what are some of your most effective exit filters? | LinkedIn http://www.linkedin.com/groupItem?view=&srchtype=discussedNews&g...

Automated Trading Strategies (Algorithmic Traders of Stocks,...

Discussions Members Promotions Jobs Search More... Discussion | Group rules

Ccil k her e t o f n
i d out m or e!

what are some of your most effective exit filters? Latest Updates
It has been said on several occasions in this group (and elsewhere) that with effective exit
strategies, even a system that employs randomized entries can be profitable. In the interest of
building the most robust systems possible, I thought it would be useful to share general Alex Krishtop and 2 more
methods in this area. I'm not necessarily expecting people to share specifics, but general commented on:
Tucker areas of focus. That way we all may gain some insight into areas of research that may be bigmoneytrade EA
Unfollow able to enhance returns. 37 comments 1 hour ago

In order to start things off, I'd like to share two methods that have produced good results with Claudio Gonzalez commented in the
a counter-trend system: 1. "adverse volatility exit": system exits/reduces when the latest bar H group on Holiday Greeting from
- L exceeds a factor of the ATR and price closes against the position. 2. "excessive standard Doron Whitman: Thanks Doron for
deviation" - system exits/reduces when price exceeds a factor of 1 standard deviation away your greetings and wish you and the
from a moving average and against the position. trading community a fruitfull , exiting
1 month ago and growing trading year 2013!!!
Like (3) Comment (8) 1 hour ago
Like Comment Follow Flag More
Victor Eduardo Lopez De Anda
commented in the group on Trading
Guy Bormann, Jikun (Jacob) Wu and 13 others like this as religion. Part 2: investors.: Great
article !!
37 comments • Jump to most recent comments
Like Comment (9) 2 hours ago

Kevin Spreier • Trailing Stop at 60% of max profit. See all updates »

1 month ago • Like 1 Ads by LinkedIn Members

Follow Kevin
Free Forex Trading eBook
Gain Trading Insights and be one of
Tucker Sferro, CMT • thanks Kevin - how do you filter positions that are underwater the the 1,000 to get the Saxo eBook.
entire trade?
1 month ago • Like Learn More »

Tucker How Do You Manage Risk?

Unfollow
Download Whitepaper: Near
Real-Time Risk Management
Kevin Spreier • Strangely enough, by position size. I found that if I limit the margin used
by a position to less than 1/2 of my balance (at 50:1 leverage), there is enough equity Learn More »
left over to wait and let the market come back to my price, closing at a profit. I have very
few losers trading EURUSD.
Follow Kevin 1 month ago • Like
Manager's Choice
1
Holiday Greeting from Doron
Whitman
Ken Jarrad • By 'counter trend' do you mean a bet in the direction of the trend but Doron Whitman See all »
entered when the price moves against the trend? If so then your -1 x stdev exit would be
a great entry, wouldn't it?
Top Influencers This Week
1 month ago • Like
Follow Ken

Tucker Sferro, CMT • it's actually against the direction of the prevailing trend however it
is a *factor* of an SD (not 1 SD), so it basically is saying that price has moved too far
from fair value
1 month ago • Like 1
Tucker
Unfollow

Ken Jarrad • Tucker, thank you for clarifying the trade setup. The reason I
misunderstood is because Michael Covel has banned all counter-trend trades.
You
are a renegade. I like it.
Follow Ken

1 of 4 27/12/2012 12:46
what are some of your most effective exit filters? | LinkedIn http://www.linkedin.com/groupItem?view=&srchtype=discussedNews&g...

Jaleh Samani

Chris Tubby • I agree with Kevin Spreier. However, I would use trailing/rolling stop of
20% from highest level it has printed. Therefore, if the price tops out at 100 points profit sean seo
your stop would be activated at 80 points or a 200 point rally would put the stop at 160
points profit.
Follow Chris Another technical method based on charts is Bollinger Bands or peak study.
Alex Krishtop
1 month ago • Like 1

Raphael Geraldelli • Price channel line. The number of periods for the channel would Doron Whitman

depend on pattern or trend characteristics.

It is also good to see which works better: price channel using highs n lows, or just
closes...
Follow Raphael 1 month ago • Like 1
Group Statistics
Director
Yehuda Aharon Zbaida • Thanks guys, I am eager to experiment with these suggestion Manager

CHECK OUT Entry

1 month ago • Like 1 INSIGHTFUL

MEMBERS
Follow Yehuda STATISTICS
Aharon
ON THIS GROUP
3,759
Zachary David • The most effective exit filters are counter signals to your original entry.
However, if you have signal that is fairly sparse, your signal's efficacy will decay before a View Group Statistics »
counter signal.

Follow Zachary For a non-specific framework, this post from TR8DR is a good way to begin thinking
about position management: http://tr8dr.wordpress.com/2012/09/29/money-
management/
1 month ago • Like 1

Zachary David • Tucker,

Your current research is definitely on the right track. If you characterize the market that
your signal works well in, then you can exit when the market exhibits characteristics
opposite of the one that "works."
Follow Zachary
By and far, the biggest influence on a particular trading algorithm's efficacy is volatility.
Some work well in low, some work well in high. If you know which type yours works best
in, then you should exit when the opposite appears—because as we all know, volatility is
one of the most persistent factors in any market.
1 month ago • Like 1

Scott Boulette • I use a series of exits but one that works very well (in conjunction with
the others) is a move to breakeven after a certain amount of time given the trade is going
against me. I have all my other exits in place but after say 2 minutes I attempt to scratch
the trade.
Follow Scott
This works particularly well in counter trend trades (I have special permission from
Michael Covel) where you expect a relatively immediate bounce but haven't gotten it after
a certain period of time.

The reason this works is to avoid what amounts to a random trade. If you expect a
certain reaction and don't get it within a certain period of time, you are effectively in a
random trade at that point. It doesn't mean you won't win, it does mean that if you do
win, it was likely a random occurrence.
1 month ago • Like • Reply privately • Flag as inappropriate 2

Alex Krishtop • Excellent technique, Scott, let me only add that it works well mostly for
relatively short-term trades, but not for positions held over a number of days.
1 month ago • Like 1

Follow Alex

Scott Boulette • @Alex very true, thanks for pointing that out. And while I mentioned 2
minutes, that could just as easily be 2 seconds.

2 of 4 27/12/2012 12:46
what are some of your most effective exit filters? | LinkedIn http://www.linkedin.com/groupItem?view=&srchtype=discussedNews&g...

Follow Scott

Tucker Sferro, CMT • @Scott this sounds like a great idea, one that I look forward to
researching/deploying. This is one of the core principles of any trading methodology,
systematic or otherwise: that we make trading decisions based on immediate
expectations driven by observed conditions. Once these expectations fail to materialize
Tucker within the expected time frame, we enter the realm of hope/randomness.
Unfollow
One question though - isn't your entry price (i.e. break even point) and arbitrary figure as
far as the market is concerned? By choosing to exit at this level once the instrument
enters what we deem to be a random phase, wouldn't our attempts to exit at an
apparently arbitrary price also be considered a random event? If this is the case,
wouldn't a simple time exit be more appropriate? (Perhaps the entry point of a counter
trend systems could be considered significant since entry points typically follow extended
moves?)
1 month ago • Like

David Fry • For intermediate trends lasting for several weeks DeMark works well for us.

1 month ago • Like

Follow David

Scott Boulette • @Tucker, this works in conjunction with other exits that are more along
the lines of what you describe. One way of looking at price movement after entry is a set
of Bayesian probabilities that collapse into a standard win/loss ratio.

Follow Scott If you look at the price movement and the win/loss ratio for every combination of price
movement, you will notice a very clear pattern. For example, if price moves 1 tick, 2
ticks, etc. against my entry, what is the probability of making the profit target?

If you go through the analysis, you see some clear exit techniques. I also use a sort of
negative trailing profit target. Why would I look for more positive movement from price
given it went against my entry then if it went for my entry?

In essence, this is what I am doing if I hold my profit target at a constant price while
price is actually moving the other direction. If that worked then the logical thing to do
would be to "pretend" I had an entry, wait for price to move against it and then enter at
the lower price looking for a profit target based on the "pretend" entry.

There are a number of variations on this but my analysis (and practice) is to use negative
excursion points to lower my profit targets with breakeven based on time being just one
of those.

I suspect this all makes more sense to me than it does to anyone else based on this
limited explanation given I use this in live trading.
1 month ago • Like 1

tom mcginnis • For a scalp from 4-hour/210-minute/2-hour w/1-minute candles,

-- 3/4s of an expected move

-- ADX peak/downturn
Follow tom -- decreasing volume or trades/minute
-- turning TICK
-- turning TRIN
-- turning Stochastic
-- turning MFI
-- turning MACD

Pretty much in that order!

(Please excuse the surprise, but I'm embarrassed at the number of times I've failed to
reinforce such a protocol. And as noted above, it's not so much the entrance which
makes for profitable trades -- despite our unending focus on it. It is instead the exit, exit,
exit.)
1 month ago • Like 2

Dmytro Shchurov • The most effective stop technique I've ever used (for manual and
automated trading) is 38% on Fibonacci Fan. The tool is placed on a breakout bar, and
the rule is to not move it if 38% on the next bar appears farther from the current price
after moving. Stop is not moved if its price on the next bar is onward of a market. In this

3 of 4 27/12/2012 12:46
what are some of your most effective exit filters? | LinkedIn http://www.linkedin.com/groupItem?view=&srchtype=discussedNews&g...

Follow Dmytro

André Maurice J. Bost • We look at correlated pair-ed legs which suggest suitable
trade entry & exit. So to say it another way, our hedge-leg correlation strength will
indicate the exit.
1 month ago • Like 1
Follow André
Maurice J.

Send me an email for each new comment. Add Comment

Ads by LinkedIn Members

Your Wikipedia Profile Win a Saxo Bank iPad mini
written by Wikipedia experts, just $295, Just vote for you favorite 2013 prediciton
results guaranteed. and get the win an iPad

Help Center About Press Blog Careers Advertising Talent Solutions Tools Mobile Developers Publishers Language Upgrade Your Account
LinkedIn Corporation © 2012 User Agreement Privacy Policy Community Guidelines Cookie Policy Copyright Policy Send Feedback

4 of 4 27/12/2012 12:46