2012 09 13 spy crash soon

Chart update 2013 Feb 28 : see flic.kr/p/dYuoLU


Chart update 2013 Jan 06: see flic.kr/p/dJ82Fn

Math update 2013 Jan 03: VXX vs SPY charts in scatter-plot form

zoomed in: flic.kr/p/dHpavN

zoomed out: flic.kr/p/dHiJzF


MATH UPDATE 2012 Nov 26: correlations are moving very far out of whack.

On a window of 4 months + 10 trading days I calculated a correlation slope of -12x meaning VXX traded down much quicker than normal as SPY rose on average. HOWEVER, I also found something on the 2 month + 10 trading day time window: a POSITIVE Correlation of 4.6 - VXX & SPY dropping together. To see everything clearly a scatterplot will be needed log-scale for each axis. Sorry I have no time yet for this, but do try it yourself.

Scatter-plot features are in OpenOffice & LibreOffice, data can be obtained from freestockcharts.com using the disk-platter icon once you specify several chart ticker symbols for 1 output chart & save as CSV for import. (end 2012 nov update text)


edit 2012 10 05 : 4:1 reverse stock split in effect, reading around 34 (not same on all sites) on vxx therefore old figure 23.68/share would now read as 94.72 maximum


Very soon SPY (DIA, dow, s&p500, qqq, nasdaq) will crash

see also: stocktwits.com/message/9521252 , flic.kr/p/bnbsG5


See also flic.kr/p/cXAU7y


Related concept: (picture) from (article)


ALSO related & nearly a copy of this picture (not a copy: it's using Bezier and NOT powers/logs to find inflection points) : article & picture


original math from 2009 calculations:


  • Fxgenius. 3y

    Hey Gp thanks for the update ! Yes I agree with your analysis that the dip will be coming very soon. I was estimating that it would start today however its consolidating still . I guess it will break this week end. I'm anticipating next week to be bearish
  • goldpricemodel 3y

    Looking at the time-scale with the boxes this break could be days to 4 weeks. The scale below is 3 grids = 1 year so around 17 weeks = 1 grid-span
  • elpuerto93 3y

    SP-500 is in a bull market, I would look to buy on pullbacks to late august lows 1395 - 1400 with confirming candles that show support.
  • lizadfuel 3y

    Think you will find this sight interesting. www.investmenttools.com/equities/s_p_500__standard___poor...
  • goldpricemodel 3y

    elpuerto that's what's called a bull-trap and heading down to the read line if you did such a thing that's what's called "catching a falling knife" or "getting your face ripped off". If you dare go long S&P 500 via futures or options make sure to hedge by having a balanced measure of SPY puts and if you can short a little SPY directly on margin please feel free. Going all long at the levels you specify is a suicide mission. It would have failed, and people called for it, 2010 march and 2011 july.
    S&P is not a bull market it's a RIGGED market. There are NO fundamentals. There are FEW inputs from anywhere in the market that are NOT central bankers.

    Candles do not confirm any show of support.

    Repeat. Candles CAN NOT confirm any show of support.
    What can show a supporting trend is the rate-of-change metric rolled forward. Set ROC to 280 days and watch it unfold as per this image - flic.kr/p/cXAU7y

    Right up to 2007 was a "bull market" for the S&P500 too. Until it smashed down a giant amount. You know why? Because the entire situation was a rigged sham designed to lead suckers in then plunder all their funds. ESPECIALLY 401k's, RRSP's and hedge funds.

    Don't be a sucker. Going long S&P500 at levels above 1200 is tantamount to voluntary bankruptcy.

    You need to understand this in terms as clear as possible.
    The very notion that support and/or resistance is a single linear value is a fraud. It's a fraud perpetrated by various market-speakers and tech-chart sites. It isn't true at any time. At the very least if you get a linear function you'll get a slope & need (usually) to apply it to a log-scale graph which immediately is not truly a linear price-scale or truly a single value. Lines on log-scale charts are actually exponential curves.

    In particular rather than going bankrupt going long S&P500, ES1, anything correlated with it at any positive leverage, I've detected 3.5x to 4.6x leverage without margin for VXX negatively correlated to S&P500. Going long VXX should get a maximum of 167% profit before fees. Your profit will vary based on the fees & dilution over the lots of shares you get at your average fill price across all lots.
    I intend to exit with 100% profit or if I get jumpy just 20% net of all fees.
    SPY is going down. Even with QE-infinity.
    You see, I understand math and I hope you do too. The Fed's new plan is a constant monthly purchase of 40 $billion USD. This will fail.
    Initially a fixed total for a fixed time was announced. This is QE1. And QE2 was Operation Twist, much the same but targeted on swapping short-term for long-term bonds.
    These failed & QE3/infinite is intended to address all front-running by announcing NO fixed total and NO time-line.
    The RATE is still detectable and generally fixed (40 billion). the only way to smooth out the S&P curve to avoid all dips is to drop the QE-rate when SPY is doing well on its own and MASSIVELY ramp it up to avoid the dips. That could mean 1 billion at peak or less (monthly) and 800 billion (in ONE month) for a calendar-target landing on a dip I outlined.
    LINEARLY purchasing a constant 40 billion a month smoothes NOTHING. It can at best push the curve up.
    HOWEVER the curve I plot INCLUDES prior QE thereby showing a CONTINUATION of new QE will not break the highs, lows, curvature-formation (logarithmic curve) at all.
    The cake IS A LIE.
  • goldpricemodel 3y

    Indeed, lizadfuel, on www.investmenttools.com/images/weq/charts/sp500log.gif I can see clearly a new pattern emerges. In addition to CPI being understated clearly a new periodic trend has formed for S&P. Without QE, something introduced only in the 3rd period, 2013 clearly sees a smashdown to the 775 level +/- 14% (I can only eye-ball from the graph but I'd suggest that covers 90% of probabilities).
    Regarding a later image www.investmenttools.com/images/de/p&f/spx.gif and also www.investmenttools.com/images/ss/cycle_length.gif I was able to check a few variations & came up with this:
    scharts.co/NwIOxT (long URL is stockcharts.com/h-sc/ui?s=SPY&p=D&yr=0&mn=8&a... )
    This clearly indicates a trend (roc 25) from +7 to -7 (1.07 to 0.93) over a period of 17 weeks +/- 1% in the band-gap.
    The shift back up from -8% to +6% (0.92 to 1.06 factor) is rapid (5 weeks) the later trend is 2% to 4% (1.02 to 1.04).
    If one was to trade in profit on such a pattern one would need good leverage or lots of funds or choose to write calls < 10 days to expiry on SPY shares and likewise on the dips in the ROC trend write put contracts.
  • elpuerto93 3y

    If I would have listened to all the top callers during this rally, I would have never entered the market. I agree the market looks like it is making a toping process, but I am not going to try and make that call. That's an easy way to lose money. If you are not sure, stay in the safety of cash.

    Looking at the rsi, we have a monthly divergence, weekly divergence and starting a daily divergence. So I would not be surprised a top is coming. To try to trade this as a top is not practicing proper money management.

    Many people make this mistake and blow up their account, because their analysis could be right, but they timed their entry wrong. Or they hold on to their losers longer than they should, because they believe their analysis to be true, while cutting their winners to early because they fear losing any profit.

    Trading is more money / trade management than your analysis being right. Basic rules to stay ITM with your account so as it grows. Plan your trades with entries, stops, and exits.
    Use appropriate trade size in relation to your account. Use proper money management. Cut your losers, and let your winners run.

    I find its better to trade an established trend where I catch the middle of the trend than try to hit home runs by calling tops and bottoms. Base hit trades can add up if you are consistent with your trade/money management.

    I agree with you QE3 is a big FAIL and the markets will not be propped up by it. I have no idea when a correction will actually happen.

    I agree a reversal could be close. Counter trend trades could be good, look at what happened to many on thursday, bear trap set, and as you said, "had their faces ripped off" because they were top calling. There is another saying, "don't stand in front of a steaming locomotive". Just because a market looks like it wants to roll over, doesn't mean it will, markets can stay in over bought conditions for extended periods.

    At this moment I would not feel comfortable going long and would prefer a pullback. 1400 seems a likely place of support, round number, former level of resistance, and at the bottom of a daily channel from june to today. If the daily chart formed a huge hammer at this level, I would look to go long. Once in the trade, I would set a tight stop so as to cut my loser fast. If it goes in my direction, then I will bring my stop to BE asap, then follow price with my stop, while locking in profits. If the daily trend channel is taken out then I would look to sell. The monthly channel puts the sp500 downside at 1200, but it also has room to go to 1500.

    Don't fight the market and try to tell it what it should do, instead let the market tell you what it currently is doing and follow the trend, even if it is against what your personal opinion is. Take all your personal opinions and put them aside and trade objectively with no preconceived points where you believe the market will go.

    Good luck with your trading!
  • goldpricemodel 3y

    Well I'm not a top-caller, I'm moving with the market. The market moves up and down and the place to make a trade is at the change-over, what's called in calculus the "inflection point", where the curve meets parallel, briefly to the tangent to its curve.
    I've simply shown for SPY what those points are.
    As it happens bottoms and tops are trends, trends are bands of trading pattern with a price being between those bands, however they may be defined by a fractal pattern, data-set or some function for approximation (which is what you see in the curves I provide).
    "1400 seems a likely place of support, "
    Again I return that no linear (a number) function of support exists. The very notion support is a single number is itself mistaken for all charts at all times.
    It's not price=1400 , it's not price= 1470 resistance, it's not price=1120. See those curves I showed?
    They are not locked to a single value.
    Each of them is linear pattern after a kernel-function transformation.

    for each price use 1.05 to the power price, base=1.05. Any base works but this one provides output numbers you can do with pen & paper, easy scaling.
    Use lines to connect peaks & dips, the inflection points in the data, showing the real trends.
    Now re-scale back by inverting the kernel-function:
    for each data-point in each line plotted (overlay to the price-transform data) you now apply price= log(base 1.05, price_transform). Voila, there's the curves - like this ( flic.kr/p/a1db7f ) and this ( flic.kr/p/avnwvp ).

    I'm telling you flat out that when you go against what's showing here you are fighting the market. What I'm showing is how to move with the market.
    "Once in the trade, I would set a tight stop so as to cut my loser fast"
    Good luck with your trading but as I said, the cake is a lie.
    NANEX has fully exposed that when you use stops you are exposing your poker-hand. You are showing what you can not tolerate as a boundary line and the HFT's read your data from your sitting orders and poach you. They deliberately pick off each of your orders using your stops as a guide and then return the market in the direction least suitable to your trade short-term.
    That's why you make a note (as you said) of your planned entry & exits and while you may use an algorithm it should disclose NOTHING to the market, or you can do it manually which is fine so long as you have planned carefully.

    "If it goes in my direction, then I will bring my stop to BE asap"
    NO, you will not.
    The markets are being front-run so you're seeing a feed delayed by fractions of a second, thousands of trades, ahead of you. Before you can react the market has already moved, even if you're using an algo to trade - you're just too far away in precise multiples of wavelengths of light from the data's origin and the HFT machines are not. It's that simple. You're too far away so all you see is a distant echo. You can't trade an echo.

    "If the daily trend channel is taken out then I would look to sell"
    There is no daily trend-channel in s&p500, nasdaq or DOW, e.g. SPY, QQQ, DIA. None. It does not exist.

    Between the curves I showed you, the extremes, there are fractal patterns (not a channel) you can apply using "expectation value" once you program in parameters for the inflection-event based on prior data fitting the fractal.
    A daily price channel implies a simple band with easily plotted boundaries into which the price falls for daily prices. This channel flat out does not exist.

    A pattern which does exist, however, is revealed in the ROC (rate of change) time-derivative.
    Set interval to "daily" and the parameter input to "280" and you can see the output chart linearize. This shows between 2 ends of a consistent time window the price movement is linearly proportional.

    What it MEANS is that a repeating time-cyclic pattern has been detected.
    See image flic.kr/p/cXAU7y
  • Fxgenius. 3y

    Hey gp I'm seeing a reversal in the trend i was blindsided today. i was biased towards the bears but it seems like we are going up. I'm now bullish as of today. I'm anticipating a bullish week. the pull back has been complete. I'm calling a buy at the current price of about 1776 with a limit order at 1799. just a bit shy of the 1800 mark to lock in that profit.This is a home run of over 150 pips with small risk. Write me and let me know that your in on the move. take care.
  • goldpricemodel 3y

    It's better to ease in, reduce risk, average in to faz, rsw, vxx and be prepared to hedge short-term with calls for spy, fas or rsu. This is the worst possible time to be bullish, I'm 100% bear on this one for s&p500, dow, nasdaq. As for gold I'm not bullish or bearish, I'm following the chart I made. Short-term drop to 1600 maybe 1500 is more than likely, repeating 2008 and 2006-7 chart activity as per the normal long-term manipulation cycle.
    The problem you're having is likely not direction, it's likely the time-risk vs margin-size risk so regardless of what direction you pick you can get knocked out by volatile moves.
    Pick both directions on paper & determine what price moves opposite are "too heavy" for you to bear the way you've been trading. Determine if either is likely & if so, do not use either/both trade-styles.
    At this time my only exposure to gold & silver price action is physical coin in hand.
    Previously (and profitable for moves of gold > $95/oz up or down in time period < 120 days) I have attained decent lower-risk profit with zero margin using a Strangle of GLD call & put, same distance from the at-the-money price. I suggest danger happens with time < 45 days. You raise price & lower risk by adding more time until a larger break-even after which adding time only cuts out all hope of profit & you're better to sit on the cash.
  • Fxgenius. 3y

    Hey gp thanks for the advice. I'll Definately bear this in mind. I've decided to change my screen name to Fxgenius but same old trader here.
  • Fxgenius. 3y

    So i changed my screen name from Fxnitro to Fxgenius :-)
  • goldpricemodel 3y

    being determined for a directional trade, reducing risk, is almost the same approach. If you can reason that one direction is say, 5x more likely than the other, and if you can put a time-frame / window on it, then so be it, take a gamble. Figure out what opposite direction is too hard for you, don't commit more funds than that, and jump in. I like Puts or calls on inverse ETF's (or inversely correlated etf's, e.g. vxx is inversely correlated to spy but is NOT an inverse etf officially), but they expire. Shares get much less leverage without margin but some etf's still offer some effective leverage & you can still reduce total dollar risk by judging the max you're willing to lose, including fees, & choose the # shares appropriately.
    I've been doing it in my head lately but 2 years ago I started up a spreadsheet - I'd put in the max spread I thought was attainable on the chart, the amount of $ I'd be willing to spend, my trading fees & it would spit out the number of shares needed just to break even. Obviously you need more shares than that & of course from the share price + fees it will spit out the total input cost you'd need. Therefore if the specified profit on top of that is too low in % or the total input $ is too high you'd reject the trade for now. The ticker might be good for later but the conditions at that moment may suggest no-play.
    The math works fine the other way too, putting in the desired maximum $ and putting in the minimum desired % gain, then seeing the chart pop out (including fees) the real price you pay per share to get in, the real chart price to get your desired net % gain, and then you can try to conjecture upon the chart if/when such a price can be achieved.
  • Fxgenius. 3y

    Thanks go this is some sound money management advice. I just checked the market and it shows that we are going up though. The USD index is in going down for a pullback today thus causing an uptrend in gold prices. My target is 1799. I'm paying close attention to the USD index and my Fibonacci swing on the four hour chart. The final extension is at 1810. When these two targets get hit and the USD index resume its uptrend I believe we will be seeing able dip in the gold market.
  • goldpricemodel 3y

    the USD index is actually not very useful. Very much it's an inverted formula of the EUR/USD ratio, highly overweight. Look at prices of oil, gold, silver each priced in all currencies & this will be far more useful.
  • goldpricemodel 3y

    MATH UPDATE 2012 Nov 26: correlations are moving very far out of whack.
    On a window of 4 months + 10 trading days I calculated a correlation slope of -12x meaning VXX traded down much quicker than normal as SPY rose on average. HOWEVER, I also found something on the 2 month + 10 trading day time window: a POSITIVE Correlation of 4.6 - VXX & SPY dropping together. To see everything clearly a scatterplot will be needed log-scale for each axis. Sorry I have no time yet for this, but do try it yourself.
  • Fxgenius. 2y

    Hey All ! I'm swing trading gold. I took a short at 1468.47 with a target at 1405. I believe FOMC minutes release tomorrow will result in this bearish correction. After this support gets hit. imm planning to resume buying gold again. Please feel free to share your charts opinions and analysis Cheers !!!
  • goldpricemodel 2y

    I think I'll go long slv calls, cheaper for the same linear gains expected vs gld calls, bought silver for delivery already. May or may not get some more but got 2 deliveries of .999 already
0 faves
Uploaded on September 14, 2012
  • Show EXIF

Additional info

  • Viewing this photo Public
  • Safety level of this photo Safe
  • S Search
    Photo navigation
    < > Thumbnail navigation
    Z Zoom
    B Back to context