I've read varying numbers, but all the empirical academic studies I've encountered agree: the majority of mergers & acquisitions fail. (Side note: if anyone has empirical counterexamples, I'd love to read them. Please post here or email me). Also, before TicketStumbler I used to work at a financial transaction services firm involved with senior lenders, M&A and private equity.
The only ones that succeed at a rate above 50%, by their measure, are ones where the Companies involved are very closely related. The measurements for success are tricky, which is why it's important to read the criteria used within the study.
Most retail bank acquisitions are successful. Additionally, Johnson & Johnson has a robust track record of successful acquisitions. Google does not.
Here are some qualitative and quantitative articles on the subject (I wasn't able to dig up my favorite academic study on it; I'll keep looking though). Additionally, if you have access to LexusNexus or something similar, there's a myriad information and studies on the topic. In short, acquisitions are more often than not a fool's game that many people & companies have strong incentives to carry out.
http://is.gd/2Fla (PDF from Wharton)
http://www.theseus.fr/03370970/1/fiche___pagelibre/
http://findarticles.com/p/articles/mi_m3257/is_n1_v43/ai_696...
http://news.cnet.com/8301-10784_3-9796296-7.html
All else equal, the usual winners of an acquisition are: the fee trolls, lawyers, the shareholders and upper management of the company being acquired (unless it's hostile or due to negative circumstances, but even then..) and the upper management team of the acquiring company.
Generally speaking the losing parties are: the acquiring company's shareholders, and employees at both companies.