The ATR Advantage

Which is better a high average true range (ATR) or a low one? The answer - it depends on the circumstances. If you are going to day trade and you have two stocks where everything else is equal and you only want to be involved with one of them for whatever reason - pick the one with the highest ATR on the minute-basis you are playing (ATR varies with time scale). You do this because you want the most "bang" for the buck and ATR is a proxy for volatility.

But if you are planning a swing trade and you are faced with several different stocks and can't choose between them then the answer is clear - take the stock with the lowest ATR.

Here are the results of testing done with the ATR set at various levels -


Obviousely the lowest ATR has both a better win percentage and ROI. And you can't help but notice that the ROI deteriorates as the starting ATR increases.

Once more this is a volatility issue - before a stock can go up it has to stop and the ATR tells you when a stock is slowing down or "stopping".

One other thing you have to remember is that ATR is also relative to the price of the stock - that is - low priced stocks have naturally low priced ATR and high priced stocks have naturally high priced ATR. So you really can't use ATR across price boundaries.

Remember also - ATR is a guide not a guarantee - trading stocks is a risky business.

Volatility Is King

Long time readers know that I hold the average true range as the key to volatility. Any other so-called volatility such as the VIX, VXN, or VXO are pretenders in my humble opinion.

Long time readers also know that whenever I make a bold statement I try to back it up with evidence - here is the evidence -


You can see where the ATR was during the "bubble" of the 90's. It began below 600 (monthly charts please) and then slowly climbed to over 900. 900 seemed to be the catalyst for the fall because once it broke 900 there was too much volatility and not enough liquidity to sustain the price and the market fell.

We've seen this picture on this BLog before - most notably on the daily basis since the 27th of February. You can see that the ATR did not break until the market bottomed - at that point it began to fall in a significant manner and the market resumed its climb. Notice too that well before the market completed its fall that volume was cut almost in half.

Now a lot of people are going to be saying all kinds of doom and gloom crap about Dow at 13000. All you need to remember is that the only number that matters is the ATR at 505.

And I know that there are hundreds of people, maybe even thousands who would disagree with my every statement - but none of them can produce a picture like I can.

But always keep Rule 1 firmly in mind - nobody knows nothing - including me.

Optimum ATR Length

Right. That title is a bit overblown - let's just say it is optimum for my purposes.

Before we get to the results let me go over a couple of guidelines – this is based on daily close data not intra-day. That will make a difference.

I tested a lot of versions but only two showed any promise and I called them ATR y and ATR z. In the “y” version I used the approach - ATR(n) < ATR(n) 3 days ago. In the “z” version I used the approach - ATR(n) < ATR(n) 5 days ago.


Some were pretty good and some were pretty bad but as you can see for this approach ATR(10) seems to perform better. The 30-day net change was about the same for each of them so this is not necessarily a good filter for a long-term approach.

I noticed that 15 z seemed out of line so I decided to step down one period at a time from 14 to 10.


Even stranger 13 z was better than 14 z by a lot but then 12 z fell off again. So I said that the results might be date specific so I changed the start date so that February 27th would be included.


And the results fell in the same manner as before – ATR 10 is the best with ATR 13 following. I can’t explain this so we’ll just accept it as some mystery of the ATR (for the time being). Point is – ATR of a length 10 seems to be best as a short-term daily indicator.

I was surfing around looking at various sites and I happened across one site where the poster had an article suggesting using MACD with the ATR. That, to me is the best way to use the ATR – in concert with other indicators – I prefer the RSI, that guy likes the MACD. The fact is they both measure the same thing just one is more complex and a little slower than the other. Slow is good from the standpoint of a long-term relationship because slow ensures that you get on board a developed trend – but that is a matter of choice.

ATR - Potent Tool

I know - I get hold of an idea and I don't want to let it go but I was playing around this morning with a couple of filters and I had a thought on how to improve my Blow-Off Bottom set-up filter.

That filter as you probably remember only outputs stocks that meet the first two-thirds of the BOB set-up and it's up to you to watch for the third third to develop and perhaps use that to day trade (or start a new, longer-term relationship).

Anyway I added a constraint that said - and ATR(10) is less than ATR(10) 2 days ago and I back tested the modification.

The results are interesting enough on the surface -

Original BOB Set-up: 58% Win rate, 27.74% ROI
Modified BOB Set-up: 58% Win rate, 66.76% ROI

Apparently the ATR makes a difference and I'm going to leave it in - but this is the newsworthy part - when I looked at the net change over time here is what I found.

Original BOB Set-up: 4 day change = .34%, 30 day change 1.51%
Modified BOB Set-up: 4 day change = .80%, 30 day change 5.76%

That is amazing. Over double on the 4 day and nearly 4 times on the 30 day.

What this says, at least to me, is that an ATR that is high and then begins to decline is potent tool for the trader's tool box.

As always I tested the counter proposition just as a sanity check and here are the results from when I required the ATR to be increasing over the ATR of 2 days ago.

Win% = 54%, ROI = -6.28%, net change over 4 days = -.05%, over 30 days 1.44%.

That clinches the deal as far as I'm concerned - a decreasing ATR (= decreasing volatility) is better for your trading account than an increasing ATR (= increasing volatility).

Some months ago I wrote a post that provided an anecdotal indicator that when trying to choose between two stocks for a day trade and only having enough bullets or desire to trade one that you should take the one with the higher ATR. This is why. This tool will help you on all time frames. And maybe even if you do have enough to cover both you should put is all in on the one with the higher ATR.

That's your choice, of course - not an invitation to speculate in the Wall Street Casino.

RSI ATR - A Practical Example

Revisiting that filter I spoke about yesterday here is a good example of what I was talking about. This is IMCL and you can see quite clearly that there were two occasions in the last month and a half when you could have taken this stock off the RSI/ATR set-up.



And at least one occasion when you shouldn't have. You can see on that occasion what I was talking about yesterday when I said that you had to wait for both conditions to be met to be able to say that the stock is bottoming with about a 65% assurance. If you had just taken this stock from the RSI(2) < 2 flag you would have been two days early.

Note that on both of the true flags the stock went on to post a blow-off bottom and that is just one more indicator to watch for - it isn't required but it makes for a good confirmation.

And confirmation is the most important thing of all - on this chart to the left you can see one more spot where the criteria were met but the stock didn't turn. Remember this business is all about confirmation and without confirmation we don't do anything foolish. No confirmation - no trade. It did finally confirm but didn't follow through - that's why you use a stop loss. On the next flag it followed through and then some.

In a previous post this morning the one about the past being prologue I said we needed to wait for the ATR to turn around and come down before we could be sure that the index had bottomed. This is not inconsistent with how I use the ATR as a proxy for volatility. In the ATR/RSI filter I am using it to look for low volatility because low volatility begets high volatility and if the stock is at the bottom then it should go up when volatility goes up. But both tops and bottoms are marked by volatility changes. If it is high it will go low if it is low it will go high.

For those of you not in possession of a capability such as being able to draw a moving average of the ATR note here's a tip. Just look and see where the ATR is in relation to its recent past and price bottoms. If it is anywhere near where it was the last time the stock turned and went up then it is probably at a bottom.

And finally - this chart is a screen print of stockfetcher.com's new charting interface, SF2 and it is a beauty. It is still in Beta but when they get it finished it will be a joy to behold.

I don't get anything from them for the endorsement so it doesn't bother me if you don't want to improve your trading ability.

Average True Range

Sometimes the simplest packages contain some of the finest results. This filter is viable but unless you have a set of fairly sophisticated filtering software you won't be able to duplicate - simply put it tests for the Average True Range less than the Average True Range 20-period simple moving average.

show stocks where close is between 15 and 35
and average volume(90) > 500000
and Average True Range(10) < CMA(average true range(10),20)

The ROI on this filter was 71.30% with a 63% win rate and a spectacular 2.25 risk/reward ratio.

Sometimes just to ensure that it isn't just luck and especially with these simple filters I reverse the premise - for this one I switched the ATR < to ATR > and reran the test.

The ROI was -1.44% with a 56% win ratio and a .98 risk/reward ratio.

Now I don't suggest that you use this filter or anything like it as a selector for your stock picking. What I am saying is if you are looking to invest in a stock then you should probably know where the ATR is in relation to the stock's history.

I write about the INDU ATR in my wraps and it is a fairly good indicator of changes in the near term trend. Currently the ATR is sky high and it shouldn't stay there for long.



You can do this with every stock that you invest in and perhaps give yourself a little extra edge. Remember - ATR is a reasonable proxy for volatility.

Wednesday's Wrap

Wrap me up - after I found out that NTRI was reporting after hours it took every bit of my strength to not buy it to the extent of my trading account at the close - I mean who wouldn't want a 12% loss on their book three days in a row? Of course NTRI will go up at least 10% tomorrow morning. That you can bank on. (Update - up 16% after hours).

Flat - in cash - no stocks in my trading account - no holdings - nada - I'm clear. As I mentioned I sold everything including CRI before noon. I reasoned for CRI that it wasn't going to recover the losses ever and it might even go down some more since that was its general direction before I took it anyway. Right now I'd rather have the cash for trading rather than being carried as a bump in the book.

Turns out I should have held NEW a bit longer but the way I was feeling at that moment I needed a winner and that was it and I'm happy with what I got out of it. I took KNOT on a blow-off bottom but set my stop way too close and was stopped out - I didn't have my heart or mind in the trade - maybe tomorrow. I think KNOT is going to get back to where it was - it, afterall, is selling dreams and not real stuff like baby wear. Dreams are a lot easier to value. (If you detect the slightest hint of sarcasm here it's probably because there is a little in the air at the moment).

But when you get bucked off the bull you pick yourself up, make sure nothing is broken, and then you either get back on the bull or you go home and hide under the bed - I don't hide. But I did make some changes and you won't see me holding any stocks in my trading account any longer than 5 hours - if that.

Nuff about my woes - I published a new method to find day trade candidates this morning using a filter that would, if completed, find full scale blow-off bottoms. The filter found 15 matches and of the 15 - four went down today and 11 went up. Needless to say today was a good day for "going up" so I won't say that will be the norm - but I will keep track of this in the wrap and we'll see how well it does.

My three favorites were ETFC, SAY, and TTM - but because this was the first time I've tried this I just watched today. I'm going to continue using this method for awhile and see what develops.

Another interesting detail of the day is that small caps went up and then retreated back to the open - or as I say - they round-tripped. I'm not sure what that portends. Is this the age of the large caps? I doubt it - it might be as simple as the small caps are all overbought. And if you look at the IWM, IWN, and IWO charts together with the stochasticRSI(2) you see the incredible variance of the IWM (the index) and IWO (growth component) both being at 1.000 and the IWN being at 0.000! Apparently value was sold off today hard and that was enough to bring the overall index down. So that leaves the only question as why is value being sold off this year? I have no clue.

GOOG, MSFT and AAPL all went up today as did the semi's. Don't start cheering because I think that is just a one day story.

Another stock that took an amazing haircut today was GIFI. Just last week Trading Goddess was talking about that stock as if it had a real chance post earnings. I didn't think much of the stock - it appeared to me to be ready to come down anyway - but apparently the company has postponed its earnings announcement and that was enough for an 11% haircut. See what I mean about the market being so fearful? I don't think the economy is in that good shape - at least not as good as some of our leaders would have us believe. Gold keeps going up and stocks keep exploding. Although it is possible that gold is starting to peak. I think Bernake scared gold traders today when he started talking about inflation being under control - the worst thing that can happen to the gold bugs is an interest rate cut. If stocks keep blowing up I think that is coming - I'm surprised it hasn't happened already. Jimmy Crack Corn Pone also thinks one is due too.

CRVL continued its droppage today. See what happens when the day traders and scalpers abandon you - absolute chaos.

I think GM is done for awhile - third gap up yesterday and a DOJI today.

Now for the most amazing piece of the puzzle - the up/down ratio actually went down today to 55% from 59% and there were about 300 fewer stocks going up today than went up yesterday. That suggests that today's exuberance was indeed irrational. Now despite that the NEWMOMO method is suggesting we are already due for a correction.



As I've pointed out - this is a relatively new method that I'm using and I really don't know what to expect from it but the icing on the cake is the fact that the ATR(10) for the INDU is at 78.99 which is less than 80. So I'm forecasting a down day tomorrow.

The coin says --- tails - also a down day - I have to start letting the coin go first - I think it is playing me.

We both got it right again so the score is Marlyn 12 - 8 and 4 and the coin is 9 - 11 and 4.

Happy trails to you.