Predicting the End of the Hope Rally

It is interesting to me how many pundits have become wrapped around an axle of their own choosing in their failed attempts to predict the end of this rally. I don't think it is that hard to apply wave theory and make a prediction. I have generally avoided adding my own count to the noise since this site is about reporting the waves not defining them. I feel compelled, however, to fill a vacuum left by the wave pundit-sphere: I think this rally ends by early Feb at around Sp1158-1170.

First, the pundits. Neely called the top in Jun. Oops. He seems to have lost his confidence since. Prechter first called the top in Aug and then went double short a couple of months ago. Ouch. Now his wave counts morph from expanding triangles to symmetrical triangles to ending diagonals. The many EWI tout boards follow the big guy in similar morphing counts. The one pundit that has held relatively steady is Tony Caldero, and I will borrow his charts to make my case.

 

Eyeball this chart of the S&P since the March 6 low. There is only one significant correction, from Jun11 to Jul8. Classic ewave would say that the rally should be divided into two up moves with that one correction in between. Neely has it part of a large triangle of a very odd shape. Prechter and EWI divide it into 3 moves (a triple zigzag). They are both getting lost in complexity. Only Tony Caldero has stuck to two moves. 

A typical relationship is wave A=C. A went to Jun11, B is the Jun-Jul correction, and C is what we are in. The distance from Mar9-Jun11 should equal the distance from Jul8 to the end.  That would give Sp1160 as the likely end point: Sp667 to Sp 956 = 290 pts rounded off. Then Sp869+290 = 1160 rounded up.  A range of Sp1158-62 covers all the rounding. 

Where Are We Now?

The second wave can give us guidance as to where and when this will end. The second wave is not hard to count either. The key is to see it as a double zigzag - two corrective waves connected by an X wave. This should not surprise us: 

• The first wave (Mar9-Jun11) was also a double zigzag ("ZZ" for short) 
• The X wave (Jun11-Jul8) was a single ZZ, but since X waves typically are simpler versions of the corrective patterns they connect, it also would suggest two double ZZs
• The whole rally looks like a double ZZ

Thus this structure has a high internal symmetry: a double ZZ first wave, a single ZZ middle X wave, a double ZZ final wave, and a double ZZ overall structure. We might have expected the final wave to be a double ZZ! 

All these ZZ's might put you to sleep, so bear with me. Within this second wave, let's call the first zigzag wave W, and the second wave Y. Here is the count:

• W goes to the Sp1080 peak in Sept, with its B wave a three wave irregular flat (meaning b of B went to a new high). This is a different count than Tony has in his chart (his red ab are my bc of the flat). 
• X fooled the pundits since it broke as an irregular flat with a higher internal wave than it started or ended: its A wave dropped to Sp1019, the B wave rose to a new high at Sp1101, and the C wave dropped to Sp1029 on Nov2. Again, this is a different count than Tony's chart shows. 
• Y began on Nov 2 and we are still in it.  

What I like about count this is it also captures the psychology: Prechter called the top near the end of wave A up in Aug, and got fooled by the irregular flat that followed; then he went double short in the middle of the X wave in October (Sp1101 in Tony's chart), and again got fooled by an irregular flat. Remember that casinos are places to separate money from people, markets are mechanisms to take money from traders, and irregular flats are waves to take track record from pundits.

Check out this second chart to focus on the final zigzag, wave Y, noting that Tony's labels are different from mine but his final count is similar:

Wave Y should break as a 5-3-5 ABC, with A and C five wave patterns. The sharp rise from Nov2 to Nov16 is A of Y, and easily counts as a 5 waver, as it should. The long plateau with the funky square waves is a triangle B wave, a "3". We started the final wave C of Y with the Santa Rally on Dec9 at Sp1086, and so far it has broken into five waves. We are in the final fifth wave. The end is near. 

How Far Might it Go?

We can use the internal waves of this Y wave to estimate the end of the whole rally: the usual pattern has wave C=A of a zigzag. This gives a target of around Sp1170: A went 1029 to 1114 = 85 pts, so C should go 1085 + 85 pts =1170. Pretty close to our other target of 1160 for the whole Hope Rally. 

Tony notes in the box in his chart two relationships which target Sp1158 to 1162. His "Major A" is the first wave from Mar-Jun, and like I did above he reaches Sp1158 as a target. His "Int. A" is what I termed wave W, the wave from Jul8 to mid Sept. Tony has that wave end in Oct at Sp1101, and gets his target of Sp1162. I end that wave earlier, at Sp1080, and count the Sp1101 level as the peak of the irregular flat. I would come up with a lower target using W=Y, except given it is an irregular X wave connecting W and Y, I instead count the equality of W=Y from the middle of the X wave. If you do that you get a target of Sp1164. 

Thus, in terms of distance, we have a range of Sp1158-1170. Tony also has Sp1168 as a major pivot derived from other relationships. A cluster of relationships points to an end in our target range. 

When Might it End?

In terms of time, we can use wave structure for guidance. Note that time is not as well incorporated into ewave as distance, so while we come up with tight time ranges, the confidence in them is less than with the target distance range.

In flats wave C tends to be faster than A. The period from 2000-2009 fits the pattern of a large irregular flat (irregular since the B wave exceeded the start in 2000). The A wave was choppy and sloppy, while the C wave was dramatic and sharp in 2008! 

Note: if the 2000-2009 period was a completed flat, does that mean we are now in a new bull market? One count floating around a lot of sites is that the C wave is not done yet; we have only done two of the five waves of a C (flats break as 3-3-5, so C is a five wave impulse). This is the Prechter count, that the Big One down in 2008 was Primary 1, pr P1, and the Hope Rally is P2, with a devastating collapse in P3 to follow. The other count, which I favor, is that the Hope Rally is a large X wave connecting the flat to a second corrective structure to follow. The whole pattern might become a large triangle, or a double flat. 

In zigzags C tends to be slower than A. This seems to be the case right now, as the rally has slowed its ascent since Jul8, and seems to be in a sloppy, overlapping topping motion since mid Nov. While in ZZs we see C=A in distance, since the C wave slows down, we instead see C = A+B or 1.6x A+B in time: 

• For the whole Hope Rally, A took 3 months, A+B took 4 months (18 weeks actually), C has taken now almost 6 months (Jul8 is six months to today!). The 1.6x ratio gives an estimate of 6 1/2 months, or more precisely 29 weeks, the week of Jan 25. 
• For the second wave of the Hope Rally, A+B (or in this case W+X) went from Jul8 to Nov2 or 17 weeks; 17 weeks after that gives a target of the end of Feb. 
• Since this is a double zigzag, the X wave could split it in two in time. To the middle of X took 14 weeks. 14 weeks after that take us to the week of Jan 25.
• For the current Y wave, A+B went from Nov2 to Dec9 or 5 weeks, and C should go either the same 5 weeks and top next week; or go 1.6x that or 8 weeks, and top the first week of Feb.

Add these together and we have a topping window from next week to the end of Feb, with a most likely range from the last week of Jan thru the 1st week of Feb in a target range of Sp1158-1170.

Comments

[Michael]

Yelnick,

Excellent summary of where we might be.
Thanks again for all of your hard work!

[jeff]

Yelnick and All,

The "astros" for early Feb look good for a significant market turn. I am looking for a "blow off" top first half of Feb and then a "Valentine's Day Massacre".

But, I can't separate the market from the socio-political realm because:

1. the market has been so resiliant - despite horrendous fundamentals,

2. the American public is so placid - despite being raped in every orifice by their gov't, and

3. the terrorists being so completely inept at causing the West any real harm - despite our best efforts to let them have access to us.

So, I believe Feb will see a good market turn along with either: a significant terror event and/or a significant political uprising in the US. If no extraneous events, the market could keep going further into 2010.

PS. I posted earlier about Obama going the way of Nixon on the 36th anniversary of Watergate.

Best,
Jeff Clark

[Hockthefarm]

Yelnick:

I read Dent's latest book and he has convinced me of the inevitable deflationary outcome to this mess.

You mentioned that you don't follow Dent, but I'd appreciate any comments here, and your thoughts on the upcoming final low:

TonyC:
A retest of the March lows in Q3/Q4 of 2010 which will put an end to the bear market in stocks. Tony is a pure wave guy, unlike Dent and Prechter??

Prechter:
5 years of deflation culminating in a Dow print of 777 (the 1982 low). Expects to purchase RE for pennies on the dollar before it is over. Full on depression. Expects a swift fall to start any time.

Dent:
Economic contraction resulting from the end of the boomer spending wave. Depression in the 2010 to 2012 time frame. See's a minor correction in late Feb/early March this year and strong down after mid year. Dow hits 3800 and completes a retest in 2012. Then mild recovery into 2020.

Questions:

Do I have it right for Tony and Prechter's P3?

How do you see it playing out?

I have to say that Dent's work is very compelling.

Thanks,
Hock

[yelnick]

Hock, I read Dent's prior book, and he got it right (the 5 year boom into 2007) while Prechter got it wrong. The other guy who got it right was Zoran Gayer, who had the Dow in a fifth wave to new highs. So while I do not follow Dent, I respect his opinion. I should point out that his book in 2001 had used the analogy I use, that 1995-2000 = 1915-1919, the auto boom, and the 2003-07 market is therefore our Roaring '20s. From that he derived his fifth wave prediction. (What he got wrong is how far it would go he expected a 6x type of wave as we had in the '20s to Dow40K. Oops!)

Tony C is like Neely without the all of the new wave forms. He follows a very numeric EW without being distracted but technical indicators that are directional but not precise (hemlines, for example). This is why he has been such a steady hand. I had found Neely to be equally steady into his Jun11 call, and then not so much.

EWI has an ultimate low around 400 - think of the 1974 low not the 1982 low.

Dent's view seems closer to mine. The market took off in the extended fifth from 1994-2000 from a range of 3600-4000, hence his 3800 target is sound. The logic here is not the fourth of the prior third, but the extended fifth (read: mania) goes back at least to its start. Hence a Dow4K target. If it goes farther, we get to the 1987 range of 1700-2700.

BTW EWI's count that starts in 1932 and then in 1974 ignores inflation. One a constant-Dollar basis the proper start of the wave we were in is 1949 and 1982. 1982-2000 would mark 49w3, and we are now in 49w4. That count suggests that this bear market is not the ultimate destruction of everything, back to 400, but a correction of the 82 wave up.

[joe]

"I feel compelled, however, to fill a vacuum left by the wave pundit-sphere: I think this rally ends by early Feb at around Sp1158-1170."

Yelnik - please continue to "fill this vacum" as often as you can.

Nice analysis -

Joe

[Sanjay]

Yelnick nice analysis. Wondering how come you don't mention Gann at all. Do you have any belief in his charting? Anyhow FYI he sees 1150, 1125-30 and then up to 1179 without timing then down to 1080-1100. Also an anonymous Barron's contributor who seems to write in with predictions that are 100% spot on says 1200 then 10% correction and a continuation of the rally into the 1300's for year end.
Meanwhile Charles Nenner - cycle guru who has been right/wrong 60/40 says cycle ended Thurs 1/7 and we go down into June and then up into the end of the year.

[yelnick]

Sanjay, I do not know what to do with Gann, since his methods border on numerology without underlying causation. I think some of his thinking can be added to ewave, and I know Zoran Gayer and Glenn Neely have done a bit of this. In effect, the speed of a move in one direction is useful to tell if it is still on or has reversed, and it can be used to give time/distance endpoints for a move. Zoran used this to tell if a change in direction had bifurcated or not. Neely uses this in his "box" method.

BTW I find these round number predictions (Sp1200, Sp1300) as essentially meaningless guesses. What is the logic? We go up a bit more then fall?

As to Nenner, the USD is down, gold is up, so I expect one opf those Monday Pops tomorrow to a new high. Not sure why his 1/7 prediction will hold. We shall see tomorrow I suppose.

[D]

Few months ago I tried to mix the essentials of Gann's and Elliott's works to get the Holy Grail, but with no result.It was to complicated (for me). Sometimes I use his time cycles. In my latest post I included Gann, Fibonacci and Elliott to predict a reversal point with respect to DJIA in near future.

[MI]

Yelnick, this is your best post I have read. As an individual investor, I love to bet against Elliott gang, specifically people at EWI, these guys have no credibility at all IMO. I have no words to describe their stupidity and arrogance.
I am sorry for anyone paying money to lose money.

[AD]

Yelnick,

a very helpful post indeed; a pity you do not give your own wave count more often; I agree that a 'simple' a-b-c defines clearly the current correction; I should add, though, that I've seen some bullish counts where we would ending P1...and they make some sense

[Rob]

McHugh's has a phi-mate turn date on February 19th and another on February 26th and thinks there could be a market turn in that time frame. He is leaning towards that being the top.

[MI]

Did anyone make a dime out of Machugh's research. I did not. This guy is another deer caught in the headlights. He had been calling for markets to go to zero! a tad worse than people worse than EWI. I wish these jokers trade with their money and prove that they can make a dime out of that so called research, which in essence is garbage, IMO.

[JT]

Agreed 100%.

McHugh's research is virtually worthless.

[Bird]

Sanjay, D--

As Yelnick knows, I've been interested in seeing where Gann and Elliott may meet. I've mentioned in prior posts that where a wave-2 or wave-b occurs at a square in price and time, then the end of the move (wave-5 or wave-c) should lock out in time or price. The July low in the Dow and S&P squared price and time as a wave b. An ideal time for an end of move "squaring out" to occur is January 14th or so. We are also within 150 points on the Dow of an ideal price lock out. While not necessary to satisfy this price-time rule, if both price and time hit at once it would be an incredibly precise pattern and great place for an end of the move. If neither price nor time tie in, then...I don't think the upmove can end in January or February but will continue to the next price or time juncture.

[DG]

Bird,

I just finished an analysis of Neely's results over the past few years. The numbers indicate that if

1. The market cooperates and gives us as much volatility over the next few years as it did over the last few

2. Neely can match his predictive record of the time frame I analyzed

3. A trader has the guts to size positions according to Kelly values (i.e., trade like a professional gambler)

Your returns would be 2600% over the next 3.5 years with minimal "risk of ruin" involved. I know that most of the trading talk on this blog tends toward methods and theories, but I have to ask, wouldn't the potential for those kinds of returns be sufficient to proclaim NeoWave at least some version of the "Holy Grail" you keep seeking?

For the record, this was accomplished with 60% winning trades and a winner to loser size ratio of 1.377 to 1. You could be wrong 40% of the time and still become a millionaire in less than 4 years (let's keep taxes out of the equation for the moment) and starting with less than $50K in trading capital.

Obviously, Neely can't control the amount of volatility in the markets and he can't give you the guts to risk 30% of your portfolio on every trade like the probabilities say you should, but the core theory of how you'd be trading and the rules you'd follow already exist.

[joe]

DG

If you flip a coin over and over it will come up tails three times in a row. Don't ever put 30% on a trade. The difference between guts and brains is a very fine line.

Joe

[Bird]

DG,

I am, self-confessedly, a seeker of the holy grail. That is what my soul seeks. That being said, we must also be providers, so I'm not against a little work just plain trading.

Consider the lowly fibonacci price retracement. If you traded the bar break of EVERY retracement that occurred in a given market over the last year, placed your stop at 1 unit and took profits at 2 units, I'm pretty sure you would be right fairly close to 50% of the time. At least that is what my tests have shown. With a winner to loser ratio of 2:1, I think you would come very close to Neely's $ record, as suggested by you, if not surpass it. Within a few weeks you'll know whether the approach is true or not and then you can do it for the rest of your life without need of any complicated rules. What makes this hard to do?

Neely has striven to predict the main high of this move and boldly issued a press release in June to do so. It was exciting! Neely is a grail seeker too, is he not? I'd like to meet him one day.

[john]

im just waiting for yelnick to get bullish. once that happens i know the crash we have all been waiting for will begin

[joe]

Yelnik, to his credit, has been a lot more decisive and accurate than STU (or Neely).

Joe

[DG]

If you flip a coin over and over it will come up tails three times in a row. Don't ever put 30% on a trade. The difference between guts and brains is a very fine line.

Joe

There's a slight difference in that the 30% is asymptotic (actually, you'd rerun the allocation formula after a losing trade and the number would go down. Given the number of trades in the current sample, let's say it went down to 25%, it would be 25% of the remainder of your original portfolio after losing 30%) and the payoffs aren't strictly speaking 1-to-1 because the winner to loser size ratio is above 1.

Anyway, yes, as I say, no one is going to give you the fortitude to allocate that much to a single trade. Although, taking a sliver of your portfolio and allocating it that way would not be a bad idea, quite frankly. Stick to a more reasonable level with the remainder.

[DG]

Bird,

What makes this hard to do?

I don't know. You tell me. I don't see anyone doing it or they're being quiet about it.

Anyway, I'm sure we can all think of simple rules that should enable us to make 500000000%/year. Apparently, though, none of them actually work.

My point in trying to discuss an actual track record is to get away from the idea of simple schemes that should work and discuss tangible results and the probabilities they derive from. To mix metaphors, the rubber has to meet the road at some point or it's all just pie in the sky talk.

Without a common basis for measuring what "works" and what doesn't it just devolves into methodological partisanship.

[Bird]

DG--

Most onlookers would say that apparently, Neely doesn't work. The "past few years" of Neely trades isn't a common basis for what "works" because the method seems to promise that it is much better than the original method (thus justifying the mind numbing complication), but it doesn't seem to be so.

You dismiss my retracement proposal by saying, "apparently, none of them actually work". So do one thing for me. Take your favorite trading instrument. Mark the first 30 trades with a .382, .5 or .618 retrace using as close to monowaves as is reasonable. Enter at the break of the prior bar. Stop 2-3 tics above the high or low that made the retracement. Take profits at 2 units. Tell me how many wins and how many losses. Thanks.

[DG]

You dismiss my retracement proposal by saying, "apparently, none of them actually work". So do one thing for me. Take your favorite trading instrument. Mark the first 30 trades with a .382, .5 or .618 retrace using as close to monowaves as is reasonable. Enter at the break of the prior bar. Stop 2-3 tics above the high or low that made the retracement. Take profits at 2 units. Tell me how many wins and how many losses. Thanks.

Bird, come on, use some common sense. Do you really think if this pattern was a consistently profitable pattern no one would have noticed by now?

I think it's a fairly safe assumption that if and when the "Holy Grail" gets found, it won't be something that could have been thought up by a child.

Most onlookers would say that apparently, Neely doesn't work. The "past few years" of Neely trades isn't a common basis for what "works" because the method seems to promise that it is much better than the original method (thus justifying the mind numbing complication), but it doesn't seem to be so.

By a common basis for what works, I mean actual trading statistics, not theoretical arguments why something SHOULD work. I'm putting actual results on the table and you are still talking about theory. Your Fib retrace trading system should be easy enough to backtest over the past few years, so why wouldn't you already have done so? It's like trying to compare baseball players by the number of letters in their names, rather than their batting averages, home runs, etc.

Anyway, I know you've got more of a philosophical bent, but in addition to philosophy (I actually know Ancient Greek, if you can believe it, and back in my undergraduate days read and translated Plato and all the other classic works, so I actually know quite a bit about philosophical seeking), I like to successfully trade the markets and make money, regardless of whether the method I'm using is the "Holy Grail".

In fact, this discussion reminds me of an old story about Aristotle's way of coming up with the number of teeth a horse has. Apparently, rather than just go up to a horse and count its teeth, he logically deduced that a horse should have X number of teeth, based on its nature as a horse. Of course, he was wrong about the number of teeth a horse has.

[joe]

Yelnik - do you think there could be some volatility this week - maybe a low in the 10,300 - 10,400ish range?

Joe

[Bird]

DG--

I have backtested it. That's why I'm saying that it holds up. But, being philosophical and naive (both strengths), I am not a statistician and my backtest methods are informal. So would you do me a small favor? It should take you only a few minutes time. Run my silly little trading system on your favorite instrument, favorite time frame, 30 trades in all. Tell me whether you come anywhere near 50% right, or more, with wins exceeding losses 2 to 1. Thanks!

[DG]

I have backtested it.

Great, then you can just tell us what the trading results were.

[Bird]

Oh and DG, I agree, the holy grail is not something that could be thought up as a child. My fib idea isn't my holy grail. It is presented as something that appears to me to give a statistical edge for bread and butter trading while one searches for the grail.

(And by the way the grail is a circle.)

[Bird]

I've run the tests many times. Probably several hundreds of trades. It would be the very rare run that doesn't reach 50%. Some runs exceed 60%.

To some extent this makes sense doesn't it, even based on what we know about wave theory? One would expect to be right about half the time by the flip of a coin. But because "the trend is your friend", it gains more when right than you lose when wrong. This is why letting profits run and cutting losses short is so crucial, and considered a successful to most workable trading approaches. This "system" if it is called that, attempts to do just that in a super simple way.

Now its your turn. Do it just one time. Run the Test DG! Do it because you know that the world is LOST without philosophers and grail seekers, as delusional and misguided as we may be.

By the way, I have since run a backtest on the the IBM 2 year chart, but don't know how to post the chart to show the results. The results exceeded 60% wins(lucky one). Yelnick, how do I attach the chart?

[DG]

Bird,

I'm not going to run your test. You ran it and if the results are good, they're good. It sounds too simple to me to be a real edge, though, because a hedge fund with access to a few hundred servers would be able to write an algorithm to monitor every listed security and trade that method. Within a few months, they would have all the money in the world, trading billions of dollars a day over and over just based on a couple of simple rules.

Sorry, but it doesn't pass the smell test. In fact, it kind of sounds as realistic as a perpetual motion machine. Whatever, though.

[Bird]

With some differences, I'm running a version of the concept in real time with real trades. I'll let you know how it goes.

[rr]

Yelnick,
Your last several weeks of posts have been exceedingly good. I look here first daily for your recap. You seem to bring the best ideas together and blend them with your own in a meaningful way.

RR

[joe]

You are being too hard on Bird, I think. He may be right in a theoretical world.

The real problem is that trying to do that in the real world runs into human emotions that will screw up the model.

I hope all is well with you DG - you seem like a stand up guy.

Joe

[joe]

Yelnik - you could probably start your own subscription service - and I would be a lot more inclined to subscribe to yours than any pundit you currently post about.

Joe

[Hockthefarm]

Thanks Y:

"(What he got wrong is how far it would go he expected a 6x type of wave as we had in the '20s to Dow40K. Oops!)"

Agreed. In his latest book he talks about it. FWIW, he is saying that the boom in the developing world plus the commodities boom pulled a lot of money away from the US markets (thus a muted boom here).

He talks a lot about the 29-30 year commodity cycle that apparently runs like clock work. The commodity cycle has peaked, it will deflate with everything else in the 2010-2012 depression (calls for a low of 10$/bbl) and then rise and fall into 2020-2023 before the next major advance.

Dent talks a lot about Keen's work (debtdeflation.com/blog). Private debt hit 42 trillion in the US at the end of 2008. Most is lent against assets like RE and business assets. In 1990 it was less than 10 trillion. The Obama garden hose is 2 Trillion and about to run dry. Good luck stopping deflation.

What struck me most is the strong correlation between private sector debt accumulation and financial sector profit. They are parallel lines. Even the children of people in congress could see the link and knew what it meant.

My bet is that in the next 5 years, stock prices hit new lows and that RE prices get trimmed another 25 to 35 percent. I hope to buy both lows. If I'm wrong, I'll work an extra few years and lower my standards. It seems to me that this year is a good one to work on my tennis game.

Thanks again,
Hock

[Hockthefarm]

Here is a better link to Keen's work:

http://www.debtdeflation.com/blogs/

Hock

[joe]

Hock - I am really looking hard at small multi-residential buildings. Less that 24 units/more than 8.

They will certainly get their clock cleaned here - and if you can go all cash, well then - you can be the only game in town as a buyer shortly.

Those loans are only starting to re-set, they will not be able to get the financing if they bought in the last six years. Most of the commercial loans in the past decade had 10 or less years due dates.

It is just starting.

Joe

[DG]

You are being too hard on Bird, I think. He may be right in a theoretical world.

I am just struggling to understand where people are coming from when I ask them about methods that can return hundreds of percent over a three-year period and the response is a general non-interest in favor of some theoretical model that can generate "better" returns. I'm not trying to be hard on Bird and he's clearly a bright guy, but I find it hard to understand how anyone can think there are trading methods that can be turned into 3 lines of programming code that will take 10 minutes to write and somehow they're going to earn market-beating returns on that idea. You got banks like Goldman employing hundreds of programmers, at a substantial cost. If winning trading methods were so simple, how could Goldman not have heard of them already, fired all of their programmers and just bought a supercomputer to pick up Fib-based trades (just to stick with this example) on a millisecond-by-millisecond basis?

I just think there is a disconnect between theory and the recognition of the amount of money and energy that gets poured into the market for even a few basis points of outperformance.

I hope all is well with you DG - you seem like a stand up guy.

Same to you, joe!

[joe]

"...how could Goldman not have heard of them already, fired all of their programmers and just bought a supercomputer to pick up Fib-based trades.."

I think they do. I also think if all us "little math nerds" focus - we can beat the pants off those a-ssholes. I really do. No one has a monopoly on the probabilites, but put together, maybe we do??

We rarely focus as a group - but if we really did - I think they will not be able to deal with us.

Thats real capitalisim.

Joe

[Bird]

Applause to both Joe and DG. I like what both of you say.

DG, I hear your point. How could some half-assed guy like me figure it out, with no research tools to speak of, and not the really smart guys at Goldman. Since I have no particularly good answer, I am thinking you are right.

I am actually testing a version that I've proposed, so I will find out for myself. One issue that I may find is that what one sees in a backtest isn't what one sees in real time. That is partly a limitation of the price movements within a bar chart, and partly a perceptual bias. On the other hand, I wasn't claiming that this is the holy grail. I was saying 50% with bigger wins than losses. Could make someone rich, but not the "holy grail" so called. Two points about Goldman: (1) they are not trying to figure it out. They are able to front-run trades so they don't have to figure it out. How exactly they justify this is beyond me. (2) Almost everyone on Wall Street believes fibonacci retracements occur because "everyone believes" they will occur. That seems to be the extent of the analysis. Wall Street, generally speaking, isn't looking under this rock. If they were they wouldn't say this.

[ponziunit]

Here's a look at a fibonacci spiral that seems to match up with this.

http://picasaweb.google.com/ponziunit/Ponzi_Unit#5425346523827738306

[DG]

Bird,

This link has a link to a Credit Suisse piece that looked at some common technical indicators, including Fib retracements. Yes, I understand you are just using that as an example.

http://vixandmore.blogspot.com/2009/06/open-thread-do-fibonacci-retracements.html

And I'm not saying it's because you're some "half-assed guy". I can tell from your posts that you've given all of these issues a lot of thought.

Could make someone rich

See, this is where I think there's a real disconnect for me. I don't think there is easy money to be found in the stock market. Even portfolio managers who produce bottom-tier returns work a 12-14 hour day. If those guys could write some simple code to buy and sell a bunch of blue-chip stocks on a deterministic basis and improve their returns while reducing their workday to essentially nil, don't you think they would? There are definitely some super-genius types who turn their attention to the markets one day and "Bam!" they figure out something with a flash of inspiration and it works, but I'm guessing that the number of people who fall into that category is minuscule and usually they are people like this guy, may he rest in peace.

http://www.latimes.com/news/nation-and-world/la-na-scientist-murder7-2010jan07,0,2571055.story

Anyway, point being that Occam's Razor tells me that IF there is a "Holy Grail", it must be at least one unit of complexity greater than anything that exists today, because if it were less complex than the most complex trading method in existence, it would have been found by now, given traders' willingness to endure complexity for the sake of profits. Therefore, unless you are doing work at the edge of complexity, your chance of finding the "Holy Grail", assuming that you are looking for it (not everyone working on the edge of complexity is), your chance of finding it is nil.

Is NeoWave complex? I don't know. I don't think NeoWave is the "Holy Grail" (I don't think it exists and if it exists, I'm not sure one could find it in time to enjoy its power, given our limited life spans. It's basically like searching a thousand haystacks for one needle, in my opinion), but I can't see how the "Holy Grail" could be less complex than NeoWave.

[yelnick]

RR, thanks, I appreciate the feedback!

[yelnick]

Joe, I don't expect the sort of volatility you are expecting, at least not this week. This market will roll over and fade, not drop like a rock, at least not at first.

[DG]

This market will roll over and fade, not drop like a rock, at least not at first.

I don't know, Yelnick. The market is priced for perfection and if Alcoa is any indication, we still live in an imperfect world.

Green closes 33 of the last 47 trading days, since the intraday price low on November 2nd. Out of 15056 chances for that to happen, it's happened 106 times, or 0.7% of the time. No volume to speak of and no retracements to "back and fill" at various price levels. I don't watch any financial news shows, but I can't believe anyone who goes on there can explain why anyone would buy this market other than because it's going up. So, once it stops going up, the only logical thing to do will be to sell.

Of course, my other market scenario is that Wall Street just keeps pushing this pig to new highs, reality be damned, in order to try to entice the retail investor back in. I honestly think the Street thought Joe Six-Pack would be back by now and they'd get a chance to sell him the stocks they bought at the March lows. Only Joe's too close to retirement to be screwing around in the market, in his mind anyway, so he's letting Wall Street keep doing what it's going to do without him.

That was the big thing I was following today, was a bunch of commentary on how the retail investor isn't interested.

http://seekingalpha.com/article/182019-left-out-of-this-rally-the-little-guy

This is the first time in a long time when the public just isn't interested and I don't think anything but new highs will get them interested, but I don't think new highs are feasible. The rush to the exits should be epic. My .02.

[vipul garg]

if somebody wants a statistical edge , a lot of trading systems have been backtested for it and the results are well documented..numerous exist which in backtesting and optimisation seem to give results.

there is little need for reinventing the wheel.

[Hockthefarm]

Joe:

Dent likes that space as well. Also starter homes. Sweet spot is cities of 1 to 2 MM people. From a timing standpoint, he feels it could come as early as late 2011. His time frame is late 2011 to early 2015 to meet the needs of the echo boomers.

Dent's book turned a few lights on for me. I now understand why my 401k doubled between August and december of 1999 (100 % in FDGRX), and why it will take 150 years for something like that to happen again. Second the process we are going through today has been baked in for many, many years. It is just time for the tide to go out. The world is not going to end, it is just time to shed the phenominal amount of debt average people have accumulated over the past 10 to 20 years.

Hock