Have a look at Neoticker from Tickquest I know there are Hedge Funds that use this software and it is years ahead of Tradestation...
Yes, this is true. There is even a resource called 'Tradestation Institutional Resources' that assists multimillion dollar and billion dollar+ hedge funds with using Tradestation.
Yes, like Excel, Tradestation is in the nooks and crannies of many big firms, used by people who "should know better."
My personal (untested opinion) is that many successful algo-systems are actually quite simple and can be modeled in Excel or Tradestation with relative ease. It's resisting that temptation to turn the system off during drawdowns that snags most would-be gurus. The real secret IMO is getting your hands on very long term, very high quality tick data on multiple markets (15+ years of data on 30+ markets) so that you can thoroughly test your ideas and have an extremely high level of confidence not to shut your system off during a nasty -50% drawdown. I see that many successful hedge fund managers did not turn the corner with their systems until they spent some time working at a large firm. My guess is that the trader was granted access the firm's expensive volumes of historical data which can cost an individual trader many thousands of dollars to purchase on his own. With access to such data, a truly trustworthy system can be built using fairly simple tools like Excel and Tradestation. It's all about the data in other words, not the software.
I've never understood this. Data are a cost of doing business, and among the lesser costs for a full-time trader. Yet many newcomers on this site gripe about the cost. Not sure why -- maybe they grew up with Kazaa and are more comfortable stealing than buying intellectual property.
I agree. If you don't have quality data you will have less confidence in your system and be more prone to to turning it off at the worst time. I've heard people argue that you don't need to drill down to tick data and you don't need more than 6 months of data. But the critical point they are missing is that, the better your data the higher your confidence will be and the less likely you will be to shut your system off when it should be left on. It's been a couple of years since I checked but one vendor was selling 30 years of tick data on all markets for $100,000 USD. Do you need all of that? Probably not. The point is, that data is VALUABLE or they wouldn't be asking 100K for it.
I think data from esignal, tradestation, iqfeed..... is good enough, and cheap. If you cannot trade on this data the problem is not the data but the trader. The amount of data you need depends of the timeframe you trade in. If you daytrade you need less data then if you trade longterm. But with every trading software you get enough data to start with. I never used tickdata, although I daytrade, so I can understand that some traders say tickdata in not needed. It depends of your tradingstyle. 30 years of data is not necessary, especially because trading 30 years ago was completely different from actual trading. If you want to learn automechanics you don't need a car from 30 years ago to practice.
There are theoretically two possibilities: data is valuable and people want to pay for it data is not valuable but idiots want to pay for it Why people pay 2-3 times more for a Lamborghini then for a Porsche? Does the Porsche give you 2-3 time more value for money? Or is the Lambo 2-3 times more worth then the Porsche?
It is based on my personal experience, so it can be different for somebody else. But maybe I am as satisfied with this data, as someone else who buys expensive data.