Can anyone help with Percent Volatility Position Sizing?

Can someone help me calculate a [B]Percent Volatility Position Sizing[/B] for a 100.00 equity, risking 2% of the account per trade? … Really any example interpreted into pips would be useful here for posterity …As far as I can tell to this point, as I don’t see it anywhere! This really surprises me!

I have garnered this from Trade Your Way to Financial Freedom by Van K. Tharp. Most to all of the examples I am finding of this are of examples involving stocks.

[B]Position Size = Equity X Risk % / ATR ???[/B]

Here isan example from the renowned InformedTrades.com folks …but it is applied to commodities.

Hi Artaudo,

You question is a really important one that should have been answered weeks ago. I stumbled on it tonight.

As you probably already know, the concept of percent volatility position sizing is a Richard Dennis and Bill Eckhardt concept that predates Van Tharp’s book[s] by some 20 years, though Richard and Bill called it by a different name.

Now as for your specific question, I am not clear about your request. Are you looking to size your risk per trade at $100.00 USD per losing trade in forex? Once you clear this up, we can get to the answer to your question. Keep in mind that choosing a specific dollar amount to stop the trade in this way may actually violate the whole concept behind using volatility to decide how to correctly position size. Using the ATR to determine the correct position size we dictate that the risk per trade be decided by other factors like account size and percentage risk. This information would then tell you if $50, $100 or $200 is the correct number to risk. Maybe I am jumping to conclusions about your question and will await some clarification from you.

As for the fact that you don’t see a post covering this subject is sad and scary. It probably says a lot about the lack of sufficient discussions around risk management. Most people talk of a 1 or 2% stoploss deciding this is all that is necessary as their risk management system used in any given trade. But stopping at only risking 2% doesn’t tell you how many contracts or lots to buy. There are countless articles about position sizing that are unclear as to what position sizing is and how it works. Here on this site in the school section the writers seem to not address your specific question on correct position sizing, but they do have a useful software tool for position sizing that doesn’t use ATR. They’ve approached the question of position sizing strictly from an account size perspective, though the final number of lots results in the same number of lots that I’d get using ATR if I give the software a dollar risk number that never changes. By the way the same formula works for forex to. You need only do an extra few steps to convert the forex prices to pip values instead of tick values in commodities.

Regards,
Bklynborn

Hello,

Well OBVIOUSLY every morning I look forward to reading your (bklynborn) posts given our ‘common ground’ and your experience in ‘the biz’.

Good points made by you (of course).

I feel that I need to add to this though:

I’d be very careful with the formula detailed i.e. taken out of context that formula could very easily get a new trader into ‘hot water’ and here’s my reasoning:

I would say that this formula is largely system dependant (I’ve read Von Tharp’s book too but I cannot remember the specific context in which it, the formula, was detailed).

Here’s an example:

Let’s assume a $10 000 account and 2% risk per trade. The monetary value of your risk is therefore (supposed to be) $200 per trade.

Looking at EUR/USD (and using Wilder’s Swing Index System as an example): a short signal was generated on 08/11/2010 @ 1.3882 (or thereabouts i.e. it would depend on your ‘order offset’ as to how close or how far away one would place your order from the low of the signal bar). The initial stop (as per the said system) would have been @ 1.4288 (or thereabouts as previously explained). The risk on the trade IN PIPS would therefore have been 406 pips. ATR(10) on the day of the signal was 0.0177 (I’ll explain WHY ATR(10) and not ATR(14), the default, a bit later but that’s really ‘neither here nor there’ right now).

Now let’s solve the formula:

Position Size = Equity X Risk% / ATR

Position Size = $10 000 X 2% / ATR

Position Size = 200 / 0.0177

Position Size = 11 299 units rounded DOWN to 11 000 units

Now a Position Size of 11 000 units would equate to $1.10 per pip movement. Multiply that by the risk IN PIPS and your potential loss on this particular trade would be $446!!! That’s more than double your 2% risk managment figure.

The reason that this formula works with the Turtle Trading System is because the stops are ‘based and spaced’ on the value of (2 X) the ATR value. This formula would cause some other trading systems to ‘self destruct’.

All I’m saying here is that stuff like this cannot be taken out of context. I’m starting to see of late that this thing of leverage, risk management, and position sizing is possibly the most misunderstood concept of all in this business.

I believe that what I’ve explained above has been addressed by bklynborn. However: I thought it necessary to clarify further is all.

Regards,

Dale.

Edit:

By the way: since you and I have been discussing Wilder again I have to say that I’m starting to ‘fall in love all over again’ with his ‘stuff’ (particularly as now, after a good long while, I UNDERSTAND what he was ‘getting at’ and it now ‘fits like a glove’). Up until now it’s only his Swing Index System that I’ve concentrated on (and from my looking at Donchian and the Turtle’s again in the past day or two, due to our ‘banter’ on that other thread, it’s a far better system unfortunately in my opinion) but after reading the initial posts above I’ve taken ANOTHER look at his ‘Commodity Selection Index’ and I must say that I believe that it’s ‘sheer brilliance’. Have you ever looked at it or implemented it??? I’m thinking that maybe now it’s ‘time’ to implement it SERIOUSLY again.

Edit 2:

Oops: I nearly forgot. Why ATR(10) and not ATR(14)??? Wilder notes that one should use a period setting equal to one half of the period being monitored (not his exact words but that’s the ‘general rule’). Delta does not have ‘Sunday Bars’ on the (Delta Trading) charts and given that the (my) period being monitored is always a month: there are four weeks or twenty days (approximately) in a month and half of twenty is ten!!! LOL!!! Also: this setting improves the performance as well as increases the number of signals generated by the ‘RSI Rollercoaster’ so that’s a good thing.

Hi Dale,

Happy Thanksgiving.

You are correct. It is necessary to look at the total system and use a risk management formula correctly within that system. When used correctly, it should protect your capital as designed. The problem comes when a system is slaughtered as tends to happen when a formula gets told over and over again losing its original purpose and design. If I were trying to control risk and use proper position sizing, I’d start over and go back to the original design per Richard Dennis. His system was nicknamed the “N” volatility and had a two-fold purpose. As for forex and proper position sizing, one needs to add another step in the formula to get to the correct position size.

Regards,
Bklynborn

Hello again,

And thanks for the good wishes bklynborn. I’m sorry: for some or the other reason I didn’t even see that you’d posted a reply here until today i.e. I was looking for this thread because for some or the other reason this formula has been ‘bugging the hell out of me’ since we started exchanging ideas. And then it ‘dawned on me’ yesterday: this formula is ideal for a trading system or methodology where NO fixed stops are used!!! Ironically: this all started because as I noted previously I’ve started to ‘revisit’ all of Wilder’s other trading systems again (with a lot more knowledge, experience, and ‘broken bones, cuts, and bruises’ ‘under my belt’) and one of his systems (IRONICALLY called the ‘Volatility System’) is one of those very systems i.e. NO fixed stops and therefore no way of knowing beforehand what position size to take according to your risk management percentage (and it’s also ironic that now that I’ve recoded the system I find that two or three years ago I had the code totally wrong anyway)!!! ‘Enter’ this formula!!!

And by the way: I stand corrected i.e. it was NOT Wilder that said to use ‘one half of the period under review’ for the setting of the period parameters (for ANY indicator for that matter) i.e. it was someone like Larry Williams or Alexander Elder or Van Tharp (Wilder has always used 14 days as the default period). I have to say though that using RSI(10) and ADX(10) instead of the ‘Wilder default’ of 14 days does appear to give better results (at least for ONE of my other trading systems). I would say though: for the purposes of THIS formula under discussion the longer the ATR period used to determine risk / position size the better as let’s face it: at best this formula can only give an approximation based on the CURRENT value of ATR at the time of opening the position. I think it was for this very reason that the Turtle’s used ATR(20)???

Regards,

Dale.

Following on from my previous post on the subject (and I think FINALLY getting to the answer to the question posted by the thread starter):

For forex:

The ‘extra step’ I believe would be to work out the $ value per pip movement of the SMALLEST lot size allowable and only THEN to continue with the calculation.

For equities and commodities it’s a ‘no brainer’.

Regards,

Dale.

Hello dpaterso… and friends…

Finally stumbled upon the responses to this post, (which I initially started as artaudo, several years ago). I had posted my questions, and got no responses. It was about the time I got distracted away from the forex process, and encountered some other projects. One leads to another, and I am back, to see if I can make some progress, and hopefully apply some maturity.

Nevertheless, wanted to share some discoveries with you (all), since it seems you’d put some thought into the posts and process… As well as being curious of how you’ve fared in the application of the system, above mentioned. I also feel that this description addresses many/most of your intuitions/ruminations of your posts, above… So here’s what I’ve come up with, recently:

There are [B]two questions I want this trading system to answer[/B]

  1. How much does the Time Frame want to give (vs… I want to risk)?
  2. How do I convert this into a position size?

I decided to start with -
100 units of EUR/USD = (Cost me) $107.00

I then calculated the pip value, based on these 100 units -
((.0001 / 1.07) * 100) = [B].0093(4)[/B]

I take the calculated pip value, and find how many pips I am holding in my 100 units -
100 / .0093 = [B][B]10,752.(68)[/B][/B]

Calculating % Risk per each hypothetical trade -
at 1% risk of 10,752.(68) = 107.52 pips or $0.99(99)
at 2% risk of 10,752.(68) = 215.05 pips or $1.99(99)
at 5% risk of 10,752.(68) = 537.63 pips or $4.99(99)

Now then, as a reference, I have in mind generic ATRs, based upon generic averages of the 5min, 1hr and daily charts, which hypothetically are moving an average of 10 pips, 46 pips, and 148 pips… respectively (from, Ann Couling’s “Forex for Beginners,” chapter on Your Trading Plan). The system I was trying to hone before looked at the Intraday, 30 min and 1 hr, to take advantage of a position on a 3 or 4 hour time frame.

The dilemma I kept running into, when mulling over explanations of money management systems, is where to place a value for a stop loss/take profit, to complete the foundational parameters for a trading system. So the reason I’ve approached the calculations the way that I have, is to:

Take the emphasis off of, what is really a generic psychological parameter for any trader to wrap their heads around, which is looking at any time frame and deciding… “what I want to earn on a trade, or my risk.”

Instead, I feel this is up to the pair/time-frame…
Enter: Average True Range…

In this way, I find by starting with -

  1. Risking 1% of the account, I can (psychologically) loose 100 times, before I’m closed out
  2. I am able to start with a grand total number of pips representing my total equity, in this case [B]10,752.(68)[/B],
  3. for the purpose of descriptions, I have taken the generic pip averages per time frame above, but I can more naturally transfer those numbers with the ATRs of Real-Time-Frame-Numbers… I have [B]pips replacing $$s…[/B]

In essence, I’m able to take away forcing what I want to get from the market, in terms of $$ amounts… And replace it with an emphasis on pip values (in this case, a trade swing, I am comfortable with, [B]between 107.52 pips toward 537.63[/B]). I feel like this translates much more easier when looking at Set Ups and Time Frames… as a take profit.

So what I am proposing is, in the case of: [B]Position Size = Equity * Risk % / ATR[/B] :
a) The correlation between risk management and time frames is positive… the slower the time frame, the greater the distance any stop loss needs to be. So, ATR must be consulted, to fall within the above mentioned swing parameters.
b) Position sizing is Set-Up dependent (price action, chartology, etc), to adjust for money management rules
c) This system also favors, if necessary, cutting a position in half, and adding a trailing stop.
d) Working with Oanda Broker, which does not demand you specify lot amounts.
e) The account size is scalable, and still emphasizes (doubling, tripling, etc) the above mentioned swing parameters…