Every year during spring training, writers, analysts and fans all try to predict baseball's final standings. Some people pick blind or follow their gut, but it has become more common to start by looking at projections. There are several different projection systems floating around, and they're generally good. I've collected such projections dating back to 2005, which gives me 12 seasons of 30 projections each, or 360 projections total. The top third of teams averaged 89 projected wins and delivered 88 actual wins. The middle third averaged 81 projected wins and collected 81 actual wins. The bottom third averaged 73 projected wins and notched 73 actual wins.
So projections have substance. At FanGraphs we project our standings based on a 50/50 even blend of two individual projection systems. That gives us these projected wins:
The teams are in much the same order as they were a year ago, but inevitably there are going to be differences in the way things play out, which is fun to try to predict. I asked the FanGraphs audience where it disagrees, and the community thinks the projections are too low on the Colorado Rockies and too high on the Oakland Athletics.