Most dangerous space weather is driven by flares and Coronal Mass Ejection (CMEs). This could be directly in the form of an Earthward-directed CME producing a magnetic storm. This could be indirectly by a flare or CME accelerating particles along magnetic field lines connected to Earth, a Solar Energetic Particle Events (SEP events or SPEs). So a major improvement in forecasting solar driven space weather, is forecasting flares and CMEs. Now forecasting comes in two forms 1) Forecasting when and what magnitude of event will occur, and 2) The expected rate or probability of an event of a given magnitude or larger. In this paper, we are talking about the second kind of forecasting, and not the first, which is more difficult and might be impossible. The second one is similar to what is the chance of rain, with the forecast of no event (or rain) can be more important than the forecasting than the probability. To develop a forecasting tool for flares and CMEs, it is useful to understand how a flare or CME occurs. It is well known that active regions that display obvious magnetic nonpotentiality are much more CME/flare productive that are active regions that show little or no twist. Evidently, the CME/flare productivity of an active region depends strongly on the nonpotentiality of its magnetic field. In addition, Canfield et al (1999) found that eruptive flare productivity was significantly correlated with active-region magnetic size gauged by sunspot area, independently of whether the active region showed obvious twist in Yohkoh/SXT coronal X-ray images. So it is reasonable to expect that flare and CME productivity (and thus SPEs which are driven by either flares or CMEs or both) will be correlated with active regions free magnetic energy, among other parameters. Thus it would be reasonable to expect measuring the free magnetic energy of an active region, will be a useful predictor of its flare/CME production rate. To develop the empirical rates as a function of the free magnetic energy gauge, we need two databases: 1) the magnetic measures of each magnetogram of each active region (a time history of the active region's free magnetic energy), and 2) the flare/CME/SPE history of each active region (an event history to develop our forecasting rules against). To develop the first database, we have developed an automated code that processes MDI magnetograms, a required step in having a useful forecasting tool, that identify strong magnetic field areas, associates them with NOAA active regions, and determines our gauge of free magnetic energy. We have applied this to all MDI magnetograms through the present. Sometimes two or more NOAA active regions are associated with one strong-field magnetic areas. These cases have been excluded as are any strong-field magnetic areas without NOAA active region association. The second database is labor intensive and we used the flare/CME history developed by NOAA (private communication Chris Balch), which has associated flares with active regions, and determined for each flare if there is an associated SME. The second database runs through Dec 2004. So our combine database consists of ~40,000 magnetograms from ~1,300 active regions, each with known flare and SME histories that run through Dec 2004. With the combine database (May 10, 1996 through Dec 24, 2004), we have determined that there are power law dependencies between our gauge of free magnetic energy and the flare rate, CME rate, fast CME rate, and SPE rate. These rates are incorporated into our beta forecasting tool, which has been delivered to NASA/SRAG.