In 1994, Standish reported a shocking 16 percent project success rate, another 53 percent of the projects were challenged,and 31 percent failed outright. Even though the new reports from Standish show better numbers still they cause a lot of heartburn to software companies and investors.
However recently J. Laurenz Eveleens and Chris Verhoef have published a paper in IEEE challenging the numbers stated by Standish report. The major problems in the Standish report seems to be around the way numbers for the Successful and Challenged projects are gathered. According to Eveleens and Verhoef, these Standish figures are:
“misleading, one-sided, pervert the estimation practice,
and result in meaningless figures”
The problem seems to be coming from the way the Successful and Challenged projects are defined. The definition seems to have loop holes due to which many valid projects are not considered leading to wrong result.
You can read the entire IEEE report here