On Sunday, January 24, 2016 at 10:05:51 AM UTC, Col wrote:
On 24/01/2016 08:54, dawlish wrote:
On Sunday, January 24, 2016 at 7:20:37 AM UTC, Col wrote:



When you made that forecast the models were all over the place. There
was simply no clear signal at 10 days. Well done for calling it
correctly but you really just got lucky.



150 forecasts and over 80% right is lucky??? Or is there something in this technique of spotting when NWP is very likely to be correct at 10 days?

You only forecast on very specific occasions and rightly so, you know
better than anyone how difficult it is.

I do.

As I said, the models were not 'all over the place' 10 days ago. Some people just said they were. There was agreement and consistency and that's why I forecast the change, when, apparently many others couldn't see it. That kind of luck has also happened an awful lot of times before.

They weren't showing any consistency at 5 days never mind 10. I guess
that once again all those people on TWO were wrong and you are right.

They were wrong. That's not my fault. You and they just didn't recognise it, sorry. How successful do you think I'd be if I guessed these forecasts?