100fills out 10k orders. IB complains the rate is so high that I use too much their computer resource. If the order is sent by dashtrader, it does not use IB resource...
No, DAS is just a front, the broker is still IB. You need something closer to what HFT-like use - Lime or Instinet. However, they may still charge you extra.
Here are some thoughts: First, gaussian had a great point with the casino experience and I think unless there are potential legal issues to combine a brokerage service with some casino-like perks, you would hit a jackpot. Games share casino mode of operation because they are addicting users through instant feedback loops. Second, check out and play for a couple of hours each some of the most followed online games and ask yourself a question why they have been active for many years and followed by many players while others have failed. I have identified some of the required ingredients: -competition with visible rankings where players can compare their performance against others (some adopt military ranks), -optimal learning curve (not too easy but not too difficult to scare many), -instant feedback - you are scoring points for achieving something e.g. points for a kill or negative points for dying, so why my broker does not congratulate me instantly on a profitable trade?, -recognition badges for reaching some milestones reflecting user's growth in the game-this is common in the RPG and MMORPG genre (You've made your first trade, congratulations!) Third, buy and sell buttons are not enough to engage users for extended time (even if making money!) because there are many fun alternative ways of wasting your time nowadays.
Someone needs to start a new company to compete with IB. Just make sure the front end isn't in Java. The online brokers are too clueless to compete with IB.
Hi Dave, I was looking at the same type of thing, as probably many have. So I narrowed down the scope and eliminated the "trader networking" aspect and just focused on the analysis. Turns out I was able to build an Excel framework based on trade data. Since it was "roll your own" I was able to make it more a "trading partner-coach" that gives trade by trade feedback on the days trading action, during the day. Built a Tilt Meter, Optimal Trade Meter, and a Cash Out Meter for the daily action. Then rolled it all up into a session over session analysis for later. What struck me is that one can use a simple table driven approach to embody trading behavior. Table driven is pretty simple but hard to maintain and test. But as long as the input-output space is limited, it does work. I think the value is more in the semantics of the behavior and the customizable nature. PM me if you want to chat. Cheers, PS: Looked at Fund Seeder, and it seems like overkill for the retail person and heavier weight than what is needed: The ability to get actionable feedback WHILE trading.
Most of the ideas floated around here seem retail-oriented. I think that tricky problem is that it's difficult to build a sustainable business around retail trading and investment. There's a higher churn rate and the market size, even with growing trend and full market saturation, is quite small compared to the overall financial industry or even just the consumer finance space. Even recent "successes" in the retail space like Robinhood are really enterprise solutions (their true customers are a handful of large flow traders). The catch-22 is that enterprise financial apps have such a sticky following and mature UX nowadays that it is a difficult space to break into, which is why most of the available opportunities seem to have a retail slant. With that said, I think centralized post-trade reporting remains a huge opportunity having come from a similar world as you. We waste so many cycles emailing PDFs from our FCMs and PBs to our fund admin, and then our fund admin generates these PDFs that they email to our LPs. Then the auditors come in and you need to gather those PDFs which number in the hundreds to thousands, and so it takes months to put together. It gets worse when you have a multi-FCM setting and have to think about the data governance and compliance issues. And also as a developer, commonly used post-trade systems, formats and protocols e.g. SunGard/Traiana suck to work with because all the protocol documentation is proprietary. At a higher level of reporting, not one fund administrator seems to offer a good platform for client-side (for both investment manager and their LPs) view into their investment, be it simple terms baked into their PPMs/subscription agreements/partnership agreements like liquidity properties, reporting frequency, crystallization etc. or ad hoc reports derived from trades and cash flow that funds-of-funds and other institutional investors often need to prepare their own reporting. The best-in-class commercial tools like Advent Geneva satisfy the fund admin's needs for a general ledger and NAV calculation, but otherwise have rather poor integration with everything else on client side. I see most of the time it's just a glorified web dashboard with monthly statements uploaded at irregular intervals, and LPs hardly bother logging in to acquire their statements or to perform their analyses. Also, fund managers are typically expected to estimate monthly net RoR before the turnaround of the NAVs, and will end up developing their own makeshift workflow for the fund admin's job, and I feel a partially open-sourced solution here would have momentum. Likewise, it takes a certain kind of skill to prepare performance reports for marketing purposes, which should really be derived from the same system - but most fund administration platforms only provide a cursory understanding of the portfolio positions, so you end up needing to roll your own, more granular, homebrew performance reporting workflow anyway. I would pay $80-150k per year to maintain this for a small to medium-sized firm. Moreover, really the same underlying data goes into your fund admin's platform and portfolio monitoring tools, but they end up being desynced processes. In the ideal world, there should be a common, secure layer that replaces this mix of email and long-running FTP servers for file interchange - I feel those have outlived their purpose. I also feel that this would eliminate many compliance costs. And then tools like Venn/Geneva/Addepar should be built on top of it, with a single source of truth but different interfaces for the PMs, LPs and marketing/operations/fundraising/support roles. My experience with regulators is that most of them have a cursory understanding of IT and infosec; a standard solution that gives them check-the-box assurance that there's some kind of encryption, secure interchange, correct financial reporting and appropriate regulatory disclosures will cut audit costs for fund managers by tens of thousands. There's a similar vertical, which is the set of devops tools used for trading: ETL, monitoring, logging and alerts. I feel that most firms repurpose open source devops tools for trading, but there's many domain-specific requirements to trading that you can package it all into one framework that comes with more sensible defaults for trading.
Check out this thread https://www.elitetrader.com/et/threads/backtest-strategy-w-o-programming.350540/
app to map a game controller to order entry. i have used "antimicro" in the past but need something more sophisticated.