The author has investigated the relationship of major teleconnection indices such as El Niņo Southern Oscillation (ENSO), the North Atlantic Oscillation (NAO), the Arctic Oscillation (AO), and the Pacific North American oscillation (PNA) to extreme variability of temperature, rainfall and storminess in the Florida dry season (November 1 - April 30) since 1997. The author's latest work has focused on the attribution of these major teleconnections to extreme variability of seasonal weather events with a significant impact on society in the Florida dry season ranging from drought and wildfire, to cold outbreaks and severe weather from intense extratropical cyclones using logistic regression techniques. In addition, logistic regression was used in an effort to calibrate the strength of relevant teleconnections (i.e., weak, moderate, strong) based on a probabilistic measure of their impact on a particular location, rather than on the strength of the teleconnections themselves. In cases where teleconnections show a strong relationship to critical climate impact variables, but are not themselves readily predictable, logistic regression provides insight into what areas of future research would have the greatest payoff for customers. One of the advantages of using logistic regression is that the customer can be involved in the database development by defining the thresholds for critical values of rainfall, temperature, and storminess that are most important to their particular endeavor with the result being customized probabilistic forecasts of the impact of the teleconnections. Although the author's work has been for the Florida dry season, the basic methodology can be adapted to any area with an available dataset and to any customer with a particular critical seasonal forecast problem to solve in a probabilistic manner. Anecdotal wisdom or traditional deterministic statistical measures that attempt to predict the impact of ENSO, for example, on a particular customer have considerable drawbacks. Consider the recent El Niņo of 2006-2007; was it a weak, moderate or strong event? What should a user of a seasonal rainfall forecast do to exploit the occurrence to their advantage? Conventional wisdom holds that an El Niņo should result in wetter than normal conditions in Florida during the dry season, and many people were surprised when the expected heavy rain did not materialize. However, using logistic regression for a variety of user scenarios based on the most recent El Niņo indicated that the event was weak in the context of its expected impact on rainfall and that there was only a 40% chance of excessive seasonal rainfall, which is a wealth of information for decision-makers beyond just saying that El Niņo is wet in Florida. Logistic regression can be used in concert with a specific customer and their critical threshold to determine the relative strength of a teleconnection event to that customer and provide a probabilistic forecast of a negative or positive impact. Examples will be shown for a number of scenarios and locations for ENSO, AO/NAO and the PNA.