The way the first half of Oct. 2012 is looking per models, it would appear that it will likely be a good bit colder than 1991 fwiw. However, that would only put a fairly small dent in the +1.6. Nov. 1991 was quite chilly. So, it will be quite difficult for 11/2012 to be colder than that 11/1991. Now, Dec. 1991 was very warm. So, if we could get a cold Dec., rather big gains could be made then. Regardless, overcoming 1.6 F in Oct.-Dec. will be very tough.