Making the right decision in limit up or limit down

Discussions about personal psychology for the individual trader.
Post Reply
Stephen Newton
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Wed Dec 31, 1969 7:00 pm
Location: London, England and Cape Town, South Africa

Making the right decision in limit up or limit down

Post by Stephen Newton » Mon Feb 11, 2008 6:14 pm

At the risk of exposing my own weaknesses (my pride talking) but in the hope of learning from this body of experts, I am seeking to get some points of view on a specific trading issue that I faced today.

By way of background, I am trading a long-term trend system using TB with Moving Averages, Portfolio filters and ATR stops. I have been trading this system live with spreadbetting houses (as my ‘live test’) since October and have seen quite good results to date. From all my research (Trading Tribe books, c.f. books, van Tharps books, this web site, Leonardo’s forex challenge and others) I made a decision to trade the system strictly (no ifs or buts).

To date I have managed this quite well (with one or two execution errors that I have hopefully learnt from).

Over the weekend my system gave me a signal to enter Chicago Wheat (Long) at 1094. All weekend I tried to place a stop order through my online broker at 1095. The market had closed per my broker at 1094. The broker was not accepting orders on this market all weekend but was on other wheat markets such as London.

I held out until Monday morning first thing (7 am UK time). At this time, I could still not place an order online and the market had moved to 1184 (give or take). I called the broker to be told that the market was limit up and that is why orders were not being taken online. However, I could place an order on the phone.

I was now faced with two dilemmas:

1) Do you enter the trade? If so: at market or do you place a limit order somewhere closer to where the planned purchase price was? From all the literature I think the right answer is - NOT TO MISS A TRADE, so enter at market. Well I did not. I entered a limit order at 1150. In isolation I knew this was wrong at the time but I made this call due to dilemma 2 below.

2) With the market at 1184 and my stop fixed at 1026.25, my risk had effectively doubled or I had to halve my trade size (give or take). (As I am using spreadbetting it is simple math: the difference between planned entry and stop vs. potential entry and stop divided into the amount of capital I would like to risk per my risk algorithm.) Thinking of another golden rule (ie money management rules always take priority in order to stay in the game) I was faced with another problem. If I reduced the position size to match the risk I could tolerate, then the position was too small for any of the spreadbetting brokers.

Not a very sophisticated solution then emerged in my brain. Settle somewhere in the middle: increase the trade size to the min allowed by the broker but do not take all the risk on by buying at the market. Put in a limit order at a price that is acceptable to your risk level and let the market decide if it fills you. This way I thought prudent money management would take precedence over following ALL system signals.

The truth is I still risked 50% more than I planned on this trade as I set the limit price at 1150 and not at 1100, the price at which my risk limits were ok.

After having this little mental and emotional gymnastics I jumped on a plane at Heathrow to land in Philly 8 hours later (2pm US time). I could not wait to get to an Internet connection to see how the market had treated me. I got filled at my limit. So I complied with my system trade signals and made the trade but took on more risk (3% instead of 2%) than I should have.

I recall a point c.f. makes in his book about judging a decision based on its outcome as the wrong way to think. One should rather judge a decision based on the quality of the thinking/process/discipline at the time.

The reason I put this on this board is to get some discussion from some of you seasoned pros on the thought process you would have gone through. I am not looking for personal validation at all, so please be as brutally honest as you like. I really would like to hear how you would have handled this situation. For what it is worth I think I should not have taken on more risk. I should have set my limit at the level of risk I could tolerate end of story. BTW I would have got filled on that trade too but that is not the point I think.

RedRock
Roundtable Knight
Roundtable Knight
Posts: 939
Joined: Fri Jan 30, 2004 3:54 pm
Location: Chicago

Post by RedRock » Mon Feb 11, 2008 7:32 pm

Every trader must weigh the risks on both ends and make a decision which he or she can sleep comfortably with. You have done this, so ... If you had missed the trade and it went LMT for the next week..., you would have kicked yourself hard. If you entered at the highs, and it sold off, you may have kicked yourself hard. Decide that kicking yourself hard feels bad, and decide to be comfortable with whatever decision you made. Perhaps some backtests would yield what has worked best in this scenario in the past. Presumably, you have done this and based your decision on the best information available to you. So, good job. What I or others may have done, must not matter.
Last edited by RedRock on Mon Feb 11, 2008 9:28 pm, edited 1 time in total.

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Mon Feb 11, 2008 7:52 pm

I would not criticize your process, which seems to be
(A) Follow the system without fail
(B) When faced with a difficult decision between X and Y, choose (half X) plus (half Y). That way you can only be half wrong, which will preserve your mental health.

Since the betting agency wouldn't let you bet at half their minimum unit, you were forced to choose between (X = larger risk) and (Y = skip the trade). One way to have done this would be to flip a coin. You did it another way, by trying to guess how other people (traders you admire) would have behaved in the same situation.

I might suggest you invent a few different trading rules that handle situations where you can't afford to trade the minimum unit. Then put each rule into your system, one by one, and backtest each of them in simulation. The difference between the best results on the best rule, and the worst results on the worst rule, may give you some insight into the extra uncertainty you will experience because you have a small account that can't take every trade. Some rules you may wish to consider, might include
  1. The first time this occurs, round the position size up. The second time, round the position size down. The third time, round up. 4th, down. 5th, up. etc.
  2. If (total risk among all positions > some_number_Z) round down, otherwise round up
  3. Flip a coin each time. Heads = round up, Tails = round down
  4. Let January=1, February=2, etc. If today's month number is odd, round up, otherwise round down.

Stephen Newton
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Wed Dec 31, 1969 7:00 pm
Location: London, England and Cape Town, South Africa

Post by Stephen Newton » Tue Feb 12, 2008 4:29 pm

Red Rock and Sluggo thanks for your thoughts.

I agree traders must be able to live with themselves, but I am also a great believer in learning the right behavior and the right way to think. If you have played allot of sport you will know that even the most gifted athletes may not make it. Some have the right mental attitude naturally ... Tiger Woods springs to mind. Others are great golfers but just cannot do the mental thing without constant training and coaching - aka Ernie Els, Phil Mickelson etc... Like Ernie and Phil in golf this applies to me in most of my life’s endeavors unfortunately... I have to keep working hard at learning and re-learning the stuff I already know. Sometimes new approaches and methods are what makes it stick for you...

Btw - the trades stopped out today!

I have been playing with some other thoughts on the whole spread betting vs. 'proper broker' thing. I use spread betting as the gains are tax free in the UK and I can get smaller than 1-contract size positions. I have done some analysis over the 30 trades my system has taken since Oct (clearly not a statistically relevant sample yet but I have what I have). I see that when I compare my position leverage (ie the amount of capital in the market for the position I own) spread betting gets me on average 70% of what the actual contract on the exchange would get me. This is particularly bad for the Forex trades with some of the spread bet positions being as low as 20% of the leverage of a contract position. This says to me that my returns will be 70% of the back tested average (Assuming back tested average plays out in real life - and I know all the shortcomings with this assumption!).

Mentally this feels ok as I get the benefit of not having to pay 40% tax. However I was wondering if there may be some advantage to placing certain trades with spread bet houses (ie grains, interest rates, metals etc) as these seem to get even better leverage on spread bets than buying the contract, while placing FX trades at the CME exchange to get the leverage advantage there. The risk is the same if you fix your stop and bet size so in my mind you should get as much leverage as you can for your fixed risk. Has anyone played with this kind of 'broker diversification'? I guess the only way to test this would be to trade it live and accumulate experience of the best leverage options and adjust accordingly. Still interested in thoughts observations etc…

RedRock
Roundtable Knight
Roundtable Knight
Posts: 939
Joined: Fri Jan 30, 2004 3:54 pm
Location: Chicago

Post by RedRock » Tue Feb 12, 2008 6:22 pm

Is it true that spread bet shops honor your stop price with no slippage even if the pit market would gap past that price? Not sure where I heard that but it would be an advantage. The ZERO TAX factor is pretty interesting as well. Of course we have no such animal here in the states.

7432
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Wed Dec 31, 1969 7:00 pm
Location: sandy, ut

Post by 7432 » Tue Feb 12, 2008 7:58 pm

I'm not familiar with the spread betting operation, so they might offer this option for your limit scenerio anyway...
but if you were trading the actual wheat contract with a proper broker and the month you wanted to buy was limit up you can always buy the back month and do the calander spread. you'd be surprised how reasonable and active the spreads can be.
in the case of entering a trade you still might be chasing, but in the case of an exit, you're only option could be to trade a back month that is not limit and do the spread. at least you'd be out.

on your assumption that your returns will be 70% of your backtested average, don't your drawdowns stay the same?
and doesn't this make your system look quite worse?
or does the 40% savings at the end of the year make up for it?

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Wed Feb 13, 2008 3:01 am

Spreadbetting companies will offer a "guaranteed" stop loss if you want it, however you pay for it with a wider spread in the price. Personally I don't use guaranteed stop losses as the spreads are wide enough already and gaps are quite rare, but the extra protection is available if required.

The no commission no capital gains smaller positions part is attractive but it can be difficult to match the fills in TB with real life.

Like everything in trading it's always a compromise!

Stephen Newton
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Wed Dec 31, 1969 7:00 pm
Location: London, England and Cape Town, South Africa

Post by Stephen Newton » Thu Feb 14, 2008 4:33 pm

RedRock I think Turtle40 answered your query on guaranteed stops for spread bet brokers. I also don’t use the guaranteed stop as it is too expensive in spread terms. Of course the Zero Tax scenario is helpful (if you are making profits!). The reverse is true if you are making losses (ie you get no deduction). Therefore in a mark to market tax regime (like the US) you would get the benefit of the tax credit if you were in drawdown at a tax year end using a 'normal' broker. You would not get this using a spreadbetter in the UK. Compromise again!

7432 – I first thought Drawdowns would be the same. Therefore why would you trade a lower leverage broker (ignoring the tax benefits). On reflection however, they are not necessarily the same as drawdowns occur as a result of the following two reasons:
1) Trades stopped out at your predetermined stop loss level
2) Open profit 'lost' as price changes against the trend you are trading (that could be recovered again if the trade is not ultimately stopped out)
If all drawdowns consisted of 1) above then the drawdown would be the same under a scenario of 100% leverage or 70% leverage. However with 1) and 2) applying clearly it is not. If the drawdown is largely as a result of 2 then of course drawdowns would be reduced by up to a max of 30% (using my numbers).

Unfortunately I do not have enough live history to be able to tell you definitively how it turns out in actual trading. Notwithstanding if you are trading a system that has 70% losers (long term trend following) then allot of your drawdown must be attributable to 1 over the long term.

From your question, I think you and I are on the same page. My logic is that (assuming a large proportion of any drawdown is as a result of 1 above) then for 'largely' the same risk, I don’t get the same return if my leverage is only at 70% (again ignoring tax)

However the interesting thing is that on some markets the leverage with spreadbets can be up to 300% that of the exchange traded contract. So I was pondering whether some mix of execution as follows might deliver superior returns over the long term?
1) For grains, metals etc (with leverage over 100% of the exchange-traded contract) take the trade at a spreadbet house;
2) For forex (where my experience is the leverage ranges from 30% to 70%) take the trade on the exchange

The other thing that puzzles me is why the leverage would fluctuate this much. My understanding of the trade size calculation for an exchange traded contracts means that for a given dollar amount of risk you can afford to buy a certain number of contracts with a stop set at a certain fixed number of points away from the opening price.

For that same given dollar amount of risk and the same stop level why would the leverage be any different buying a number of $ per point (calculated by taking the dollar risk you require and diving that by the difference between the opening price and the stop)? I have thought of two reasons:
1) Slippage
2) Spread of the spreadbetting house
Both of these reasons do not account for the difference by my calculations. I need to do some more analysis on this as I am not comfortable just accepting this difference. When I get an answer I will report back here but if anyone has some insights I would appreciate hearing them.

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Sun Feb 17, 2008 4:03 am

Sometimes there are market conditions when the floor accepts MARKET orders only. It happened during Katrina event on NYMEX, I was trading Natural Gas options back then and filled at the market. If Limits are accepted then I'd go for limits during the day session or At The Close order, again what if there's not enough sellers when you wanna buy or not not enough buyer when you wanna sell? That means that your At The Close order is not guaranteed for a fill. All depends how STRONG do you are biased about the market's direction. If you were to trade Silver March 1980 contract from the long side when Hunts cornered the market in Jan 1980 and wouldn't get filled that'd hurt right? But what if you were to trade Silver May Contract 1980 during March of 1980? I liked the RR's response with a "can sleep risk limit". You could also hedge yourself with options..

freeman
Roundtable Knight
Roundtable Knight
Posts: 114
Joined: Tue Oct 04, 2005 7:12 pm
Location: Switzerland

Post by freeman » Thu Feb 21, 2008 5:32 pm

If u trade daily bars try to enter at the best price the next day by using a intraday system to enter your long position...eg ur next long signal on a 15m or hrly,use your intraday stop for the day,u can then start to merge your longer term stop in by start stretching your stops futher on time frame up,like 15,hr,2hr, 4hr daily until it matches your original stop.

This is just a quick note for you, it will give u somthing to think about :twisted: .

Psychology :?: mental health :?: whats that :?: ,you cant be human in this game, its better this way :wink:

Good Luck

BARLI
Roundtable Knight
Roundtable Knight
Posts: 650
Joined: Sat Jan 17, 2004 6:01 pm
Location: USA

Post by BARLI » Tue Feb 26, 2008 5:49 am

Stephen, I'm curently reading a book of Mark Ritchie "God In The Pits" and he explains all those moments of limit up, limit down moves of Silver madness in 79-80. Good read, I recommend this book

http://www.amazon.com/God-Pits-Confessi ... 0964695227

Stephen Newton
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Wed Dec 31, 1969 7:00 pm
Location: London, England and Cape Town, South Africa

Post by Stephen Newton » Tue Feb 26, 2008 6:56 am

Barli - Thanks I have placed this onto my list of books to order from Amazon.

Post Reply