Shutting down our small fund

Discussions about Money Management and Risk Control.
Roscoe
Roundtable Knight
Roundtable Knight
Posts: 250
Joined: Sat Jan 24, 2004 2:06 am
Location: Houston TX

Post by Roscoe »

Hey Cymond. Agree, the markets have changed but what has changed more is the notion of trust - used to be that you could trust your broker and their FCM, in fact the whole business was essentially founded on trust. Corzine and Wassendorf have done all they can to destroy that concept. I guess that FCM failure fits Talebs 'Black Swan' definition.
Chuck B
Roundtable Knight
Roundtable Knight
Posts: 481
Joined: Thu Apr 17, 2003 6:34 am

Post by Chuck B »

Roscoe wrote:Hey Cymond. Agree, the markets have changed but what has changed more is the notion of trust - used to be that you could trust your broker and their FCM, in fact the whole business was essentially founded on trust. Corzine and Wassendorf have done all they can to destroy that concept. I guess that FCM failure fits Talebs 'Black Swan' definition.
I would add that what destroyed trust at a much higher level is the response to the MF Global theft by the authorities/regulators. It's one thing to have a broker steal from supposedly segregated customer accounts as MF Global did, but to have said customers left high and dry by our "system" is beyond reproach. Welcome to the new normal. I suppose we have more and perhaps worse circumstances out in front of us.
Roscoe
Roundtable Knight
Roundtable Knight
Posts: 250
Joined: Sat Jan 24, 2004 2:06 am
Location: Houston TX

Post by Roscoe »

Hi Chuck. Well put! If my understanding is correct, MFG customers now have something like 90+% of their money back, while PFG customers are not likely to ever get more than ~50% back. I can't speak for the MFG case, but with the PFG case the NFA and CFTC were asleep at the wheel for 20 years and yet they bear no responsibility and cannot be held accountable. That does nothing to help the trust quotient either. Might as well have no regulators for all the actual good that they do. Mumble....
Chuck B
Roundtable Knight
Roundtable Knight
Posts: 481
Joined: Thu Apr 17, 2003 6:34 am

Post by Chuck B »

Hey, yeah, I've received 80% back on two accounts, so those are still in their Corzine imposed 20% drawdown, and I'm thinking we're not getting back much more, but time will tell I guess. It took long enough to get to this level. One account that was 100% US seg had received back 80% level about two months ago, and the second account which was in US seg, and US secured and EUR secured, just received the additional amount to get it up to 80% level. James Koutoulas has been instrumental in leading the charge on *so* many levels with this MF Global debacle -- very impressive guy.
Roscoe
Roundtable Knight
Roundtable Knight
Posts: 250
Joined: Sat Jan 24, 2004 2:06 am
Location: Houston TX

Post by Roscoe »

Glad to hear that you are getting something back. I wish there was a PFG version of James Koutoulas! I don't think that anybody is leading anything there. Wassendorf seems to have assets stashed offshore but he is not being particularly helpful to the liquidator about them, guess he is planning ahead for when he gets out of the joint (which will be when he is about 200 years old unless he has Corzine-like connections).
MarkS
Roundtable Fellow
Roundtable Fellow
Posts: 70
Joined: Wed Dec 15, 2010 12:10 pm
Location: Chicago, IL

Post by MarkS »

Chuck --

Excellent posts and thank you for sharing. With regard to orders on the first day, my partner and I also had that discussion; we ended up doing probably the least scientific thing imaginable: we pulled up all the open positions the night before our trading "start" and eyeballed the trends versus what we knew were our average trade length in our simulations. The trends seemed to be long in the tooth on that basis, so we made the executive decision to start by taking new trades only.

Your thoughts on using as much data as possible are also something I have given much thought to. Makes me wonder if one could have millions of data points, and stick to a single system that works across all of it... are you condemning yourself to using something that works "on average" across all markets and thereby guaranteeing mediocre results? Is it even possible to find something that could work across a vast spectrum of economic cycles, technological changes, and political maneuvers (be it deflationary policies, war, etc)? I find myself currently being drawn to what the folks at O'Connor practiced before they sold to UBS -- running a hundred or more systems every day that test over the past X days/weeks/months, and then trading the top ten or twenty. As the market changes, the rankings change; as long as your lag is not too great or the market changes occur too quickly, you should find yourself trading the "right" systems. Intriguing concept now that I'm back on the sidelines.

Roscoe - thanks for sharing the stats of the alternative start date. So based on your comments, you have not gotten any recovery to date? And 50% is what people are expecting? I cannot imagine that the other clearing firms wouldn't pony up to make customers whole for the sake of the industry. I hope that is the case, or there is some other remedy. Like you said, for years I remember the CME and the CBOT bragging that no customer had ever lost money to a failure by a party on the other end of a trade, nor a clearing firm failure. They were very vocal and proud about that. Apparently times change...
:roll:
jas-105
Roundtable Knight
Roundtable Knight
Posts: 130
Joined: Sat Aug 02, 2008 2:32 am
Location: London, England.

Post by jas-105 »

MarkS - Thanks for the "postmortem" on your fund, it's kind of you to share this info. I'm starting to believe that, for long-term systems at least, 30 years of data just isn't enough and that most of us have built parameters based on a market period that was particularly suited to trend following. This isn't to say that our models won't work in the future but that our assumptions about risk adjusted returns may be over optimistic. If you had been able to test your model over 1000 years and it gave you a 15% return but with many 40% drawdowns would it have changed your approach ? It's still a great rate to compound money given the relatively low barriers to entry (assuming the 1000 years of data wasn't too expensive !) but it wouldn't be easy holding investors through the bad times.
Somebody recently posted a paper of 100 years of trend following which was well researched but indulged the reader by taking a very long-term view and I think the worst drawdown length was in the region of 8 years. This is way too long to be able to keep investors unless you already have a decent track record.

On a separate note, I was over in Chicago a few weeks back and I dropped by Attains's office and heard the PFG story. I doubt anything is as painful for a trader to be a victim of fraud and I feel dreadfully sorry for those affected. The long term effect on the futures market, together with the HFT issues, doesn't look great, I just hope this is the darkness before the dawn.
Roscoe
Roundtable Knight
Roundtable Knight
Posts: 250
Joined: Sat Jan 24, 2004 2:06 am
Location: Houston TX

Post by Roscoe »

H Mark. Nothing to date from PFG, the wholesale transfer of all PFG accounts to Vision Financial happened over a month ago and I still can't get any response from anyone at Vision, although to be fair I am sure they are very busy. A friend received his Vision statement a month ago but I have seen exactly nothing from them apart from acknowledgement of my new account forms on Oct 26.

Nobody has even hinted at anything like making us whole and the best (and only) guess from my broker has been "expect a 50% haircut".

Yeah, the PFG and MFG failures will hurt the futures industry but it seems that, as Chuck said, this is the 'new normal'. Nobody gives a rat's patootie.
Moto moto
Roundtable Knight
Roundtable Knight
Posts: 427
Joined: Mon Jun 01, 2009 4:12 am
Location: once again in the UK

Post by Moto moto »

thanks for peoples feedback.
given this is in the money mgmt section...
one thing i have wondered about in some random thoughts----

given most fund managers are expected by investors to have the "bulk" of their money in their own funds, and its shown that (maybe with the exception of arbs) nothing works over the long term all the time (there are drawdowns) and that there are risks from MFG, PFG etc.....then the only answer to all of these is diversify, diversify, diversify - --- but then you run into the problem of flat returns as nothing really makes a lot.....

as I said...random thoughts about why we are really doing this. If its flat returns maybe its worth while investing the bulk with others, as oppsed to running it yourself, unless of course you wish to build a business around it. Spend the time on looking for the outliers and apply the same principles of TF (without the diversification) there....
Chuck B
Roundtable Knight
Roundtable Knight
Posts: 481
Joined: Thu Apr 17, 2003 6:34 am

Post by Chuck B »

MarkS wrote:Your thoughts on using as much data as possible are also something I have given much thought to. Makes me wonder if one could have millions of data points, and stick to a single system that works across all of it... are you condemning yourself to using something that works "on average" across all markets and thereby guaranteeing mediocre results? Is it even possible to find something that could work across a vast spectrum of economic cycles, technological changes, and political maneuvers (be it deflationary policies, war, etc)? I find myself currently being drawn to what the folks at O'Connor practiced before they sold to UBS -- running a hundred or more systems every day that test over the past X days/weeks/months, and then trading the top ten or twenty. As the market changes, the rankings change; as long as your lag is not too great or the market changes occur too quickly, you should find yourself trading the "right" systems. Intriguing concept now that I'm back on the sidelines.
Thanks for your thoughts on this as discussions like these are essential (to me at least) mental stimuli that prompt additional thought paths.

On the latter concept of trying to fit a group of many systems based on recent performance -- I would imagine that you'd need a wide range of strategies that depend on completely different methods to generate returns versus the risks they assume. It's an interesting concept which I suppose is solely based on the belief that "what has happened recently will continue to happen." This is actually a reasonable belief to hold in markets over the short term I suppose. I assume the variable in this methodology with the most impact on results would be the ranking and switching logic? Decisions made within that might likely yield the highest variability in results going forward, hence it might take a lot of thought and testing in that algorithm.

On the thoughts of a sufficient amount of data -- kind of makes one wonder about the whole concept of "trend following" and what it really means to some extent. What appears like a great trend on one timeframe might look awful on a different timeframe all in the same market, so we end up with all sorts of qualifiers on what a "trend" really means, and I guess it turns out it means many different things in some respects.

If I have a 12ATR trailing exit and follow wheat from $4 to $15 over the course of 18 months while someone else with a "more traditional" TF exit (20-30 day low, 3-5ATR trail, etc) is chopped into and out of wheat over that same time period losing 6 out of 6 trades, did wheat "trend" over that time period? It seems that the common thought, so to speak, of whether wheat "trended" over that time period is whether or not the vast majority of systematic trend following funds/CTAs made money or not which is directly related to what "tests well" on historical data since the majority are using similar exits to try to stay with the trend, so to speak.

Taking these thoughts a step further, what if on the very, very big picture, the time period between 1960 and 1990 was an economic anomaly as far as markets are concerned in some fashion, and "everyone" jumped on board after learning what worked during that time frame from the few who prospered? In other words, what if we looked with a biased eye on the naturally selected winners, who only won because they happened to have a strategy that was successful during that time period? We then take that forward and try to replicate that biased sample of success and project it forward with great confidence, confidence based on an actually small sample of data that has existed since that time?

What if you looked at 1000 years of market data in the year 3012 and found that the original Turtles system had drawdown periods of many years and 75%, but over the 1000 years returned a humongous sum of money? Maybe such systems (i.e. similar to many long term TF) go through long periods of horrid performance (long in our measuring window which is perhaps 1/2 a human life, 40 years, etc), but over the very long term are robust in concept enough to massively make up for those periods during the boon years. What if 1970-1990 was one of those boon year periods?

I realize these are all just wild thoughts/questions, but this is the type of crap I lay awake at night pondering. I've tried for many years to get the really big picture in my mind on all of this, and it is slowly evolving -- not that it's been hugely helpful yet. Using some of Einstein's methods of looking at problems is helpful -- I've imagined peering down from space at the planet and watching all the human interactions occurring though time, the biases, superstitions, successes, failures, etc, trying to get at the core thought, the core belief, that underpins this whole subject. Sorry if this sounds a bit whacko.
Chris67
Roundtable Knight
Roundtable Knight
Posts: 1052
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 »

Ok here's a spanner in teh works - Id like to say Im playing Devil's advocate but it really is what Im seeing
Take a simple donchian B/O system i.e. a number of days break out >100
run it for 30 years on 100 markets
look at teh results

Is there really a problem ? seems to me the equity curve looks very sexy and this is a mildly difficult losing year ? all looks very much within teh realms of what shoudl be expected

That being said the real worry for me is not that this is a tough time - the REAL worry [ and I know I'll be accused of bias/ opinion, etc] is that I really dont see things changing from where they have been for the last 2 years for the next few years
The markets have become almost paralysed into a sense of trade every headline, believe every comment
The truth is that the Global situation [particularly the U.S and Europe] are getting worse by the minute - but these guys will bullshit and lie and carry on plodding through and the really bad bit for us is that the market will continue to believe it until there is an almighty shock - at that point [ abit like pregood run in 2008] I envisage we'll probably, as trend followers, be on the wrong side of that move initially anyway - so there seems to be very little upside right here right now
Ive drawn down capital base aggressively and I really dont care about missing an explosive upside move right here right now [which just aint coming] - i'd rather survive and make 5 % even if my systems make 15% in that period - its just , for me, a Golden rule - if in doubt get out - I doubt markets will get back to what we percieve as "normal" for a very long time
best
c
Chris67
Roundtable Knight
Roundtable Knight
Posts: 1052
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 »

chuck b - great comment by teh way - i share those thoughts entirely - there seems to be a great sense of "the original turtle system is dead" on this forum
How could you possibly know that - it may prove to be the greatest system ever invented when traders sitting in their offices in 2190 look back
they may say wow that system made a humongous amount of money 1983-1998 - it then did nothing for 20 years - it then spent 50 years blowing the dorrs off , had another 10 year flat / losing period - and THEN - Made consistent money for 100 years !!!!!!!

We just dont know - but one thing we know is that the "rules of trading" have woked for centuries - and we may be in a 5 year crap period
I would wager this will revert to normality - but re my post above - probably not for a while yet
AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2071
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Post by AFJ Garner »

A very good post indeed Chuck and I wrestle with exactly the same questions; unfortunately usually also in the middle of the night. Any answer one might arrive at is unfortunately mere guesswork; there do not appear to be many certainties in this universe. For anyone who has not come across the site there are interesting reflections on "Investing Demons" here http://www.cxoadvisory.com/investing-demons/ which make for amusing reading.
jas-105
Roundtable Knight
Roundtable Knight
Posts: 130
Joined: Sat Aug 02, 2008 2:32 am
Location: London, England.

Post by jas-105 »

Chris67 wrote: Ive drawn down capital base aggressively and I really dont care about missing an explosive upside move right here right now [which just aint coming] - i'd rather survive and make 5 % even if my systems make 15% in that period - its just , for me, a Golden rule - if in doubt get out - I doubt markets will get back to what we percieve as "normal" for a very long time
best
c
Sounds like forecasting !
Chris67
Roundtable Knight
Roundtable Knight
Posts: 1052
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 »

yeah it is - shame Jon Henry and one or two others never tried it
cymond
Roundtable Fellow
Roundtable Fellow
Posts: 50
Joined: Fri Jul 10, 2009 1:45 am

Post by cymond »

Chris67 wrote:yeah it is - shame Jon Henry and one or two others never tried it
At least I doubt he "doubled down" as the advice in the link (quoted above) bizarrely suggests: http://www.attaincapital.com/alternativ ... alysis/494. Sure, it works just fine every time on an index that, by definition, never goes to zero.
rajivm
Roundtable Knight
Roundtable Knight
Posts: 131
Joined: Fri Jul 25, 2003 12:50 am
Location: New Delhi, India

Post by rajivm »

cymond wrote:
Chris67 wrote:yeah it is - shame Jon Henry and one or two others never tried it
At least I doubt he "doubled down" as the advice in the link (quoted above) bizarrely suggests: http://www.attaincapital.com/alternativ ... alysis/494. Sure, it works just fine every time on an index that, by definition, never goes to zero.
Even if returns come back -- What use for the people who washed out..
LTTF is similar to option selling..Basically if you practice both for long times they both take you out finally :D
Market Wizards in first books had already predicted this would happen..:D
I want to sue Covel for making me believe that Trendfollowing was the Best thing to happen to science since Newton was hit on head by an apple..
When I first read his site I was wondering -- How could anyone ever lose after knowing Trendfollowing... :)
Chris67
Roundtable Knight
Roundtable Knight
Posts: 1052
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 »

no one can make you believe anything - unless you live in the land of make believe ?
Chuck B
Roundtable Knight
Roundtable Knight
Posts: 481
Joined: Thu Apr 17, 2003 6:34 am

Post by Chuck B »

In a way we all live in a land of make believe. :wink: When you think about it a bit and distill it all down, EVERYTHING we believe in, expect to happen, know is the truth, etc, is all simply beliefs we have adopted, either consciously but mostly unconsciously through our life's experiences.

The key imo is to be able to observe oneself from a dissociated viewpoint, dispassionately, and see those feelings arise to life, peak in intensity and decline back to quiescence. Part of being able to trade properly, to follow any plan/system/program is being able to manage oneself effectively.

At the end of it all, it's all made up in our brains. It's all programming, some instinctual, the rest environmental experience, that we have running constantly. The more subroutines we have running subconsciously that are actually detrimental to our daily life, our professional life, etc, the more likely we are to be totally unaware of their existence (and of course blame all kinds of outside sources as the cause of "our" problem).

(like one little routine bothering me greatly right now and is trying to drag me away from this post is my "get ready for the open" programming...I'm telling it that it has 20 minutes, so pipe down :D )

Cheers,
Chuck
doubleR
Full Member
Full Member
Posts: 15
Joined: Thu Jun 12, 2008 3:25 pm
Location: London

Post by doubleR »

it is also possible that the unprecedented joint intervention of all Central Banks into FX, bonds and stocks has killed volatility and being a trend follower means to be long volatility. Those in 'charge' are forcing the hand of investors into things that, maybe, they would not invest in 'normal' conditions. Still, the price is the price and TF is all about following the price. However, I also believe, that this crisis is rightfully pushing TF on more sophisticated shores with better asset class mix, better managemente of the size and risk.

The Turtle idea to use ATR for position sizing is brilliant and still valid however it is a static concept when the trade is initiated. More sophisticated TF techniques may be required to manage the size DURING the trade. Apologies for stating something mentioned here already...

Happy 2013.
Post Reply