Wow, I did not mean to start a flame war or anything like that... I believe it's better to have a variety of tools in your toolbox, and R, like Java / C++ / Python / MatLab and many other programming languages/platforms environments has their own strengths and valid applications. The only one i'm not so sure about anymore is .Net Not every strategy requires analysis of tens of millions or records, and if yours does - R is probably not the best tool to use, there are many others like Hadoop, Spark, or even MatLab with their parallelization package (if you have deep pockets). Having worked for nearly 20 years with "regular" programming languages listed above - I like the idea of using statistical data analysis platforms like R or MatLab with their higher level programming languages (which are basically what people in the industry call DSL - Domain Specific Language) for a "quick and dirty" slicing and dicing of data and statistical analysis. Once you become comfortable with R's syntax, libraries, modules, and the "weirdness" (sorry ) of the pass-by-promise approach - the productivity can easily increase by an order of magnitude over a traditional imperative language. The main benefit is that this increase in productivity will allow you to iterate over a large number of strategy ideas/candidates much faster. However, when I find a promising strategy confirmed by preliminary analysis and backtest in R- I will switch to Python/Java/C++ to implement a more realistic event-driven backtester, which can then be converted to a production trading algo runner. This is what I'm planning to do anyway, and from reading many posts on this forum I feel like this is a pretty common approach to automated trading.