditions up till early June have been consistent with a possible late-season collapse, the only prediction I can make that may hold my theory up to anything approaching a test is to go low.. This was the case in 2013, 2014 as well. If I say low and the weather remains favorable to melt, my theory can be disproved, or at least sorely dented. That did not happen in 2013 or 2014.
If I had said high in 2013 or 2014 I would have been just dodging an opportunity to test the theory. I don't think that it's fair to contrast this with WTFWT - their position is based on ideology backed by cherry-picked data.