Wednesday, February 07, 2007

Adding to Your Edge: A TraderFeed Poll

In this post, I'd like to ask a favor of blog readers. I'm taking a poll regarding an idea, and I'd like your sincere opinion. I'm not looking for folks to agree with the idea and I won't be discouraged if there's a lack of interest in the concept. Your honest opinion and comments are all I'm looking for. You can post your reactions as comments to the blog, or you can feel free to email them to me at the address at the end of the "About Me" paragraph on the TraderFeed home page. Your comments can be anonymous if you like, and no email addresses or identities will be shared with anyone else.

Three groups of participants are particularly sought for the poll:

1) Developing traders who have not yet established consistent profitability and are open to adding to their edge;

2) Successful traders who have worthwhile trading methods, but have never attempted to actually backtest and improve upon those ideas;

3) Successful traders who develop mechanical trading systems by appropriately backtesting entries, exits, stops, money management strategies, etc.

The poll question is this: Suppose it were possible to create a cooperative structure uniting these three groups of traders in which three things would happen:

1) Developing traders could subscribe to real time signals from a variety of backtested trading systems. They could use these signals for decision support, with the tested ideas providing ideas regarding market direction, guidance for position sizing, price targets, and stops. Over time, the trader would become familiar with these signals as a new set of market indicators and would develop a feel for how to integrate those into his or her existing setups and skills. For a single, reasonable subscription price, the trader would have access to signals that, otherwise, would require many months of system development or many thousands of dollars to purchase.

2) Successful discretionary traders could collaborate with system developers to backtest and improve their trading ideas and adapt them to new markets and time frames. The successful discretionary traders who lack programming skills and/or the time to backtest their ideas could form a partnership with successful system developers, creating a win-win. The discretionary traders learn about what is working in their methods and how to improve returns. The system developers gain access to new market approaches and inspirations for trading new markets and time frames. I recently wrote about a successful trader who shared ideas with Henry Carstens. I have since talked with both, and both have greatly benefited from the collaboration, creating methods more robust than either could have accomplished individually. This could greatly add to the edge of a successful discretionary trader.

3) The successful discretionary traders and system developers could make the signals from their collaborative efforts available to developing traders, creating a constantly updated portfolio of trading methods that provide edge. When new systems are developed jointly between traders and programmers/statisticians, the revenues from subscriptions would be shared fairly between them. This would create a diversified income stream for both groups. Moreover, as the developing traders become more experienced and generate their own ideas, they, too, could collaborate with programmers and make their work available by subscription. The result would be an ongoing flow of new trading methods that would benefit all participants.

Now for a few important additional considerations:

* All intellectual property would be kept confidential. Signals from systems and overall system logic would be shared, but traders would be able to keep their specific methods proprietary if they so choose.

* Subscription prices would be kept low to developing traders, given that they often have limited capital. This is meant to empower traders, not exploit them.

* The signals would be transmitted electronically (email, IM), keeping overhead down and ensuring timely broadcast. This would permit global participation among traders and, over time, would encourage traders from various parts of the world to collaborate on ideas covering the entire range of global markets.

* Every new system would be accompanied by hands-on education regarding how to use the signals. That would be included in the subscription.

* The vision would be to create a Web 2.0 approach to the trading process, in which system development and education would occur in a social network. The "customers" who would utilize the signals could also be the "business owners" who develop (and benefit from) new methods and signals.

In short, what I'm proposing is a kind of trading co-operative, linking system developers, experienced traders, and developing traders. My vision is for a collaborative structure that benefits all parties, instead of the all-too-common "charge a big fee and make big promises" approach to trader education and development prevalent at conferences and the media.

My poll question is simple: Which of the above three groups of traders best describes yourself, and would you have an interest in pursuing something like this?

I realize many operational details and legalities need to be worked out. At this point, however, I'm just interested in determining traders' interest in becoming part of a network that would blend research and trading skills and seek mutual growth and success. I look forward to hearing your honest feedback and will summarize the responses (anonymously, of course) in a future post.

Thanks so much for your interest and support.

Brett

52 comments:

Marc said...

Hi Brett,

Wow, you're really thinking big now.

I'm of the #2 variety of trader. I have tried "systems" in the past and have met with limited success. They all seem to breakdown at some point.

What I've found is that the human brain is so adept at recognizing patterns that it is the best trading system. Relying on a mechanical system is almost like saying "I don't need to think anymore, I'll just let my computer make the decisions". It may seem that I'm biased against trading systems but I'm really just biased against blind trading systems.

That being said, I think a global sharing of ideas and backtested theories would be very exciting. I'd be interested in sharing my ideas and hearing about others. I may not be open to paying for a proprietary method but I'm always open to paying for good information.

Perhaps you could share if you have a more specific plan in mind or if you're just looking for feedback in general.

Thanks,
Marc

Yaser Anwar, CSC (Trader- Equities & FX | Quoted 7 Times In WSJ, Twice in NYT & FT) said...

As a "developing trader" I'd be more than happy to pay up if I could learn from experience, especially if your leading the initiative sir!

LifePost said...

I would be interested. I would classify myself as a developing trader. Your description of the service sounds vague right now, but I like your approach to trading futures and could learn a few things from you and other experienced traders. Having a solid network of experienced traders sharing ideas and their research would be very helpful.

Sabretache said...

Brett

I'm a successful proprietory retail trader. I undertake constant reasearch and backtesting. Principal backtesting/optimisation sofware is Ward Systems Neuroshell Day Trader Professional + their other AI suites. I also use a range of other software. It's my living.

I accept that I will always be learning and that my both my modus operandi and results could probably be improved upon. BUT, I have sweated blood to become consistently profitable and frankly I would not wish to participate in your proposed venture. That's not to say that the idea does not have merit; just that I've found my niche and it pretty much excludes teaching others. Over the years I've found that most newcomers seem to be searching for the holy grail; they have unrealistic expectations and tend to endow 'indicators' and 'patterns' with magical/mystical properties rather than slogging to understand the how and why of what it is they may be 'indicating'. As for understanding the instruments traded and the technical/operational details of markets, forget it. 'Don't need to know about things like that', as if it were anything but absolutely fundamental to the job - hey ho. I find countering that sort of thing a distraction and these days participate only in the briefest of fashions on a few chat sites and BB's.

FWIW, yours is one of my must read bloggs. Best of luck with it

Brett Steenbarger, Ph.D. said...

Hi Marc,

Thanks for the feedback. You make an important point about systems breaking down. I think it's less well appreciated that discretionary traders break down equally frequently as markets shift. One advantage of having multiple systems and methods available is that, as one or two draw down or start to break, others are performing up to par. One feature I'd love to see in this idea is a real-time monitoring of the signals from the systems and their profitability, so that traders could gain a sense for when they're behaving normally and when they might be breaking. I appreciate the interest--

Brett

Brett Steenbarger, Ph.D. said...

Hi Yaser,

Thanks for the vote of confidence. Having read your blog for a while, I have a feeling you may have more to contribute to this effort than your modesty allows you to acknowledge!

Brett

Brett Steenbarger, Ph.D. said...

Hi Lifepost,

Yes, lots of details remain to be spelled out, but I appreciate the initial interest. It's the prospect of mutual learning that really excites me about this. Thanks for your comment--

Brett

Brett Steenbarger, Ph.D. said...

Hi Sabretache,

I can only wish more prop traders approached their craft the way you have done. Congratulations on your work with the neural network and AI programs. I completely understand the desire to keep one's work proprietary. If you would ever be interested in using your system development skills to develop ideas contributed by others, I'd be only too happy to have your involvement. Thanks for your note and the excellent points you make re: the desire for holy grails. Getting signals from systems will never substitute for understanding markets!

Brett

Michael said...

Count me as interested.

Brett Steenbarger, Ph.D. said...

Thanks, Michael. I appreciate the feedback--

Brett

Lord Tedders said...

Brett,

I have to agree with some earlier posts that this is a very ambitious project - and an admirable one.

I consider myself both a #2 and #3 (a rare bird perhaps).

So here are a couple of immediate difficulties I think you will face:

1) email signals are too slow and the best (especially short-term) automated trading systems simply give too many signals for email to work with

2) black-box methods cannot be successfully traded by anyone other than those who create them. I don't mean physically possible - I mean psychologically possible. If I didn't know what made my systems tick I would have a very tough time sticking with it. Newbie's would have a very hard time with this IMHO.

3) many of my automated systems are "grey box". In other words I as the discretionary trader must "work the order" to reduce slippage and get the best position. how do I convey that to a newb?

4) automated traders by their very nature are super secretive group. We're like a cult. And there are cults within cults :) And I know from experience that if you haven't achieved a certain amount of programming mojo - you'll never be invited to work with the best automated traders.

5) many of the best automated systems require nearly constant tweaking. I have several systems that are reoptimized daily. Someone who was unfamilar with these adjustments wouldn't be able to understand why the system was constantly changing.

6) As one other person said earlier. I've spent literally thousands of hours to get where I am (and I know guys who have WAY more mojo than me). Why would I share that? Not only time, but I've also spent many tens of thousands of dollars developing these systems, purchasing software, etc. Again, I would need a huge incentive to make it worth my while.

7) The best discretionary methods cannot be fully coded. Yes, you can code a great deal of it, but not all. Additionally, a good discretionary trader is perpetually "reoptimizing" their methodology based on new data. How do you incorporate that into code?

8) Speaking of code - which platform and coding language should be used? I would vote for NeoTicker (they have the most programability for a mass produced software), but even then I have arguements with guys who insist on coding in C# or VB. Going beyond that, even the NT users I know who primarily use Delphi have 3 or 4 different structural layouts. How do you standardize that?

I hope I haven't been too overly critical of your idea. I actually like your idea a great deal. However, I just don't think it is realistic because the type 3 traders will be giving the most and receiving little.

Edward

NO DooDahs said...

I'm interested.

Re: patterns – the human brain is wired to find patterns. Is that dark area in the jungle a tiger hiding behind the tree? Or not? Unfortunately we are so wired to find patterns that we find them where they are not, such as the constellations or the three peaks in a domed house.

Personally, I think the scientific search for indicators can do much to aid in the understanding of market behavior if one approaches it with an open mind and one is willing to discard "conventional wisdom" in the process.

My difficulties with extensive backtesting rest in the scarcity of data available for fundatechnical approaches. Testing purely technical ideas is relatively easy and relatively inexpensive by comparison.

heywally said...

Wow - wadda you kidding? Sign me up! I am between #'s 1 and 2, trying to inch up to #2 so that I can make a living at this; trying to get to that comfort level - that I mostly know what I'm doing - that will allow me to increase the # of contracts that I'm trading. I am trading the NQ's mostly, full time but watch all of the key contracts. I back test only in the sense that - for my daytrading - I visually/manually scan 3/5 minute charts to see how effective my trading ideas work at the places on the charts that I would buy or sell. I try and merge the price action/volume signals with the big/recent market picture and make decisions. This is rocket science, isn't it? BTW Brett, I greatly appreciate your efforts and always wondered about Naperville (I used to get off of the train in Western Springs).

Steven said...

I am as new as the day is long...so I am interested as a developing trader.

Flatwallet said...

Brett,

I too am willing to participate. However, there is much truth to what Sabretache says. Far too many of us young traders want a black box holy grail systems. As you've said market dynamics change and render systems to be worthless. There are two side to every trade. Eventually the market will adjust and eliminate the edge, mostly because some deep pocketed trader/computer will pile on the other side.

However, if your approach is learn and distribute market knowledge and behaviors, then I'm all for it. I say this because all the info you've been position about the mean , averages and volatility has helped me understands what's wrong with my trading. I find this to be much more long lasting than a system per se. Remember, many have tried and eventually failed. Not because it's not a great idea but because life is just too dynamic and random.

LP

John L. said...

Hi Brett,

I'm a Developing Trader transitioning into a Successful trader. I'm not fully supporting myself yet from my trading, but feel the methods I have will allow me to finish the transition. I have been persistent and working hard on my trading for a while now.

In my opinion, there are many ways to trade and make a profit. People need to find methods that compliment their trading style and personality.

I’m always on the lookout for ways to tweak or enhance my trading. So I guess I’d be really interested in the collaboration piece of what you are proposing. I’m willing to share and also interested in learning from others as well.

Some of my entries are very time sensitive. A delay in the entry would increase the risk of the trade being a loss as some trades are scratched out at par. The entries are also not automated at this time. I’d love the automation, but I have not coded all the possibilities and may not be able to code every case.

Count me as interested.

John L.

Stephen said...

Hi Brett, I would be interested. I consider myself of the #1 variety but I have a system that I am working on that has made me some money but nothing to quit the day job over and it is definitely still a work in progress. I would love to compare/learn from the #2 and #3's but not completely copy them. But as someone said earlier they have invested a lot of time and work just to give it away so I can see where they are coming from. Love trader feed, thanks for ding what you do. - max

heywally said...

BTW, let me mention my skepticism regarding trading systems and back testing, as the current market is always a unique HODGEpodge amalgam that is at best difficult to relate to past circumstances, though I understand and use patterns in the short run.

My main interest in this would be as it relates to learning and sharing ideas - I have no interest in canned signals though I would subscribe to a service that enabled me to learn more.

dupaski said...

As a member of the #1 crowd I would love the oppurtunity this presents. It has the potiential to take time off of the learning curve.

Brett Steenbarger, Ph.D. said...

Hi Edward,

You raise very good points re: the challenge of efforts to integrate system and discretionary approaches. What I've found so far is that system developers benefit from dialogue with successful discretionary traders and discretionary traders benefit from the performance tweaks in their trading that can come from the developers. You're right; the issue of reoptimization is a tricky one, but will have to be addressed. Thanks for the insights!

Brett

Brett Steenbarger, Ph.D. said...

Hi NO Doodahs,

Very good point about the challenge of integrating fundamental and technical data into systems. I agree that anything that brings an element of science to what we do can greatly aid subjective judgment. Thanks for the interest and the note--

Brett

Brett Steenbarger, Ph.D. said...

Hi HeyWally,

Thanks for the note. Naperville is not so far beyond Western Springs; definitely worth a stop to hit the Riverwalk (when it's warmer outside!). I like the way you're using charts to identify how signals would perform. A formal review from a system developer might lend additional insights to you--and to the developer!

Brett

Brett Steenbarger, Ph.D. said...

Thanks, Steven, for the interest. I'm also getting a good response by email.

Brett

Brett Steenbarger, Ph.D. said...

Hi John L.,

Great point. Very short term signals wouldn't work for alerts; the time frames would have to be wider, which might not interest all traders. Thanks for the comment and the interest--

Brett

Brett Steenbarger, Ph.D. said...

Hi Stephen,

I totally agree with your concern re: not giving away hard earned ideas and systems. Clearly there would need to be contractual guarantees regarding the integrity of intellectual property. Signals would be made available by subscription to traders, but systems would not be fully disclosed (unless a system developer/trader elected such disclosure). I do think that traders with good ideas can learn from testing them out and programmers can learn from the insights of traders. The hope is that all will be better off as a result of collaboration. Thanks for the comment and interest--

Brett

Brett Steenbarger, Ph.D. said...

Hi HeyWally,

I also am skeptical re: purely mechanical trading. But systems do identify non-random historical tendencies, and I think those are worth knowing--even if you choose to override those based on factors that might make the current trading session unique. Thanks for the perspective--

Brett

Brett Steenbarger, Ph.D. said...

Hi Dupaski,

You're raising an important point. Seeing how other traders think can provide learning experiences and shorten the learning curve for developing traders. That is why I think it's important for education re: use of the signals accompany the broadcast of those signals. Thanks for the interest--

Brett

mOOm said...

I use a mix of my system and discretionary trading which greatly improves performance. I would never use anyone else's signals but am interested whether others are seeing things the way I am seeing them which makes me check my assumptions and analysis. I use a combination of Excel spreadsheets and an econometric model which is optimized each day. No way anyone else could run it without a huge learning curve and I haven't got the time to consistently give other people the signals... So I guess bottom line I'm not interested. At this stage I am trying to be a type 2 trader - getting more consistent. But mostly it is about my confidence in using my model...

Zeidy said...

Would be interested as a developing trader (although have followed and participated in the markets for a long time but started trading es and nq futures only last year). Worked as a professional programmer for financial institutions for over two decades, but not programming anymore for the last few years.

Yaser Anwar, CSC (Trader- Equities & FX | Quoted 7 Times In WSJ, Twice in NYT & FT) said...

Sir-

You should also talk to www.epchan.com he's a phd like yourself and a pure quant. Could be helpful in this endeavor.

Brandon Wilhite said...

hmmmmm.... This is definitely an interesting idea, but only because you would be the instigator. I'm a lot like Lord Tedders as far as my category of trader goes. If you had to pigeonhole me, I'd say I'm in category #3, but I still feel like I have a long way to go. That is, I have some profitable systems that I have developed and actually I am trading full-time since late November.

As far as sharing my systems goes...that's a tough one. There are some things I'd be willing to share and others which I would not.

For me one of the major concerns, frankly, would be that interacting with the newer traders might be a waste of time and psychological capital. It's hard for me when I share something with someone and their reaction is "so?" They usually just have unrealistic expectations and/or fail to understand the rarity of what I'm showing them, which is a real edge.

Again this is interesting; but only because, and to the extent that, you can improve upon what is already out there.

BW

sage08 said...

hi Brett,

I'm your typical #1 type. A novice who started trading in Oct06 and have since chalked up some losses. I certainly need some help here. Currently subsribe to professional market timers' newsletter for direction but nothing like a real time system.

Greg Lepiaf said...

Hi Brett !

With your blog, everyday there is something new to think about !

I am akin to class 3 since I am mathematically inclined by taste and profession.

I could possibly cooperate in a similar way you described the recent collaboration with Henri Carstens : I would not do it for fees but for fun and because trying to put in a mathematical model/computer code the ideas of others can be a way to start new ways of thinkings. And as you so often emphasis we must always keep flexibility.

Best regards

Greg

Michelle said...

As a moderately successful discretionary trader and also as an admirer of your expertise and generosity, it would be both a pleasure and a honor to be involved in such a trading co-operative.

The global reach, the sharing of different styles, generating additions revenue streams, using the strengths of both discretionary and automated styles, utilizing scientific principles, and promoting a viable trading community are very appealing and worthy to implement.

Incorporating a Wiki into the website for such a trading co-operative can support the real-time, communication interconnectivity among the members via user pages.

The resident trading counsellor chez nous--my husband--suggests in his typical out-of-the box style that in order to motivate the owners of automated trading systems to share their goodies is for those owners not only to get subscription fees but also to receive a certain percentage of the wins of traders using their systems. He acknowledges that the legal and practical implications of his idea might be a bit tricky!

Michelle B

T said...

Brett,

Great idea. I think I would be interested in participating in all three.

mdbllbr said...

Any trading system that is profitable will loose the "edge" as soon as it becomes popular. So instead of focusing in the trading systems based on TA we could concentrate our efforts in the fundamental analysis side where the most of us (individual traders) lacks the resources to develop and keep updated in a consistent and comprehensive way.

Instead of providing trading signals, I propose share analysis, reports, market statistics and models using a file sharing mechanism like people do with mp3 files and programmers do in the open source community.

What can we achieve in the financial markets if programmers in the open source community can beat large corporations?

Of course its challenging because some of data is protected by copyrights like music, but we can join and update a large database in real time and use it to feed our systems and models it would certainly make our life easier.

We can also jointly develop and update a large model that can help our decision making. I realized that my trading can be improved if I use momentun or trading indicator provided by a fundamental based model.

Instead of paying money to access people would pay providing regular and timmely services to the community. You can have access as long as you are contribuing with something other people want to have.

Tyro said...

Brett,

I'm probably a type 1 pushing towards 2 with varying success on different strategies. I've tried to identify the areas that most need work and a black box, even one that you approve of, isn't something that sounds valuable. I want to become a successful, independent trader, not a trained monkey.

It doesn't sound like there will be provisions for the new traders to learn the details of the strategy or, more importantly, how the strategy was developed in the first place.

If someone explained the nature of the signals and I understood their history, the underlying theory and I had confidence in it, then I would risk my money on them. As a beginning trader, I am not sure I would make use of this service.

However, I have at least one strategy which has many mechanical features. I would like the opportunity to work with some people to refine and optimize it and in exchange, I would have no problems sharing the signals it generates with others.

Eyal said...

Hi Brett,

That's an interesting idea. My initial thoughts are that I might be interested but in a slightly different angle: a marketplace rather than a cooperative. For example, "purchasing" development work.

From my perspective the IP rights management/guarantees and the fact that successful traders devote so much time to getting there are the main issues to work around.

Good luck.
Eyal

drphone1 said...

I am in the first group think the idea is great. If groups 2 or 3 are concerned about helping newbies it would seem the more group 1 traders in the fold would add to the overall volume of trading and expansion of the markets.

Cucca said...

Count me in, I wish I could say I'm a number two, I'm sucessful but too discretionary to make any logical contributions. However, my youngest son is going to be joining me in the trading world, and I am looking for a way to educate him that will keep his old man, and my deadly built in predjudices, out of the mix. I want him to only trade futures, something I wish I would have started with, rather than stocks. Anyway, it sounds great.

Brian said...

I am in group 1. I'm not sure I would participate. Receiving signals from a "black box" doesn't teach me anything - other than to be dependent on the black box. One thing that I would be very interested in learning more about is the money management side of trading - scaling in/out of positions, risk amounts, stops, etc. I think successful traders could provide info along those lines without "giving away" their hard earned edge.

-Brian

Jay said...

Brett,

Your idea intrigues me, even though I have reservations. I am a systems developer and trader, and I, too, have spent countless thousands of hours creating, coding and tweaking applications that allow me to “observe” what the human brain can’t. It is a never-ending process, but so too is the learning process and required creative input that allows for continued development and improvement. That said, I also feel that if it weren’t for the existence of a few select blogs that I read (including yours, which has become a “must-read” for me), I might not have ever been afforded the opportunity to stumble on a few critical seeds of an idea that has since become central to my process. And so while I am protective of what has essentially become a very valuable “black box” for me (which, in reality, is not “black” to me at all, since I’m the creator), I also see merit in your proposal, which is what you are seeking to confirm.

In response to several of the comments that precede me, I agree with Sabretache in that it would seem to run across the grain to teach other people proprietary trading methodologies, because, at that point, it would no longer be proprietary (unless it was a one-on-one situation with a special person whom I might want to teach, which is a very different matter). Second, I also agree with much of what Edward said. It’s hard to imagine a black-box being successfully traded by anyone other than the person who created it because, at least for me, I would imagine they’d have to know what made it tick (and as Edward also said, I believe “automated traders” are a secretive bunch). So that, in itself, would be a problem, that is unless only a portion of the process is revealed, leaving a gap in terms of specifics. And while the seemingly never-ending process of optimizing and tweaking might also be a concern, in the end, it may really just turn out to be a relatively minor issue compared to some of the other, more daunting, challenges that would probably soon become apparent.

In the end, I believe that developing, and then trading, a systems-based proprietary methodology is a very personal enterprise, which, at least for me personally, has become meaningful only after a pretty fair amount of experience, some very hard lessons and losses, an acquired appreciation of the “power of the beast,” followed by a sincere desire to master the necessary skill-sets. I think this is pretty basic. However, I, like you, also appreciate the learning process, and as long as I’m dealing with a thoughtful, seasoned, considerate, intelligent (but not arrogant) person who values innovation and collaboration (which only begs the question, ‘how do you really know’?), and has something to offer, I’d be very open to working with a group of like-minded individuals to achieve a common goal. So, initial resistance notwithstanding, if a program like this could be properly administered, and could incorporate a well thought-out, but not necessarily harsh or severe screening process, this might very well be worthwhile endeavor with the added opportunity to grow.

Jay

Brett Steenbarger, Ph.D. said...

Hello, and thanks to all for the many comments. I'm afraid I can't respond individually to all the emails and comments on the blog, but rest assured I'm reading them carefully and taking them in. Lots of thoughtful observation and worthwhile points. One point made by several people that's an important caveat is that receiving the signals from a system doesn't really teach you the reasoning behind the system and thus is limited as an educational tool. It could lead traders to become dependent on the signals rather than learn to see markets in new ways. I'll need to think this challenge through and see how it might be best to address it. I appreciate the interest, and I also appreciate the tough questions and issues--

Brett

tan said...

Dr. Brett,

I've been trading for over two years and I read your blog every morning. Out of all the blogs out there this is definitely one of the best if not the best. I probably still fall in to group #1 as a developing trader. I would be interested in this project. I found my trading lack of sturcture and rely mostly on discretionary decisions. As result, my day to day P&L has not been consistant. I'm working on developing into a more sturcture trader with a sound method to provide consistant results. Hopefully this project will help me get there.

thanks,

Tan

tan said...

Dr. Brett,

I've been trading for over two years and I read your blog every morning. Out of all the blogs out there this is definitely one of the best if not the best. I probably still fall in to group #1 as a developing trader. I would be interested in this project. I found my trading lack of sturcture and rely mostly on discretionary decisions. As result, my day to day P&L has not been consistant. I'm working on developing into a more sturcture trader with a sound method to provide consistant results. Hopefully this project will help me get there.

thanks,

Tan

Brian said...

Hi Brett,

I really like your idea. This structure more or less is already in place at a certain site. (site name emailed to you as to not violate your blog policies)

The problem inherent in this model is that system developers on the site, in an effort to sell their system to the most users, offer inherently flawed systems that showcase unrealistic returns. You'll often see systems claiming backtested returns of 1,2,3..10 fold over the past 12-18 months. However, for obvious reasons, the systems are either highly unreliable, excessively risky, or just completely unstable in real-world conditions. Many of them fail, though the survivorship bias naturally removes the losers from the site, leaving a preponderance of "winners" on display to the average visitor.

So the question remains, how do you attract the "true" system developers while weeding out those who are in it for the quick buck? I'm not sure if you can. As others have mentioned, those who have spent thousands of hours developing their systems and honing their edges may have little vested interest in sharing them.

I am in the #2 camp. I would like to interact with system developers as I head in that direction, but I understand their perspective, and that there are no easy shortcuts, a beginner must walk the long path first.

adam said...

I'd love to participate/help, but not sure which group I am in. I'm experienced, but not in this sort of trading, so i'd probably be more "developing". I think alot of the same principles in assessing options are useful in short term trading

cbaer said...

Hi Brett,

excellent idea. I would consider me a type 3) and type 1) trader. That sounds strange? Well my basic believe is that the markets are constantly changing so ultimatly every system will stop working at some point. Be it that the markets have changed (as they constantly do) or be it that the participants have adapted.

So no matter how much time we have invested or money we have spend on developing a system one should get used to the idea that at some point it will not work any more the way it used too. Up till then enjoy your profits :-)

Fortunatly the one thing remaining constant is the question of demand and supply changes. So the task would be to contantly examine what can help us to find demand supply imbalances and adapt or reinvent our systems to profit from them.

I always think of this as a cannon ball approch (fixed system) versus a guided missile approach. With the latter being able to hit farer and moving targets better.

Probably that is the reason why black box models do not seem to be very popular in the poll as they do not teach you anything on how to think for yourself and to figure out your own 'system' that fits to your personality.


A community sharing ideas (see mdbllbr)and resources is very appealing too me so that we could all constantly improve ourselfs and our systems or at least see that our own ideas are unique and therefore not likely to loose there power in the short run.

Great Idea, great blog, great work.

Thanks Brett,

Carsten

steve said...

Hi Brett,

Here are a few additional thoughts and refinements on your proposal.

First my situation. I find myself mostly in Category 3, although by no means an expert, probably more of an intermediate level. After many years of backtesting various trading ideas I have found that many ideas will work under certain conditions and then fail miserably under other market conditions. Intraday systems, especially, seem to fall apart once a little slippage is factored in. Anyway, long story short, I now trade a system which seems to be robust in various markets. It appears to be very robust and I trust it. I trust it even when my account or the market looks really ugly.

So here are my thoughts on your proposal:

- Yes, I really like the idea and would welcome the opportunity to work with other successful traders, either proprietary or systems traders.

- On the other hand, I am not so hot in the idea of selling signals. Selling signals may not help beginning traders that much. What is needed is a deeper understanding of how markets work and the patience and perseverance to get there. I don't think it comes from buying signals. (I should confess I have never bought any signals). Also, there is already a website which offers this...collective2. There are also a number of logistical issues as some of you other posters point out. Finally, the collaboration which creates the systems would then be a money making venture instead of a learning and expanding experience. This opens up a whole world of others issues, personal, legal, and financial which are eliminated if the effort is just for skills development.

- Mechanical systems are somewhat limited in a number of ways. The first set of limitations is due to the way these environments are designed you are stuck with what they provide or you have to build it up from scratch yourself. The second set is due to the power of the human brain over some computer code. I have been programming for 25 years and I can eyeball my best exits a lot better than my code can. (this is one of my personal performance criteria...did I enter and exit better than the mechanical model).

- As I hinted in my personal blurb, there may be additional gradations within your three categories.

- The key to your idea is to help people to find others who are willing to collaborate and where each party has something to offer and something to gain in return.


Finally, Thank you for providing such a great service to the struggling trader community. Reading your book, especially, has been so helpful to me in putting my journey into perspective.


Thanks Again,

Steve
TradePerformance Software

OptionPundit said...

Hi Brett,

This sounds like a great idea. I'm sort of in #2 but to slightly lesser extent.

I personally don't like the ideas of signals for couple of reasons 1) I am a person who enjoys thorough analysis before investing, so the "signal" dependency will take away the passion part, 2) I personally belong to inefficient market theory camp and therefor can't appreciate the "signals" much ...

I like the bigger picture i.e. the idea of sharing expertise broadly and testing the theories.

I am predominantly an Option Trader with a taste for stock trading too. (i.e. following my asset allocation guidelines). And in the options world, lot of strategies are dependent on how market is behaving currently. General strategies are broadly defined and tested over and over. Pls check out my blog at www.OptionPundit.net and let me know if I could be of any help. I shall be happy to give back to you as I have been reading your blog for quite sometime and widened my knowledge horizon.

All the very best and good luck with the idea,

Best regards,
OptionPundit
www.OptionPundit.net

Brett Steenbarger, Ph.D. said...

Thanks again to everyone for the excellent comments and observations. I think the one theme that is coming through loudest and clearest is the need for education/mentorship from successful traders and the limitations of system signals in providing that. I will post my response on Saturday 2/9 in an attempt to address this useful feedback and refine my ideas for collaboration--

Brett

Phil said...

Hi Brett,

I'm a big fan - your stuff is great and your idea is right in line with what we do over at my members site, kind of a collaborative trading community.

I also have Web 2.0 type visions of the future and right now we are collaborating on a book - you can see an example on the free site of one of our educational chapters (Monday mornings).

We've just initiated the eductaional component for new traders and we only allow a certain amount to join each month so our type 2 and 3 traders can give adequate support to the type 1 traders.

Our goal is to move everyone up the ladder as each trained set of eyes gives us another idea machine for market opportunities.

Please drop me a line at admin@philstockworld.com and I'll give you a pass for the members site, we've got hundreds of very sharp people and perhaps we can all put our heads together and move this idea forward!

- Phil