This page has been archived and commenting is disabled.
SPY Hits 3-Sigma Divergence From VWAP
All those who have had a fishy feeling about this whole volumeless rally may be curious to take a look at the following graphic which indicates that unprecedented divergence between the VWAP on the SPY, which since the beginning of the rally in March has hit just over 91, and the actual SPY which as of last night was at 107. The difference between the two data series is now roughly 15 points, or about 150 on the S&P, or just under 1,500 DJIA points. This differential is entirely due to the low volume aggregation on the way up, in essence the entire run up has been a lite version of a 6 months long gap move up. What the chart also explains is the propensity by the market to see every potential sell off have a dramatically broader volume participation than the computer driven trickle higher, as all market participants realize there is insufficient accumulation interest to justify this 3 Sigma divergence.
Of note on the chart are the red boxes which represent the lateral move in the pure SPY needed to generate a 5 point move in the VWAP. As the width of the boxes increases it demonstrates, once again, that with time the rally has been of lower and lower quality (i.e., volume), meaning a much greater move in the actual index is needed to move a volume weighted representation. All this merely highlights that if and when sentiment turns, the whiplash will be dramatic as programs are reversed and instead of diverging from VWAP they start chasing VWAP. And, as expected, a synthetic and algo driven 150 pts move in the S&P should be sufficient to start a negative feedback loop of people running for the exits.
»
- 8414 reads
- Printer-friendly version
- Send to friend
- advertisements -
That's sounds just about right. My charts are calling for about a 1500 point drop in the Dow.
Tyler, I agree with the take away but I don't agree with the methods you are using to arrive at your conclusion. SPY is in someways a derivative of the broader index. Would a more accurate reading be obtained by using the index price and the exchange volumes or composite volumes.
Great stuff Tyler. I absolutely agree with the implication, the change in trend will likely be dramatic.
I have to wonder though, if these VWAP calculations are skewed by dark pools since they hide volume.
That so made me think of HFT - CIT is cutting it close hovering around 1 to 1.01. Sure can't roll that if it's < $1. Time to call out some more QE to keep that casino rolling.
EURUSD unresponsive to equity bounce, going lower
Get ready. Really bad eCON numbers = tank the dollar = ramp the equity markets = buy carry trades...
I only wish that was sarcasm.
John Brady on the dollar. Take note.
http://www.cnbc.com/id/15840232?video=1309977455&play=1
When we look around Casino Royale and realize nearly every other player is a shill, why play? There are alternatives.
thumbling dice, I think you might be very right. dark pool vols might make a great deal of a difference.
About 2 weeks ago I ran out of trading ideas. I saw that Exxon was in the mid 60s and I decided that between XOM and USO those were the 2 things that would work. Since the dollar is still on the floor it will probably be the only thing that works for a while, until policy is changed or the oil traders all suddenly go on one long simultaneous holiday. Oh, and playing SPY every 10 minutes up and down.
I too played XOM in the mid 60s a month or so ago, and sold out last week, a few weeks after dividend payout. However, as the dollar short has become consensus, MSM, CNN, WSJ, I have become very skeptical of the crowded trade. Bad economic news will cause oil/commodities to drop as the sheeple become scared that the all mighty hyper-inflation story doesn't play out in short time. After Bushs' tax cuts expire, and the Obama WH has to pay for everything, look out. I am not entirely convinced about the hyper-inflation trade. Yes, the Fed has printed $ worse than Parker Bros. But the Obama WH hates success, and the one way for Govt to get bigger is to ensure more people need it. Taxes rise > less $ for sheeple to spend > smaller deficits > stronger $.
But the Obama WH hates success, and the one way for Govt to get bigger is to ensure more people need it. Taxes rise > less $ for sheeple to spend > smaller deficits
...
Someone last night was talking about playing the game "Which of these things is not like the other..."
That would be a positive feedback loop...
I think he means positive feedback with negative slope.
Hi
What is the "average" trend in the graph?
How do you compute the boxes?
Thanks
Nice and the dollar books it back up to $120. All is well here. Nothing to see. Move along now.
Despite the well measured sarcasm, you may get your wish after Kenny's bullish count runs it's course. ;)
Hi
What is the "average" trend in the graph?
How do you compute the boxes?
Thanks
Please excuse my lack of understanding and bluntness in the following technical questions. VWAP stands for Volume Weighted Average Price? Are the prices and volume at the close used to compute this weighted, arithmetic average? If so how many days of data are used in this average? On the other hand, is this VWAP using all the trades, volume and price, for a given day? Is the horizontal axis time, frequency, cumquats, days, weeks, or hertz?
That was a little long winded. In summary, how many points are used to compute these averages?
Thanks. Cheers.
"...cumquats, days, weeks, or hertz?"
i grew up with a cumquat tree in my back yard. delicious little yellow fruit every summer. after the age of 13, i could never eat that fruit again. if u know what i mean.
Could be that a horizontal divergence hertz cumquats.
Often the horizontal divergence hertz cumquats.
I think the Presidential Plunge Team/Treasury will keep buying equities to support their political agenda. When Health Care and Cap/Trade bills pass; they will let the market collaspe. Perhaps that's when real accounting results will be allowed and QE will dissapear?? Its all a game.
I've been pointing out "price to volume" on many stocks (especially the financials) for the past 2-3 months.
3-Sigma Divergence, Delta Kappa Epsilon, Phi Delta Theta...ah yes...brings back good memories of the ol college fraternity days. :))
btw..is it just me or is this site getting really slow?
wtf? my math question to post this was: -29 x 37=?, with the answer being -1073. i got an error message saying: Math question cannot be longer than 2 characters but is currently 5 characters long. uhhhh...how do i abreviate -1073 down to 2 characters?
I've been stating for a while regarding the divergence in "price versus volume" -especially in the financials....since the March lows, the financials have been going up higher and higher on lower volume. The same is true of other stocks in other industries as well....
Nice analysis. Is there any doubt that the Fed and Treasury are behind (implicit or ex) this rally and will stay behind holding the market at this point at all cost. They cannot afford a double dip in stocks; too much rides on it. Goldman is likely handling much of the boiler room activity (and being well rewarded!) and using secret electronic Fed funds to play with. Bernake and Geitner know the severity of this economic decline; they have pulled out all the stops to stem the tide as they realize their culpability in its' formation particularly because they believe (a bit arrogantly I think) that great depressions can be avoided through knowledge of the past.
The .DJT is in the process of taking down the whole market right now.
Someone at the PPT pulled a Homer Simpson and fell asleep with a box of donuts on their chest over at the nuclear plant.
What the heck is the VWAP? Speak English man!
It's volume weighted average price. It's meaningful in that it tells you what the average unit of something sold for, which can be different than the quoted price.
For instance, consider the following prices and quantities traded for XYZ each hour during a 4 hour day:
$10, 1,000
$11, 1,000
$12, 100
$13, 100
the closing price would be $13 and the day's range would be $10-13. The "Average price" in the day would just be (10+11+12+13)/4 = $11.50
but the VWAP, as adjusted for volume would be (10*1000 + 11*1000 + 12*100 + 13*100) / (1000+1000+100+100) = $10.68
the average share of XYZ traded hands at $10.68, not $11.50. If the two diverge a lot, it means lots of marginal trades at the edges are "painting the tape" on price.
If tomorrow you rushed to sell your 2,000 shares of XYZ at $13 or even $11.50, the previous day's VWAP would probably tell you the demand might not be there.
The implication of all this and the article is that SP500 is up nominally, but on marginal volume at the higher ends that all looks like tape-painting, and so if and when people decide it's time to sell, the steady price climb (ahead of VWAP) hides the fact that buyers just might not be there at a price they like!
Thank you!
http://www.lmgtfy.com/?q=what+is+VWAP
Just curious:
Since March the Toronto Stock Exchange had a similar steady ramp-up since March on low(ish) volume (400-500 mln shares vs 600-799 mln up to then). However, the last two days have seen a sell-off in the TSX on even lower volume still (not even 400 mln today). Anyone?
>>the average share of XYZ traded hands at $10.68, not $11.50. If the two diverge a lot, it means lots of marginal trades at the edges are "painting the tape" on price.<<
Thank you! Makes perfect sense in this market. In "real bull markets" one sees the market chewing its way higher on good volume and pulling back on light volume. This market has wafted higher on light volume, and pulled back on higher volume.
The past few days the volume has really picked up in the big ETFs (SPY, QQQQ, and IWM). Downside volume. Not a good "tell."
As long as we're talking VWAP--if one dollar-cost averaged into the stock market for the past 10 years, what would be the "average value" of the DJIA or S&P 500 for "break even."
If the market shoots up, kind of plateaus for a few years, then drops back, most people would be underwater, no?
I dont know. Maybe someone can answer. Is it possible that the "dark pool" trading has pulled more and more of the volume out of the measure? Any idea what percentage of total reported volume might equal?