Larry, and anybody else interested,

I posted a message here a while ago inquiring about how you felt about the science of trend following. I read Michael Covel's new book on trend following and thoroughly enjoyed it. I've been doing my own independent study of trend following to see how it holds up "in the real world" and I have become a real believer in this practice.

Anyway, Larry, you told me a while back that you would welcome a discussion on this topic. Well, I felt that I needed to do my own due diligence on this subject before I felt that I could discuss it with any level of intelligence. Anyway, I really respect your views on these matters and I would love to hear your views on this topic.

For starters, I'd like to pose a couple of initial questions:

1. It seems that all the "trading greats" from Jesse Livermore to Ed Seykota have been trend followers. Does it make sense to jump on the bandwagon and implement certain aspects of trend following into your own trading approach?
Why or why not? Like I said, I have already decided to implement trend following into my own market approach, but I would love to discuss its merits and drawbacks with anyone interested.

2. The COT systems you have developed are wonderful, and I suppose they could be classified as "quasi" trend-based systems that use the COT data as an exit/entry bias, but have you explored the possibility of implementing certain aspects of trend following into a COT based system? One thing that immediately comes to mind would be that of using a trailing stop as your exit rather than the COT bias, and then continue using the COT bias for your entry... just wondering.

I like trend following because it is a tested and true approach to the markets that reacts rather than tries to predict what will happen. I also like it because conceptually, this approach can be used in any market from soybeans to currencies.... all you need is a trend.

Anyway, I'm looking forward to any feedback you may have!

Thanks!

-Ben

personally, i like to have more than on style of trading. that’s how I achieve diversity. trend-following is one style that i like very much but it does have it's drawbacks.

it just makes logical sense that, if you catch a trend early and ride it (perhaps for years) you will make big gains. what you're buying (or selling short) really doesn't matter. real estate, stocks, commodities, baseball cards, etc. doesn't matter.

finding a system that will catch trends, that makes logical sense to YOU and, that you will follow - will work. that AND proper risk control (for when you're wrong) is a winning combo.

i find moving average cross-over systems to be slow so i don't like them much. also, they don't work in flat markets.

a break-out entry system such as the turtles or CANSLIM i like better but it's somewhat of an "art" reading charts. it's also fairly labor intensive so you've gotta really LIKE doing it.

trend-following, also, can be very boring. that's not a bad thing - in fact it's prudent. but you have to have the discipline to leave the trade alone and not "play" with it too much.

it is my belief that trend-following is the best way to make the big bucks. but it's a process. entry is often wrong (so you get whipsawed) and it requires patience, vigilance, and on-going management. you've got to find your style and love trading - then it'll work.

-ja

I have read quite a bit on this topic as well and have come to be a believr. Because of this I am partial to the COT strategies that use trend following elements (using a moving average to determine leverage).

I have examined a lot of trend following approaches (they are countles) and am using a couple of them. I would be interested in a deeper look at breakout systems and understand that a majority of trades will likely be losers when using such a system.

I would be willing to run a daily scan for breakout candidates in metstock and post them to this thread. My intention is that this would lead to a better understanding of this approach and eventually a full set of tradeable system rules. We would need to set some parameters:

1. breakout period.
2. volume filter? (breakout of higher than average volume)
3. price filter? (to filter out penny stocks)
4. average volume filter? (to limit us to stocks with adequte liquidity)

Any other ideas?

I have been using breakout systems a while.

IMO the best and most profitable breakout trades occur from 1 of 2 different scenarios. [1] A breakout from a previously tight trading range. Or [2] a break against a previously over-extend trend.

Scenario [1] is my favourite and where I personally have made most money. Amongst other things it enables you to have a close stop and thus achieve a very high risk / reward ratio.


Regards
Andy

to gary m,

1. breakout period.

the turtles use a 55 day new high or new low - i look for familiar patterns such as cup w/ handle, head & shoulders and channels.

2. volume filter? (breakout of higher than average volume)

100% increase (or greater) from it's 50 DMA (or 60 DMA)

3. price filter? (to filter out penny stocks)

$10 or higher

4. average volume filter? (to limit us to stocks with adequte liquidity)

100,000 avg trading volume.

these should be good basic guidelines to get you going.

also, i believe that you should trade in the direction of the overall market (i.e. if the s&p 500 is up, trade from the long side) how do you determine if the market is bullish or bearish? you can use the COT but i find it imprecise for this type of entry. moving averages are slow. P&F charts are slow too.

i like to read "market watch" in investors business daily on the front page every issue. it's pretty accurate at calling market turns. you can learn the process they use in the book "how to make money in stocks" by willian o'neil... but i cheat and just read the paper.

also, also... if you buy IBD, look at the "52 week new high, new low" list on the industry group page. if you buy long, buy from one of the top 6 groups making new highs.

when buying long you want: a bullish market, top sector, fundamentally superior stocks breaking out of a base on above average volume... these are your edges!

also, also, also... i like to buy companies that are new (ipo within past 3 years) and have fairly LOW trading volume (above 100K but below 500K on average). when the institutions "find" them, i'm already on board. of course, they are volitile!

hope this helps.

-ja

an e-zine i get and find interesting comes from:

http://www.innerworth.com/

a sample from today's report:

============================================================

Sticking to the Plan

Novice traders often report that they have difficulty sticking with their trading plan. There are many possible reasons. A common issue is not having a clearly defined plan. When a trading plan is not clearly defined, it is hard to follow and easy to abandon. When you trade by the seat of your pants, even just a little bit, you can panic at the wrong moment. You may be prone to over-thinking your plan or you may become easily consumed with self-doubt. When you clearly define a plan, in contrast, you can implement it more automatically. You can enter and exit more effortlessly. Many novice traders say, however, that even when they painstakingly delineate a trading plan, they still have difficulty following it. Depending on your personality, you may or may not have difficulty maintaining control and discipline.

============================================================

if you think you'd like to get this just go to the site and sign up.

it's free.

-ja

did you notice PAAS today?

i'd say that looks like a break out.

-ja

Joe,
I did notice the silvers have been acting pretty good lately especially today. I know I've kind of let the group down the last month what with the broken leg and all but I think I'm starting to feel the force again. So far this week I've managed to push the DOW down about 200 points. Not to shabby

Gary

Andy,
Would you consider the narrow range on PAAS enough for your scenario 1? Would you use the september low for your stop?
h

down side break outs...

FRE, AZO, AF

all look pretty good to me. i know larry mentioned being short FRE for the long haul. now might be a good time to enter.

this is not a recommendation... just a tho't. i'm wrong about half the time.



-ja

PAAS long & AZO, AF, & FRE short...
all pretty good calls (so far).

i think FRE will be the winner!

guess we'll see.

-ja

I know I'm probably wrong, but I was looking at FNM and instead of wanting to follow the trend I was thinking that at least this time, someone's going to try to prop it up. Especially since I've been hearing about bad finances at fnm, etc from normally bullish people.
h

also FNM and ERTS for short candidates

New Short Candidates
CL, CNP

Watching
FRE, FNM, ERTS, AZO, AF (short)
PAAS (long)

as for me, i like to hop on a stock when it's first breaking out, not later. for example, CNP was a good short july 21st when it broke to the downside on big volume. as of today, i think all these stocks are a bit extended.

a good way to play these stocks now (my humble opinion) is to wait for the 9 day RSI to close above 50 - (or, close BELOW 50 for PAAS). then, if you're still wanting to hop on board, that would be a lower risk entry. it's not fool-proof so use limit orders to get in and stop/loss orders to get out. just a tho't.

-ja

p.s. take a look at AH. good fundamentals (per investors.com), top industry group (top 25%), broke out of range on strong volume.

my ONLY "complaint"... it didn't close above it's 52 week high. not a deal-breaker, but not perfect.

FOSL looks like a channel break-out to me.

-ja

Thanks for voicing your humble opinion Joeaaron. I have a lot to learn and appreciate what you have taught me so far.
In regards to designing a breakout system, I have thought about a few approaches and would like your opinion, as well as those of other board members.
We could run a scan daily for breakouts and enter only upon the occurrence of a more extreme price. For securities breaking to new highs, enter if the new high is exceeded the next day on a stop. Opposite rules for new lows. An exit stop could be set above/below the first breakout day (the day it came to our attention). This would involve a higher risk entry, but might also increase the chance of success per trade.
One could develop a list of securities to watch daily and set stop entries beyond obvious congestion points. This would achieve a lower risk entry, but might have a lower success rate since you enter before you know what the day-end volume will be, or if the high/low will even be sustained until the end of the day. To alleviate these potential issues these, securities might be reviewed daily before the market close and entered with market orders. Although I could see this limiting the size of your watch list if time is a concern.
Another option could involve entering on the open on the day following a signaled breakout. Probably somewhere in the middle of the above two entries regarding benefits/drawbacks.
My current money management strategy involves setting an exit stop away from recent support/resistance or using an ATR derived stop. I look at support/resistance in the direction of my intended trade, if there is a potential reversal point too close to my intended entry, I will not take the trade. Once the exit stop has been determined I can calculate how much I can purchase while risking < 2% of the account. A trailing stop is determined based on 3 times ATR or an SAR, whichever is farther away from the current price since I don’t want to get stopped out. I haven’t developed a systematic approach to pyramiding the position, although I know I should.
I would like to develop a system with sufficient detail to allow testing then trading with confidence.
Maybe too many scatterbrained ideas in this post, but I would appreciate anyone’s ideas/thoughts on any part of it.

Gary M

gary m.

i think you're smart to put this much focus and thought into your money mgmt/position sizing strategy. once you've got your risk and your emotions under control you've very nearly "idiot-proofed" your trading! it is my humble opinion that an average entry system combined with an ABOVE-average money mgmt system that is faithfully followed is a winning combo!

there are many ways to control risk. i have decided that simple is best for me so i usually just place stop/loss orders on my positions.

re: pyramiding; anyone who's read this post for a year knows i've been looking into this for a while. here's what i'm doing currently... in case you care. I have decided that 5 stocks is all I want to hold at any given time. when I get a buy signal (based on the criteria I listed above) I take a 13% initial position and use an 8% stop/loss. this puts 1.04% of my account at risk. if I have 5 stocks at risk all at once my portfolio “heat” = 5.2%. however, I try NOT to have 5 positions at risk all at the same time.

if a stock goes my way 4%, I’ll add to my shares. I look to see how many shares I bought initially, then I figure 70% of that and add that many more shares. (i.e. if I bot 100 shares initially, I would now buy 70 more shares).

I now figure the AVERAGE BUY PRICE and move my stop/loss order up to 8% below that. I now have 1.32% at risk on that stock. if I have 5 stocks risk all at once my portfolio “heat” = 6.6%.

notice my portfolio heat INCREASED. my reasoning is; I’m willing to take on more risk since the trades seem to be going my way. also, the chances of my having all 5 stocks fully loaded at the same time are very low. it’s never happened yet tho’, theoretically, it is possible.

I’m now fully invested (with approximately 2 to 1 leverage!) yet my portfolio risk is never greater than 10% even with broker fees and “reasonable” slippage figured in.

I do not trail my stop/loss every day. the next target I look for is a 15% gain from the average buy price. if my initial buy price was 100 and I added shares when it reached 104 (4% gain), my average buy price is now 101.65. when the price of the stock goes up 15%, or to 116.89, I move my stop/loss order up to breakeven (101.65). now I’ve got a “free trade”. this is where I want to be! 2 to 1 leverage and 0% of my money at risk!

once I’m in a free trade I’ll widen my stop/loss to 25%. understand, I do NOT move my stop!!! I just allow the price to drift in my favor further and further without trailing the stop/loss order until it’s 25% away. then I’ll move it in my favor, keeping it about 25% away – never back-tracking!

this gives the stock room to wiggle and I won’t get “shaken out”. I watch the price/volume action daily and will sell before the 25% stop gets hit if I feel the easy money has been made and/or there are better opportunities out there! having a “5 stock maximum” rule keeps me constantly “pruning” my portfolio.

as you can see, my money mgmt system is more “sophisticated” than my entry system. in fact, I believe a superior money mgmt system can make ANY entry system profitable. ANY!

here’s the secret: you can control your risk profile. you can’t control the price movement of a stock. most people who fail as stock traders never figure this out. they think that you need to “know” the direction of the stock and bet accordingly. their “need to be right” bias kills them.

the truth is: you don’t have to be right on any given trade. you just have to make lots of money when you ARE right, and lose small amounts when you are “wrong”. this is how Vegas casinos make money. the house has the edge – so they will, in time, ALWAYS win! tho’ they may NOT win this hand. a slight edge in your entry system and a rock-solid money mgmt system will skew the mathematical odds into your favor so that success is inevitable! you just have to keep playing.

if you like what I’ve talked about here you should really read “how to make money in stocks” by William O’Neil. he’s the market wizard – not me.

-ja

p.s. regarding break-out entries… go to stockcharts.com, click on chart school and read about “cup with handle”, “channel” & “head and shoulders” break-outs. I also like “on balance volume” to confirm my ideas. you can read about that too.

JA,
I've been looking at OBV for a while. Have you noticed it to be more effective on a daily basis than on a longer term chart? At least as a sort of "pre-breakout" signal. (I know it's not good to use it before the breakout, I'm just asking about time scale.)
h

i'm not sure i fully understand the question but...

technically speaking... i look at price and volume. if a price is breaking out of a base that i'm familiar with on strong volume THAT'S the most important thing.

i look at parabolic sar dots, on balance volume & chakin money flow (80) as "confirming" indicators.

the most important things to me are:

1. market trend
2. top sector
3. top fundamentals
4. weekly price/volume action
5. daily price/volume action

hope this helps.

-ja

Ben,

This is a good thread. Thank you for starting it. You asked a couple of question in your initial post that I want to answer.

Ed Seykota is definitely a die hard trend follower. Jesse Livermore was as well, although Livermore would try harder to get in early on a trend with a small position and then add to it if it went his way and kick it out if it didn’t. Also, Livermore was a tremendously talented risk taker whose downfall was position sizing. He made and lost huge amounts of money.

But it would be a mistake to say that “all the trading greats” are trend followers. If you’ll read “Market Wizards” you’ll find that they use very different styles and strategies to achieve their results.

Trend following works extremely well when you can find a trend. It doesn’t work well at all when there is no trend. It’s as simple as that. So you just have to use methods to probe for a trend and stay with it when you find one and get out fast when you don’t. Therefore, you have to be prepared to take more losses than profits and hope that your occasional large profits will be large enough to more than offset your many small losses. As with all methods of investing, money management is critical. It is a way of investing that requires a relatively large amount of capital because you have to spread your capital over multiple markets in order to find some that are trending.

By the way, the COT is not a trend following system. It’s more of a trend setting system. The commercials are not trend followers. It seems like a trend following system sometimes because the trend more often than not ends up going the way of the smart money.

Larry

I guess it doesn't lend too much credibility to me when I started this thread by misspelling the word "revisited"! How embarrasing!

Anyway, I'd like to thank everyone who has taken the time to respond to my original posting in this thread about trend following. I have some follow up comments that I'd like to make after reading some of the responses.

I would like to point out that there were several things that were said about the concepts behind trend following that I must disagree with.

First of all, in the true spirit of trend following, we are buying high and selling higher with a success rate of about 40%. I realize that there are many ways to trade, but many that rely heavily on T.A. are trying to predict price action rather that react to it. I have definitely embraced the concept that I do not have a crystal ball and I cannot predict where the market will go. My job as a trend follower is to try to recognize trends and then follow them, in a very mechanical, systematized approach. From what I’ve read, famous trend followers like John Henry don’t use limits as trend followers, but rather employ trailing stops that they use as their exists when the trend has run its course or didn’t turn out to be a trend at all.

With these basic concepts being understood, the leading question for me would be the following:

1. Depending on the time-frame, what are my criteria for defining a trend, and where should I set my trailing stop so that I can ride the occasional trend “to the moon?”

I think that if we can answer the above question in specific terms, we could really be onto something! As I mentioned before, I know that there are many wonderful trading systems and ways to approach the markets. Right now, I am merely trying to get a grasp on the concepts that will help me become a successful trend follower. Also, another beautiful thing that I’ve found in the world of trend following is that instead of spending my time learning about companies, fundamentals, and technical indicators, I can trade successfully in ANY market by spending my time looking for trends that match my criteria.

So, if anyone reading this would be interested, I’d like to have a discussion about what criteria we can use to define trends that we would like to try to follow and where we should set our trailing stops.

Any ideas?

Amazing, Larry! We both posted at exactly the same time!




--Ben

Ben, this is a great thread. I am looking forward to the conversation we are going to have on this topic.

Have you read the free PDF from the following website? It contains the original "turtles" trading rules. I've heard that the original turtles have gone on to better systems, maybe we can design and test a a comparable system during the course of this discussion.

http://www.originalturtles.org/

Gary M

Gary M,

I read that .pdf some time ago and found it very informative. Thanks for bringing that to my attention again!

I really like your suggestion that we try to come up with our own similar system. A few posts ago, Larry mentioned that in order to be a true trend follower, you'll need extensive capital so that you can enter trends that you find in the various markets. This seems like an accurate assessment to me, but being that I don't have extensive capital to do this, I'm hoping to find some kind of happy medium. Perhaps with less capital we could spend more time waiting for trends to appear in the select markets that we choose? Some markets, like the currencies, trend much better than others it seems.

Conceptually, trend following works for me, but I'm still hung up on how to define the trend and where to set my trailing stop. Do I just eyeball the chart and take obvious trends? Or, do I define the trend like the turtles by looking for a higher "high" after 21 periods, or something like that? I'm still searching for the answer, so I'd be willing to test a number of different scenarios. Obviously, the wider my stop gap, the longer the trend I might stick with... but what works best?

--Ben

here's an idea i had but have never tested. i'll throw it out and, if it strikes anyone's fancy they can look into it further.

this is a combo of larry's parabolic sar trend-following strat & something i read in alexander elder's book "come into my trading room".

RULES:

decide on a universe of index funds. they can be ETF's, fidelity sector funds, profund, rydex... whatever. in your charting software enter ALL the funds into TWO folders, one for daily price movement, one for weekly.

in the weekly folder track the price, the 40 week moving average and the parabolic sar dots (using the default criteria).

in the daily folder track the price, the 50 day moving average and the parabolic sar dots (using the default criteria).

every day, check the weekly charts. all of the funds that are ABOVE their 40 WMA and have their parabolic sar dot BELOW the price goes on your short list!

from this short list, check the daily charts. look for funds with the price ABOVE their 50 DMA but have parabolic sar dots ABOVE the price (bearish). delete all the funds that do NOT fit this criteria. what's left is your BUY LIST!

the buy list is a group of funds that have bullish trends on the weekly charts but the current price is in a short-term bearish move. your "bet" is that the long term (weekly) trend will resume and the short term trend (daily) is just a temporary wiggle and NOW is a good buy point! if you're a wuss (like me) you could wait on the daily parabolic sar dot to flop to bullish (i.e. go BELOW the price) THEN go long. whatever you want.

if you get too many buys, a tie-break might be, buy the fund whose weekly price is closest to it's 40 wma. (i.e. not too extended).

personally, i would NOT try this with futures, options or even individual stocks. i trust the trend of a GROUP of stocks that comprise a sector or an index much more that i trust the trend of a single stock. my 2 cents.

if anyone does any testing on this strat please share it with us! i'm too busy... (okay, i'm too LAZY - you caught me), but i'd love to know how it works.

-ja

p.s. i do recommend the book by alexander elder i mentioned above

I think what you will find is that there is no such thing as a right method for trailing stops. Any strategy that you come up with will have its strengths and weaknesses. So you just have to choose one that you’re comfortable with and go with it.

Ben mentioned Ed Seykota’s name in his original post. Seykota probably is the ultimate trend follower and one of the most successful traders ever. But I think what you’ll find is that people like him don’t spend much time worrying about what might be the best trend following system or the best way to place a trailing stop.

Seykota has a web site. I’m going to give you some links from his site. You will find that there is very little discussion (almost none) about specific systems and stop loss strategies. It’s all about psychology (he apparently is a big believer in NPL) and risk management. Hopefully, that’s not a surprise to any of you.

Here’s the link on risk management…

http://www.seykota.com/tribe/risk/index.htm

What he has to say about psychology is covered in his FAQ’s (Warning: be prepared for your eyes to glaze over). It’s very different from what you might expect…

http://www.seykota.com/tribe/FAQ/index.htm

To the extent that Seykota has anything to say about specific strategy, it’s covered on a page about the technical analyst from the 1930’s, Richard Donchian…

http://www.seykota.com/tribe/Resources/Donchian/index.htm

You may also find helpful the comments he makes on system development…

http://www.seykota.com/tribe/Charts/System_1/index.htm

I think the most important comment he makes, from what I know of Ed Seykota, is this one: “Rule of thumb: simple systems work best.”

The main thing I hope you get from his site is how differently he thinks from those who are searching for some sort of Holy Grail. The Holy Grail is YOU.

Larry