Melf's Workshop

[Melf Elf]

If 3.52 or so holds than we will have some consolodation, with higher lows and more promising to look above 4.20.  If 3.30 breaks to the down side NVAX is in a world of hurt...  Thoughts?

Joe,

My apologies, but I'm using this thread for "general purpose" discussions, using the stocks that I've mentioned to apply the particular points made, to stocks, in general.  If I start focusing on stocks that others are interested in and spending the time to analyze them, I'm going to get myself into trouble, time wise.  I hope that you understand. 

Are you shorting FPP or CPST, (on paper?) at these low levels?

No.  Neither of my brokers will allow a short under $5.

 Following trends we want to take a bite out of the middle and would you agree that these two if not at their bottom are to far over sold to trust the trend in a short trade!
Cheers Joe

Completely agree.

Using FPP as an example, good places to short were:

1. The break of 6.75 support

2. The re-test of that break, although that trade got a little hairy on the Bearish Island Reversal fakeout to the upside.

3. The break of 6.04 support.

4. The failed re-test of 6.04, that failed at 6.00

5. The break of 4.80 - 4.81 support

6. The break of 4.25 - 4.35 support

At 2.90, this would be "shorting into the hole."  From the first break of support at 6.75, the stock already is down 57%.  The trade is much too far gone, in my opinion.

Be aware, also, that these thinly traded stocks can be very difficult to short.   Not only are shares often not available to short, it's difficult to get decent fills.  If you put in a limit order, they tend to move the Bid and Ask away from you.  If you put in a market order, they can fill you wherever they want, and the fills can be really, really lousy. 

I'd suggest shorting more liquid stocks.  The closer to average daily volume of 1,000,000 shares or more, the better.  Much easier to enter and to exit.

[akclide]

Hey Melf,
           First of all, thanks for all the work you have put in to improve the site.A quick question. In you opinion Which is the best technical signal that has better than say 70-80% results and how do you (MELF) use it, I would certainly appreciate it if you wouldn't mind sharing it.
Thanks AK

[Melf Elf]

In you opinion Which is the best technical signal that has better than say 70-80% results and how do you (MELF) use it, I would certainly appreciate it if you wouldn't mind sharing it.
Thanks AK

akclide,

If anyone has found a technical signal that has better than 70-80% results, I'd like to know about it, too.   

I don't use anything mechanical, like "Always buy MACD crossing the zero line," etc.  I don't have the programming knowledge or knowledge of statistics to conduct any kind of research, but after years of studying many, many charts and indictors, I can tell you that I've stopped searching for "The Holy Grail" of indicators. 

Probably everyone who is new to technical analysis tries Stochastics and Relative Strength, etc., and finds out that often, they don't work.  For example, "Stochastics went above 20...or 25...or 30, so it said BUY."  Then, the trade doesn't work out! 

Here's the problem: the indicator doesn't know what the price chart looks like.  Was the buy signal given after the chart had formed a nice bullish base, or was the buy signal given when the stock was ready to smack right into resistance? 

Big difference, right?

So, if we say, "Buy signals in Stochastics only worked 55% of the time, so it's a lousy indicator,"  Stochastics might want to say back to us, "I might work a lot better if you'd stop using my signals that are against the trend." 

That would take a lot of research to conduct a study of under what circumstances we should take signals.  Should it be at least a 4 week base?  A six week base?  Should there be a discernible pattern?  Should we wait for a Stochastics "buy signal" AND a pattern or trendline breakout?

More questions than answers, I realize, but we can see why "it ain't simple."  But, I also try to make it as simple as possible, and not use too many indicators.  I primarily use my Fibonaccii measures of Relative Strength, MACD, the Ichimoku Kinko Hyo charts, and the rest is combining those with chart pattens, trendlines, and support and resistance.

Example: GIGM, using the 13/21 RSIs (apologies that this chart is difficult to look at). 

I put vertical lines to show when the RSIs came into synchronicity (readings within about a point of each other), and what the chart was doing when the signals were given:

1. March 31, 2006 - Bullish Synchronicity - GIGM re-tested 5.62, which was former resistance, and held, so "former resistance now acted as support," as it should.  The close was 5.68.  The high on the day was 5.85, so the 13/21 RSIs would give a "buy pivot" signal the next day if 5.86 got printed, a penny above the prior day's high.

The 13/21RSIs don't have a clue what the chart looks like.  We do. 

When 5.86 got printed the next day, do we want to take the buy signal, or not?  Yes.  The trend is bullish.  The chart looks fine.  GIGM printed the 5.86 the next session, April 3, and closed on the high of the day, at 6.30.

2. April 11, 2006 - Bullish Synchronicity again, which I view as a "repeat buy signal" when it occurs this close in time.  High on the day was 6.60, so a print of 6.61 the next session would be another "buy pivot."  Do we want to take the signal, or not?  Yes.

3. May 15 - Broken H&S Top in the 13/21RSIs.  Sell signal.  What does the chart look like?  Parabolic rally into a Bearish Rising Wedge.  Do we want to take the signal and sell the break of the Bearish Rising Wedge, or sell short a re-test of the broken Bearish Rising Wedge.  Yes.

I'll skip the remaining signals since this is so long, but you get the idea, about combining signals from indicators with what the chart itself is doing.

Hope that helps, akclide.  Good luck!

NOTE: I didn't buy any of the three bullish signals in this indicator in August because of the downside reversal at earnings on heavy volume, and because the pattern looked like a Bear Flag.    

Even with indicators and charts, it still ain't easy.   

[Melf Elf]

Was the buy signal given after the chart had formed a nice bullish base, or was the buy signal given when the stock was ready to smack right into resistance? 

Big difference, right?

akclide,

Here's an example of that, using ESLR.

On August 17, MACD had a positive divergence AND it crossed above its signal line, so a "buy signal" was given.  Do we want to take the signal, or not?

What's the chart doing, at that point?

ESLR closed at 9.24, and was up 15.6% off its recent 7.99 low, which would be the "logical stop" on the trade.

Where is resistance?

The blue trendline from broken 10.00 - 10.25 support, which was at 10.34 on August 17, when the MACD "buy signal" was given, or 11.8% higher from the 9.24 close.

The trend is bearish.  The stock hasn't formed any kind of base to attack all of that overhead resistance.  Support is 15.6% below the close; resistance is only 11.8% above the close.

Do we want to take the buy signal?  No, thank you. 

Do we want to short ESLR at 10.42, at trendline resistance, and at the declining 50DMA, or short it at 9.92 on the break of the Bear Flag, when MACD ticked down from just below zero? 

What's the trend?  Bearish...

How is the stock acting?  Bearish... 

[Melf Elf]

What's the trend?  Bearish...

How is the stock acting?  Bearish... 

Topic: The Trend Is Our Friend

We all hear that when we're new to the markets, but we tend to ignore something that simple, and tend to complicate it by trying to do things like guessing tops and bottoms, for example.

Scroll up and look at the Bear Rising Wedge (or Bear Flag, if you will) rally back to resistance in the ESLR daily chart that I posted this morning, then look at this Bear Rising Wedge (or Bear Flag) in the Hourly chart.

Given "the trend is our friend," how would we have expected this pattern to resolve?  Once the pattern broke to the downside, does it seem reasonable to expect it to go lower?

It doesn't ALWAYS.  I can't give a percentage, but more often than not, this is the result when a bearish pattern breaks to the downside in a bearish trend.

[akclide]

Thanks For your help and wisdom.I have always enjoyed reading your posts and have learned from them.Keep them coming may be one day everything will click.

[nullzero]

ESLR looks bearish right now but I believe very soon will be a prime entry for long term hold im looking at adding between 6-8. I believe ESLR will be a leader in the solar industry in the upcoming years ahead, their cost effective manafacturing and a primary location in the U.S. will be key to success over the long term. A weakning USD will help boost the demand for industrial manufacturing in the U.S. (making it more economical to produce goods in the U.S. and export domestically and internationally.) Right now ESLR is getting hit hard because of the energy correction... However in the next 5 years its safe to say oil and energy prices will be even higher. Oil doesnt grow off trees . In the next 5-15 years its a safe guess there will be a major shift in world energy sources; Solar, bio fuels, and other alternative energy methods will be used on a mass scale at sometime in the future. Anyway I think ESLR has lots of potential.

[joet]

Melf, thank you again!  You keep me studying!    I have been thinking about your bulls and bears duking it out across a line of resistance/support that you mentioned a few post back.  Your "overhead resistance" as the bag holders sell to get out?  Couldn't this also be looked at as bagholders holding to break even there by supporting the rebound price?  I pose the ? because I am thinking about it... as price is mostly driven by BIG investors not us small retailers  would you say that they are more likely to try to "cut losses" and move on to a stock that is setting new highs?  (did you see SEBFR's post on shorting GLW, it fits this logic...)

You have been waching far longer than I, I do need to abserve the trend more closely.  Yet my observations have noticed that often if RSI hits 80+ or 20- QUICKLY than a bounce in the opposite direction is more probable than not (GNTA this week), if the extreem RSI is met gradually than the trend seems stronger and less likely for a profitable short term trade (FPP) (and VLO somwhere in between)

That is my 2 cents on the RSI indicator.

I am curious to see where HSOA goes, deffinate down trend, and i am holding a bag... I am hoping that this rebound gives me a chance to get out above 6 but the little guy is saying take the money and run NOW!  NVAX is similar but i feel has more upside potential in this market so will hold.

I bought some ISIG on the pull back today, here is an up trend for ya!

Cheers Joe

[Melf Elf]

Thanks For your help and wisdom.I have always enjoyed reading your posts and have learned from them.Keep them coming may be one day everything will click.

You're very welcome, akclide.  Glad if any of it helps you.

In the next 5-15 years its a safe guess there will be a major shift in world energy sources; Solar, bio fuels, and other alternative energy methods will be used on a mass scale at sometime in the future. Anyway I think ESLR has lots of potential.

nullz,

I'm surprised that it hasn't happened a lot sooner, given the problems that we had in the '70s.

Melf, thank you again! You keep me studying! I have been thinking about your bulls and bears duking it out across a line of resistance/support that you mentioned a few post back. Your "overhead resistance" as the bag holders sell to get out? Couldn't this also be looked at as bagholders holding to break even there by supporting the rebound price?

joe,

I'm not quite sure what you mean by the last question, but yes, when stocks get back to resistance, bulls and bears "duke it out,' and if there isn't much of a base bulit, the bulls often get handily defeated, like on September 1 in ESLR, when the Bulls got turned back right at resistance (blue trendline and Red #4) on a Bearish Gravestone Doji Star Hangman.

The Bulls weren't able to make any progress whatsoever. 

Topic: Managing The Trade

ESLR broke below another bearish pattern, and also made a "lower low," below the 8.75 low of this little Bearish Rising Wedge.

When that occurs, we can lower the stop on a short trade to a penny above the high of the pattern, which in this case would be 9.51, a penny above the 9.50 high.

[Melf Elf]

[more often than not, this is the result when a bearish pattern breaks to the downside in a bearish trend.

Topic: "Don't Step Over Dollars To Pick Up Dimes"

ESLR has played out in textbook fashion.  Bearish patterns.  Failed re-test of broken trendline support.  It's come down this morning toward the August low target of 7.99.  It might not get down there exactly, and we can't expect to get every last dime out of a trade. 

This is close enough to the target to "take profits when targets get MADE."

[Melf Elf]

Topic: Fighting The Trend

After closing back below former "quintuple bottom" support at 1.25 on September 6, GV ran into the 8-day Tenkan-sen (green), and has walked down the under side of it to a new low for move yesterday, at 1.00.

In addition to the stacked resistance beginning at 1.49 that I outlined in my post this morning, near-term resistance, basis the 10-minute chart below, is at 1.44. GV has been trying to get through there for the past few days, and repeatedly has gotten refused.

The high for the move came in two weeks later, on August 2, at 1.49 exactly , which was the low end of stacked resistance. GV is down an addtional 32% from there. 

While it's tempting to try to catch a bottom, or to buy a stock because it looks "cheap," it's also tough, fighting a trend.

UGH. 

[Melf Elf]

Topic: W-Bottoms

Alister,

W-Bottoms, generally, look like the letter "W," although sometimes, as in the case of INSM below, the middle of the pattern is a little more complex, with the 1.29 and 1.28 highs.

Generally:

1. The stock comes down and forms a low, for the left leg of the "W" (1.03)

2. It then rallies to a high, to form the middle of the "W," which is the key to the pattern.  In this case, it's 1.29 and 1.28.

3. Then it comes back down to form the Right Leg of the "W" (1.02)

The breakout occurs when the middle (1.29 and 1.28) of the pattern is taken out to the upside.

To measure the target IN PLAY, take the high of the pattern (1.29 and 1.28) minus the low of the pattern (1.03 and 1.02), then add it to the breakout.

Since we have more than one choice of both the high and the low, I always use the more conservative of the two numbers so that I don't over-estimate the target.  So...

1.28 (the lower high) minus 1.03 (the higher low) = 0.25 points

1.28 + 0.25 = 1.53 IN PLAY

This is a good example of using the conservative numbers because the target MADE within one penny, but if I had used the 1.29 high minus the 1.02 low (0.27 points) the target would have been 1.29 + 0.27 = 1.56 IN PLAY, and that would have missed by a few cents.

As you know, though, targets aren't exact anyway, but rather, measures of "what we're aiming for."

[ScottishTrader]

Hey Melf,

I was just wondering if you have time, if you would be able to elucidate a little about how you set targets, as your targets are alway very specific, and you  seem to have a very good grasp on exactly how to set targets from different types of patterns.  For example, how exactly do you set the target for a falling wedge breaout - is it the wedge high minus the wedge low from the pattern breakout?

I would be very interested to hear exactly how you set your targets for various pattern breakouts (e.g., wedges, head and shoulders, triangles etc.)  I think it would also be of huge benefit for all of use in setting our own targets on stocks we trade.

Anyway, thanks buddy, hope you have time to get round to this, and as always, god trading

Scotsman

[tokyopua]

Hey Melf,

I was just wondering if you have time, if you would be able to elucidate a little about how you set targets, as your targets are alway very specific, and you  seem to have a very good grasp on exactly how to set targets from different types of patterns.  For example, how exactly do you set the target for a falling wedge breaout - is it the wedge high minus the wedge low from the pattern breakout?

I would be very interested to hear exactly how you set your targets for various pattern breakouts (e.g., wedges, head and shoulders, triangles etc.)  I think it would also be of huge benefit for all of use in setting our own targets on stocks we trade.

Anyway, thanks buddy, hope you have time to get round to this, and as always, god trading

Scotsman

I agree with ScottishTrader here, and I have mentioned it to you before that its impressive to me ow often you hit your upside targets so soon, and how often those stocks go down right after reaching the range of your target.  That is why I call you the Stock Surgeon, I also would love to hear some more about how you do it, too!

[Melf Elf]

Scotsman and Tok,

Sure, I'd be glad to explain.  Once in awhile, I do the Math in posts, like in the SIGA folder this morning, so you might want to look it that folder, as well.  My posts already are way-y too long, so I try not to bore everyone to death with so many details. 

I think that I explained the Math for this Symmetrical Triangle target in the AAPL folder, but we'll walk through it again here:

70.00 - was the high on August 3
62.58 - was the low on August 11 (Purple #2...I didn't have room to write it)

70.00 - 62.58 = 7.42 Points

Now you need to calculate the slope of the top of the pattern (#1 and #3) to find out the data point for the trendline for each day, so you'll know exactly where the breakout will occur.

That's the primary reason that I won't draw trendlines that go slicing through candlesticks, cutting through tops and bottoms.  You haven't got any way to know where the breakout occurs, nor can you take an accurate measurement, to derive your target IN PLAY.

In this particular case, on September 5, the trendline (#! and #3) was at 68.634, so we add the 7.42 points to that number on a breakout:

68.634 + 7.42 =Target: 76.054 IN PLAY.

I agree with ScottishTrader here, and I have mentioned it to you before that its impressive to me ow often you hit your upside targets so soon, and how often those stocks go down right after reaching the range of your target. That is why I call you the Stock Surgeon, I also would love to hear some more about how you do it, too!

On September 21, APPL printed a high of 76.06, 0.006 above the target, then it sold off quickly to 72.58 before rallying again, then forming a Bullish Falling Wedge.

I would have to guess, Tok, that the reason stocks sometimes hit targets or trendlines so exactly like that, is that technicians know where those targets are, where the support and resistance is, and they step in and buy, or sell, accordingly.

By the way, Target #1 always is the top of a pattern, to defend against a Double Top in the stock, so in the case of this Symmetrical Triangle, Target #1 would be the 70.00 high.  I always take some profits near there, "just in case."  And, in those cases, don't hang on for 70.00 exactly.  Sometimes, you'll get only 69.96...or, 69.92.  Close enough.  Take some profits.

"Don't step over dollars, to pick up dimes."

I calculate Bullish Wedges the same way Scotsman, like the one on this chart, although I've read that "you can't."  Those might be a little less reliable, as far as the target getting MADE, but it worked out fine here.

As you both know, targets don't always get MADE, especially when it's a target against the trend, coming out of a weak base.  Targets are just "what we're aiming for."  Sometimes, too, they get exceeded by a good bit, too, like this Symmetrical Triangle.

The target just is for that particular pattern.  It can go plenty higher, like AAPL has done.

Let me know if that helps, or if you need more information/examples.