John, there are several major errors in your numbers:
1. Your capacity factor assumptions for fossil-fired power plants are way too high. Correcting for that error alone brings the data back to its original finding: new renewable energy is a major source of new generation, rivaling the contribution of new gas generation. As you note at the end of your post, EIA reports actual capacity factor data for power plants. However, you neglected to use that data for fossil-fired power plants, instead using maximum potential capacity factor assumptions for fossil while using the real-world data for renewable generation, providing an apples-to-oranges comparison. Here is the comparison between your numbers and EIA's real-world numbers for fossil plant capacity factors:
EIA: Combined cycle: 46.5%
Combustion turbine: 4.1%
Steam turbine: 10.7%
EIA: 11.7%, 0.9%, and 6.7% for steam, CT, and ICE respectively
Admittedly EIA's numbers are a fleet-wide average capacity factor, not the capacity factor for new plants alone, though there is no reason to believe new plants should depart significantly from the fleet-wide average, aside from likely having a slightly better heat rate. A cursory sampling of EIA plant-specific data supports the conclusion that new build gas plants have capacity factors that are roughly consistent with the fleet-wide averages indicated in the EIA data above.
2. Your sampling of installed capacity data for January-June 2013 and January-June 2014 is a highly misleading snapshot that misses nearly all new wind installations. Wind was the largest source of newly installed generating capacity in 2012, and wind is likely to approach or even regain that title over the next 18 months because a record amount of wind generation is currently under construction. 2013 and the first half of 2014 have been an extreme anomaly for the low number of wind installations, driven by the very late extension of the PTC at the end of 2012. In addition, the first half of the year is always by far the slowest for new wind installations, as the summer is typically the peak time for wind plant installation which leads to the vast majority of wind projects being commissioned in the fall or winter. Using a more representative timeframe would correctly show wind to be one of the largest sources of new generation.
3. Without going into too much detail on this separate topic, the main reason why DOE's EIA Annual Energy Outlook projections almost always underestimate new wind build is that EIA uses wind cost assumptions that are nearly twice reality. EIA's levelized cost assumption for wind was around $80/MWh last year, based on a theoretical estimate from a consultant that has virtually no experience with actual wind project costs, while the real-world average wind PPA price last year, after removing the impact of the PTC, was around half of that. Combined with EIA's self-acknowledged long-standing bias of underestimating fossil fuel costs, it is little surprise that they continue to underestimate future growth of renewable energy. http://www.eia.gov/forecasts/aeo/retrospective/
American Wind Energy Association