The big key to the success of any platform isn't that it is open source - it's that it develops a large community of people using it.
TradersStudio (which is $599, btw, used to be $499) is a very interesting product, but I'm a bit frustrated by the lack of community. I know Murray is working on this issue - but to me it is critical to the success of the product. The product is a little rough vs. SmartQuant imho, but it is also a bargain. Tradersstudio also has some features that are very unique - such as the ability to deal with unique issues associated with stock splits relative to system trading. Their optimization is fairly simple at the moment but I'm sure that will be increased/improved.
RE is an interesting product and seems to be developing a good community - to me, this is the key for whether or not it will be successful. But the developers are great - very responsive, very thoughtful. If they add a portfolio level to the application, they will have a strong product. They are going to include a level of optimization for the 1.0 release, but I expect it to be fairly simple.
Tradestation has a very strong community, despite the fact that the product is relatively primitive in term programming model/structure. Not that you can't do amazing things in TS, but it lacks portfolio-level constructs. I expect they'll add them - but they don't have them presently. They also charge quite a bit for their platform if you don't have a brokerage account ($300+/month or over $3k a year). If you have a brokerage account, you can reduce that to around $1k a year (or $0 if you can met the minimum trades - not hard but not a guarantee). Lastly, their brokerage operation makes most of their money off of the lack of interest they pay on cash in the account - never a good sign. On the other hand, that great community means you have access to add ons like genetic algorithms, etc.
Wealth-lab has a fantastic community - but being a Fidelity product, I can't buy into it. Also, as far as I know, they don't support futures (but I could easily be wrong - people have comments on this?)
SmartQuant has a polished interface, although they have a strange split between the two types of systems development templates that doesn't make any sense to me. It is very expensive - I had not heard that they weren't doing retail anymore, so that make take it out of contention. I do like their object-based model - works very well. I can't really comment on the community but support seems pretty good.
TradingBlox: I don't like using a custom language, but TradingBlox has an excellent community. This has a lot to do with the fact that they have a forum that is about systems development rather than necessarily about TradingBlox. So they've got a strong community. The software, if you want to make it truly programmatic, is very expensive, but it also very capable.
There are a host of others that I haven't checked out:
- Neoticker: heard it is very good but hard to use.
- Amibroker: seemed nice to me but I couldn't find portfolio-level tools.
- NeedForTrade: the free version of their software (Lite) is pretty darn good. But they are not encouraging a community at the moment. Their use of C# as a programming language certainly makes things interesting. I will also be keeping an eye on these guys.
- TradeSignals: as I mentioned earlier, I'm keeping an eye on these guys.
In my opinion, if InteractiveBrokers would dump their current java charts, buy a company like RE or NeedforTrade and give away the software (charge for feeds), they would own the market completely in about 20 minutes. They would almostly instantly have a large community, and if they picked a product that used a standard language like C#, then people would just flock to them. But I am, of course, not in business development at Interactive Brokers.
Let me know if I'm missing features on these products - I've spent some time on them but I'm certainly not an expert on all of them.
