Well-stated. Clearly, if becoming a profitable trader were just a matter of shelling out dollars for a trading platform, there would be a lot more profitable traders. In fact, for me, I had to break the boundaries of an approach of a mentor of mine, which, although it worked, i.e. had positive expectancy, left me wanting more. Due to the unconventiality of my approach (I use that word reservedly because I am not really the proper judge of whether my approach is unconventional, but I use it also because I've never seen any discussion here of the things I consider most important in trading, so I take the absence of those discussions as indicative that they are not important to others, which is fine by me), I had to build my own tools from scratch and Excel is my program of choice. Nothing I do is visual nor does it require any real-time analysis beyond some simple math. I also have Excel databases which houses all of the background data which drives my trade logic and my past trade results for post-trade analysis, but even for that Excel is more suited than a relational database would be.
For a long time Excel was my program of choice too, but mainly because I was familiar with it. Most tasks don't require coding work, although some VBA would make them less convoluted. It is also very simple to learn VBA, record some macros, and get things moving along. If that is all that is needed, then there's no problem. At a certain point though, you might start adding to your trading repertoire in a way that might require real-time analysis. For example, there isn't any reasonable way that one of my "indicators" could function from a simple database. This is because there is no realistic way to compile a reference list of every particular outcome from a bunch of Monte Carlo sims. The database would just be too massive. Better then to simply run the sim in real time. This is just not possible in Excel.
Listen up guys: Excel (especially 2010) is a heckuva flexible data analysis tool. However, it's NOT APPLICATION SPECIFIC...aka "trading". Your high level tool for that is either Tradestation or Multicharts. These are trading-specific programmable platforms. Each has a robust language with a strong framework. I would include Ninja Trader but it's had a rough release history. It took them almost 2 years to pull-off release 7 and I think it's still unstable.
Sure, if you're looking for something "free". But if you want quality and aren't afraid to pay for it, there are simply better platforms out there than TS and MC.
I'm curious in what ways something like Deltix (there are others, I'm just picking one at random) doesn't meet your needs?
But this goes back to what the other poster was saying, which is that those "trading" platforms only give you the tools to re-arrange the same old indicators and approaches that everyone is losing with, when what you really need to do is something that is outside those particular boundaries. Unprofitable bells and whistles are of no value in the long run. If what you're doing actually "breaks the mold" relative to what other traders are doing, a "trading" platform might be useless. Now, Excel isn't really of great help in actual automation, which is part of what the OP was seeking.
I love threads like this... A bunch of bickering by people who come across like they have no idea what they are talking about. The the strategy or algo needs to be matched to appropriate platforms and vendors - ninjatrader may be perfect for some platforms while it is terrible for others. Every part of the system needs to be complimentary and also make sense - in terms of speed, robustness and price. If your strategy would work on ninjatrader but you still run out and pay for colocation services and data at an exchange that is overkill and may ruin a perfectly good strategy.