Climate warming forecasts grotesquely incompetent according to actual professional forecasters:

Those who were responsible for making the forecasts had no training or experience in the proper use of scientific forecasting methods. Furthermore, we were unable to find any indication that they made an effort to look for evidence from scientific research on forecasting. It is perhaps not surprising then that their implementation of their forecasting method was inappropriate.

But even the best forecasters are out of their depth with climate:

In other words, if one were to recruit the cleverest climate scientists in the world and give them access to all of the available facts about climate, and ensured that all facts were true and all data were valid and accurate, the experts could do no better at forecasting climate than people with only minimal expertise. And their forecasts would even be less accurate than those from a simple heuristic. This finding is astonishing to those who are not familiar with the eight decades of evidence in the peer-reviewed research literature, and nearly all who learn of it believe that while the finding might apply to others, it does not apply to them.

The IPCC seems uniquely lousy at it, however:

The forecasting procedures used by global warming alarmists were not validated for the situation. To address this oversight, we conducted an ex ante forecasting simulation of the IPCC forecasts (from the organization’s 1992 report) of a .03°C per year increase in global average temperature.

We used the period from 1850 through 2007, a period of industrialization and exponential growth in human emissions of carbon dioxide. In a head-to-head competition involving 10,750 forecasts, the forecast errors from the IPCC model were more than 7 times larger than the errors from a model more appropriate to the situation, the aforementioned naïve extrapolation. More importantly, the errors were 12.6 times larger for the long-term (91 to 100-year forecast horizons).

Governments love this kind of incompetence, unfortunately:

We are conducting an on-going study to examine earlier forecasts of manmade disasters such as the global cooling movement in the 1970s, and the environmental movement’s campaign to ban DDT. We have been actively seeking such analogous situations, especially from the people responsible for promulgating alarming forecasts of manmade global warming, to see if there have been any widely accepted forecasts of manmade disasters that proved to be accurate or where the forecasted disaster was successfully prevented by government actions.

In all, we have identified 72 analogous situations, and we judge 26 of them to be relevant. Based on an analysis of these 26 similar alarms with known outcomes, we found that none were based on forecasts derived from scientific forecasting procedures, and all were false alarms. Government actions were sought in 96% of the cases and, in the 92% of cases where government action was taken, the actions caused harm in 87%. (“Effects and outcomes of the global warming alarm: A forecasting project using the structured analogies method”).

The actual best way of projecting climate is to extrapolate last year’s climate, whatever it was, for the next 50 years:

In the case of global climate change over policy-relevant time scales, there is little uncertainty. Proper scientific forecasts provide extremely accurate forecasts. [i.e. –] Climate varies, but our validation study showed that simply extrapolating last year’s global mean temperature resulted in a mean absolute error of only 0.24°C for fifty-year ahead forecasts. It is difficult to imagine how policy makers would benefit if this error were reduced further, even to 0.0°C.

http://www.climatedepot.com/a/6451/Team-of-Scientists-Counter-US-Govt-Report-Global-warming-alarm-will-prove-false–Climate-fears-based-on-faulty-forecasting-procedures