Question about money management

Discussions about Money Management and Risk Control.
Post Reply
rs

Question about money management

Post by rs » Sat Apr 19, 2003 11:26 am

Hi All,

I am trading in a long term trend following style on global futures markets and have dones so systematically for just over a year. I have a question for anyone who would like to answer about the initial risk/ pyramiding side of a trade.

1) The trader decides that total risk on a trade is 2% and uses 2N (where N is as described in the original turtle trading rules) as an initial risk stop.
2) The position moves against the trader by 1N the next day and the trader exits.
3) The trader does this consistently on all positions entered.

The question is : Is this not in effect pyramiding since when one loses, one only loses 1% initial risk per trade, but when one hits a winner one has twice the position size on? I would also be interested to know from the experienced traders on the forum as to whether this is a valid method of position sizing and in effect, pyramiding.

I hope I have explained myself clearly! If not, please let me know and I will try again.

Thanks

rs

Forum Mgmnt
Roundtable Knight
Roundtable Knight
Posts: 1842
Joined: Tue Apr 15, 2003 11:02 am
Contact:

Post by Forum Mgmnt » Tue Apr 22, 2003 6:23 pm

rs wrote:Is this not in effect pyramiding since when one loses, one only loses 1% initial risk per trade, but when one hits a winner one has twice the position size on?
I don't think this is pyramiding at all. I view pyramiding as scaling into a position. Adding some contracts and then adding more when it goes in your favor.

What you have is a different system with a different risk level and a different stop. Be very careful about making these sorts of changes without doing some serious testing. Changing the stop can mean the difference between being winning and losing.

CRASS COMMERCIAL PLUG: VeriTrader was designed specifically to answer these sorts of "what if" questions without requiring any programming.

rs

Post by rs » Sun Apr 27, 2003 3:59 am

Hi Forum Mgmnt,

Thanks for the reply and I don't think I said it before but many thanks for setting up this forum. It is definitely of great quality and very helpful.

I see what you are saying. I guess what I meant was that within the context of a successful system using this position sizing approach, it was a way of having a larger postion on when you have a winner. Of course, you would need to test the efficacy of this to be sure that it works.

Thanks

rs

Learner Trader
Contributor
Contributor
Posts: 4
Joined: Mon May 12, 2003 9:01 pm
Location: Karachi, Pakistan

Question on leverage & pyramiding

Post by Learner Trader » Sun May 18, 2003 1:14 am

Well its a very awakening experience browsing through the forum. Its turning out be a good teacher.

I wanted to ask is that how do we know how much "leverage" to use or if what I understood what should be the funding of our notational Account.

Should it be static like "2L" or dynamic and changing for every new trade taken. Or pyramiding takes care that u leverage winner and vice versa for losers. How can we come up witha proper notational account.

Thank You for reading and giving your valuable time.

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Sun May 18, 2003 2:16 am

Generally speaking your bet size is determined by your equity at the beginning of each new trade (dynamic). Depending on their systems different traders will use:
- the open trade equity (total equity reported by your broker)
- the closed trade equity (OTE less the equity of open trades from the entry point of each trade), or
- a hybrid (OTE less the equity of open trades from the stop for each trade).

OTE is frequently used because is simple to calculate (read your statement) but the hybrid gives a more accurate assessment of real risk.

John

Learner Trader
Contributor
Contributor
Posts: 4
Joined: Mon May 12, 2003 9:01 pm
Location: Karachi, Pakistan

Thanks John & to check i got it right

Post by Learner Trader » Sun May 18, 2003 3:22 pm

Well what a quick and informative reply. Thanks John. :)

What I understood from your reply plz correct me if I am wrong is that that ones's betsize should be based on equity in one's account.

You generously explained all the three types of equities.

1. Open Trade = Total equity in your account both in cash equity in any trade. ( would be the largest estimation) or (it shows how much the trades have gone to one's side)

2. Closed Trade = Open equity minus open trades at entry price or simply what i understand is the equity one has before one puts on trades. Or equity u start out with. (

3. Hybrid= Equity minus the price at which stops are placed or expected equity when things goes against one trades.

I believe basing our bet size on Hybrid would a conservative but effective strategy. while basing on Open trade would be aggressive . the middle one the Closed trade equity and should be ( you may object or advise against) use when making betsize education.

Well betsize will be based on equity but how I am ascertain my Equity. for instance I have 50,000 but could twice leverage and get to 150,000 on margin. How would I calculate that when I have to leverage. To leverage, not to leverage what I am getting to know is more or less discretionary upon how presently trading is going.

In this regard is it better to decide leverage to used on annual , bi-annual monthly basis , how your account has done or we can fix the leverage annually and multiply by equity left after every month and follow the turtle rules on going 20 percent down on equity and so on. So we not over stretch ourselves.

About pyramiding I have heard that one should always buy lesser contracts or number or shares then initially bought but as turtle rules tell us it can be equal to the size of the preceding position,

Well sorry again for such a long reply, i have a problem in condensing my idea, need to work on them

Thanks again

Rai Omar

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Sun May 18, 2003 6:22 pm

Rai,

I confused you slightly by not giving an example. The hybrid is less conservative than the Closed Trade because as you move your stop up your risk declines and you have more money to bet with (with the CT approach you don't get more money even though you moved your stop up) but I'll give you an example using a 10% bet size (in real life 2% is a better number but 10 is easy to calculate with):

You start with 100,000
You bet 10,000 ( this is the risk to your stop)

At this point if you were to make another bet all options (O,C,H) would bet 10% of 90,000 because your equity by all measures is now 90,000.

After 5 days the bet has moved your way 20,000 so you move your stop up 10,000 to break even. This means that your:
OTE is 110,000
CTE is 90,000 (set at the beginning of the trade/doesnt change)
Hyb is 100,000
so the next bet would be 11,000, 9,000 and 10,000 respectively.

That should clarify things a bit. Obviously if the trade carried on your way the OTE will move up and when you tighten the stop up the Hybrid TE will also move up but the CTE doesnt change.

To add complication, some will bet higher on equity over their starting equity at the beginning of each period using the argument that they are now betting the markets money.

Your equity is your total money assigned to trading this system. If you are trading futures you get 97% margin in some cases so you have to be very careful not to get this bit wrong (buy Van Tharp's book, "trade your way to financial freedom" to get a full understanding). So if you have 100,000 then 50,000 might be in your brokerage account, 50,000 might be in the bank in case a drawdown requires it. BUT you bet size (the risk you take on each bet) based on 100,000. The risk is usually the difference between the entry price and your stop times the number of shares or contracts bought.

You can recalculate your equity as often as you find convenient. This depends on your tools.

Yes, pyramiding is normally on a smaller number of contracts after the first bid but can be the same. If the first bid was a test only then you can increase but unless you really understand what you are doing I'd stay clear of this.

Good Luck. To improve your luck by Tharp's book!!
In fact buy all the books in our reading lists as they are cheaper than a single currency futures loss.

viewtopic.php?p=1046#1046

John

Learner Trader
Contributor
Contributor
Posts: 4
Joined: Mon May 12, 2003 9:01 pm
Location: Karachi, Pakistan

Thanks to mentoring Round Table Knight

Post by Learner Trader » Wed May 21, 2003 6:59 am

Dear John thanks again for a quick reply and sorry for my late reply.

The example and details you gave cleared my concepts


To add complication, some will bet higher on equity over their starting equity at the beginning of each period using the argument that they are now betting the markets money.
Well since a person makes his betsize calculation on equity he has if he doing well his equity gets added so I believe he should utilize the equity at hand. Although what I have heard it is prudent to set aside some part of gains aside. You are right it all comes to the system you are trading and equity you have.

You can recalculate your equity as often as you find convenient
I would really appreciate if you can elaborate albeit if you have the time what in your opinion should be the time to recalculate ones equity would it be better to do it on a on a daily basis or I am using the Hybird measure I should calculate my equity after every trade I close.

Actually I am not trading futures, and would not trade them for some time due to fact that my present equity is not sufficient as I believe I could not trade multiple contracts.

I am trying to replicate with some success a trend following system (based on Turtle Rules) in the local stock market. I believe the Hybird measure of equity will be a better since it most accurate as you have rightly explained.

I have heeded to your advice and have put orders for the books on Amazon particularly Tharp’s Trading for a Living. They will reach in a few weeks time. You are right education is less expensive than the smallest potential loss.

Thanks Again

Rai

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Wed May 21, 2003 7:25 am

Time to recalculate would be after every trade or every stop adjustment if it is easy and not too time consuming. I have a system where I recalculate every 30 minutes but thats because the trade decision making process generates an equity calculation.

If I was having to do it manually I might recalculate on Saturdays or even once a month. Why not try doing it every trade to start with and then drop back to a lower frequency if its too much trouble. If you get a big shift in equity (win or loss) then do the recalculation immediately :-)

If you want to trade futures do a search of this site for spread betting or spreadbetting as the British firms offer you access to virtually all futures markets (and major US, UK and European stocks) at GBP0.50 per point. Some of the futures (look at jy and sf) are much better trenders than most stocks.

Enjoy the books.

John

BUNTER
Contributing Member
Contributing Member
Posts: 6
Joined: Thu Jun 19, 2003 3:12 pm
Location: WILTSHIRE, ENGLAND

Percent Volatility

Post by BUNTER » Thu Nov 20, 2003 5:17 am

Assume I have two stocks and a £100,000 account.

Stock A Price 200p ATR= 10p
Stock B Price 150p ATR= 12p

Stock B is more volatile at 8% than Stock A at 5%.

Using a 1% volatility I can calculate the number of shares to buy as follows

Stock A 1% of £100,000/10p= 10,000 shares at a cost of £20,000

Stock B 1% of £100,000/12p= 8333 shares at a cost of £12,499

Less money invested in Stock B as it is more volatile.

However each position represents a large % of my total account, 20% for Stock A and 12.5% for Stock B.

Yet I often read only 1% of equity should be risked on each trade, what am doing wrong in the calculations.?

Any help gratefully received. This is an excellent website! After being a professional fund manager for 15 years, this site has radically changed my thinking. Thanks to everyone.

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Thu Nov 20, 2003 5:38 am

Hi Bunter,

Interestingly you have defined the volatility and the capital invested but you have not defined your risk. Your risk is the gap between your entry price and you exit (incl average slippage and commission imo) multiplied by the number of shares you take.

Volatility doesnt tell us this. We would need to know your stop.

Let us say that entry-stop+expenses+slippage = 0.20 in both cases. Your risk is (10,000+8,333)*0.20 or gbp3,667 which is 3.7% of your total account somewhat over 1% per trade.

Does that make sense?

John

BUNTER
Contributing Member
Contributing Member
Posts: 6
Joined: Thu Jun 19, 2003 3:12 pm
Location: WILTSHIRE, ENGLAND

Post by BUNTER » Thu Nov 20, 2003 5:52 am

Many thanks Kiwi,

I would use a 2ATR stop for each poistion.

Referring to 1%, I have read that each trade should not be more than 1% of equity, in this case £1000.

Hope this helps.

Regards,
Bunter

Hiramhon
Roundtable Fellow
Roundtable Fellow
Posts: 98
Joined: Fri May 09, 2003 12:45 am

Post by Hiramhon » Thu Nov 20, 2003 8:52 am

Kiwi, Bunter: there's a guy who could really use your help. Take a look at this message by Joe, it sounds like your most recent exchange of messages here fits his problem exactly:
http://traderclub.com/discus/messages/1 ... #POST15922

enigma
Roundtable Fellow
Roundtable Fellow
Posts: 99
Joined: Tue Apr 29, 2003 6:56 am
Location: UK

Post by enigma » Thu Nov 20, 2003 11:09 am

The point that Joe made is similar to the problem that I highlighted quite some time ago about Van Tharp's position sizing example with %-vol.

viewtopic.php?t=377&highlight=van+tharp

c.f.' paper already show how to use the %-vol method together with the fixed-fraction method. Generally, if we use ATR for the volatility measure, to convert the %-vol algorithm to a FF method, the stop loss strategy needs to be:

Entry Price - (b/a)*ATR

where b is the fixed fraction and a is the %-vol.

If the stop-loss strategy is something different, b will be dynamic, which I believe is the case in Tharp's TYWTFF example. I think he did mention somewhere in the book about this issue... can't remember.

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Thu Nov 20, 2003 4:48 pm

Hiramon,

I have not bothered trying to find my way past the russians recently. Chuck's garden has become too wild to waste time in.


Bunter,

OK. With a 2atr stop you get 22p on the first and 26p on the second (2p slippage and commission in each case). So your risks are:
10000*22 = 2200
8333 *26 = 2169

This is roughly 2% of your equity being risked on each trade. 1% is conservative. If you want to only risk 1% then you need to roughly halve your position sizes.

John

BUNTER
Contributing Member
Contributing Member
Posts: 6
Joined: Thu Jun 19, 2003 3:12 pm
Location: WILTSHIRE, ENGLAND

Post by BUNTER » Fri Nov 21, 2003 5:04 am

Many thanks for the replies.

I now understand the % risk to total equity, however the amount invested in a position as a % of the total portfolio would just be too high. You need to be able to sleep at night.

If I were to run a 30 stock porfolio, I would invest 1% of the portfolio in each trade and as the price advanced would a further 1% per 1 ATR, and then another 1% per another advance in ATR to a maximum of 3 trades.

I could set the stop loss on the first tarde at say 2ATR to give the trade some leeway to make money, and with further additions move up the stop loss.

I know this is fixed fractional trading but the risk is controlled and it does allow pyramiding into a profitable trend.

Does anyone have nay thoughts or criticisms.?

this is a real life example, I have been asked to put forward a proposal for a maximum 30 stock fund which seeks positive return, and is not tied to any index.

BUNTER
Contributing Member
Contributing Member
Posts: 6
Joined: Thu Jun 19, 2003 3:12 pm
Location: WILTSHIRE, ENGLAND

Fixed Fraction versus % volatility

Post by BUNTER » Fri Oct 22, 2004 10:08 am

Could someone please explain how to convert the % volatility position sizing to fixed fractional. Someone posted that c.f. had written apaper on this but I cant seem to find it.

Risking only 1% of equity sometimes means that I have to invest over 10% in one single stock position which for me is a bit rich.

Any help apprecaited

Jake Carriker
Site Admin
Site Admin
Posts: 1493
Joined: Fri Sep 12, 2003 10:32 am
Location: Austin, Texas

Post by Jake Carriker » Fri Oct 22, 2004 12:30 pm

Hello Bunter,

I will try to clearly outline the two betsizing strategies (commonly called % risk and % volatility) that you are asking about.

First, % risk - To size positions using this method you need the following inputs:
1. a fixed percentage of trading capital that makes up the numerator of the expression. "A fixed percentage" means just that, an arbitrary number that you choose as a risk amount per trade. In this example let's call it 1%. "trading capital" also needs to be defined. It can be one of several different numbers such as total account equity, total equity less any existing "risk", total equity less the amount of capital currently commited to positions, etc. Again, in order to keep things simple, let's pretend we are working with total equity (the net liquidation value of the account ignoring commisions and slippage to close out existing positions). Let's assume this value is $1,000,000. Therefore, the numerator in our position sizing equation is 1% of $1 million, or $10,000. Let's call this number "Account equity at risk per trade".

2. Entry price - This is the price at which your entry order is filled. Actually, since we like to size positions BEFORE we are filled, this represents our assumption about the price we might be filled at, or our desired order price. In this case, let's assume it is $50.

3. Stop loss price - This is the price at which we intend to exit the position should it go against us. Let's assume this price is $45.

Using these three pieces of information we are able to size the position, but first let's manipulate the Entry Price and Stop loss data to produce a number we will call "$ risk per share" (or risk per contract if trading futures). This is the denominator in our sizing equation. Simply take the difference between the Entry and Stop prices (for shorts the entry price is lower than the stop price, of course). In our example, the risk per share is $50 - $45 = $5.

Now plug the numbers into this equation: Account equity at risk per trade / $ risk per contract = position size.

Simple enough. Our example gives a position size of $10,000 / $5 = 2000 shares. Notice that we are making several assumptions in this calculation and that our $ risk per share is actually variable due to entry and exit prices that differ from our assumptions. However, what we basically have is a defined way to size positions such that the amount of capital exposed to loss on each trade is approximately equal.

Note that you control the position size via your stop placement and the fixed percent of equiity you choose to risk. Closer stops = larger positions and vice versa. When using this position sizing method, it might be helpful to have a consistent method for setting stops, otherwise you might take a very small position that has virtually no chance of being stopped out, or a very large position that is well within the "noise" range of the timeframe you trade.

The position sizing method that most people refer to as % volatility helps to solve the stop placement issue, but introduces some other potential problems. It is exactly like the % risk method, with the exception of the denominator, though you might find that you need to use a smaller fixed % number in the numerator to attain similar volatility to the other method.

With this method we are going to substitute some kind of volatility assumption in place of the $ risk per contract of the % risk equation. There are an unlimited number of ways to try to describe volatility, but for the sake of this example I will use the one most commonly used in a % volatility sizing scheme, which is ATR.

If we measure the average true range of the trading instrument over a certain time period (let's say 20 days), we can come up with a "daily volatility" number that represents how much we can expect an instrument to range on an average day. By multiplying the average daily move by the point value of the instrument ($1 for one share of stock, differing amounts for futures) we now have "average daily $ volatility". For the sake of example, let's say that our $50 stock has a $ volatility of $3. That is, on any given day price is likely to experience a $3 true range.

If we use this number in the denominator of our previous example, we get $10,000 / $3 = 3333 shares. Notice that our position size is no longer dependent on stop placement (though stop placement might usefully pay attention to volatility), but it is dependent on our assumptions about volatility. Also note that the two position sizing methods are not equivalent. It does not make sense to ask what fixed fraction is equivalent to what volatility fraction, etc. Depending on your volatility measurement and your stop placement, the two methods can be manipulated to give practically the same sizing output or vastly different output.

The point of % risk sizing is to expose a fixed percentage of your capital to the risk of loss on a given trade. The point of % volatility sizing is to expose a fixed percentage of your capital to daily volatility on a given trade. % volatility does not have any information about your stops.

You mention that when trading stocks, your sizing algorithm sometimes indicates that you put a rather large fraction of your account equity into a position in order to obtain X% risk or X% volatility. This is often the case with stocks, and it is a major factor in the design of a stock trading methodology.

You might want to think about how much you trust the risk and/or volatility assumptions of your sizing algorithm if you are feeling that position sizes are "a bit rich". The implication is that you have reason to believe that you are actually risking much more than your % risk figure when you take a position. For example, if you are risking about 1% of equity per trade, and you are comfortable with this, why should it matter whether you have to deploy 3% of your available funds or 30% of them in order to achieve that risk profile? Obviously, there are issues with being able to take new trades if your capital is all deployed into existing positions, but your concerns do not seem to be of this nature. Rather, you imply you are uncomfortable with the actual size of the position in terms of percentage of your total trading funds.

Certainly, there may be good reasons to be uncomfortable with a large chunk of capital being commited to a single position. We have all seen stocks cut in half on an earnings or other news announcements. If this is the case, you might want to revise your risk / volatility definitions in order to find a way to size positions that you are comfortable with and that also meet your portfolio heat goals.

Walking the fine line of how to control risk, maintain diversification, take as many trading signals as possible, and still take meaningful position sizes with a stock trading system is a topic for a whole different discussion. I wish you the best of luck in your trading endeavors.

Hope this helps a bit.

Best,
Jake

BUNTER
Contributing Member
Contributing Member
Posts: 6
Joined: Thu Jun 19, 2003 3:12 pm
Location: WILTSHIRE, ENGLAND

Position Sizing

Post by BUNTER » Fri Oct 22, 2004 5:51 pm

Many thanks Jake for that clear explanation.

You are right that it should not matter what % of total account is invested in a stock as long as the risk is just 1% od the total account.

If I want a maximum of 20 stocks in the portfolio (which according to EMH will diversify away the specific risk) I will have to experiment with the parameters of the system, perhaps lowering the at risk percentage to say 0.25% of total account to lower the amount invested in any particular stock.

Once gain many thanks for your help.

Post Reply