This page has been archived and commenting is disabled.
SPY vs ES: Who’s leading the S&P 500?
Recently, the Pragmatic Capitalist (ht ZH) questioned who the mysterious large buyer of eMini S&P 500 futures (ES) was, when there has not been a corresponding large buyer of SPY ETF shares over the same short periods. We decided this warranted a further investigation to assess, in the larger context, whether it is SPY or the ES that has historically had the greatest impact on the S&P 500 cash index, and the results were surprising.
Methodology and Notes: After normalizing the volume for both the ES and SPY over one minute periods going back to June, 2001 (when suitable data begins), the relative difference in volume was multiplied by the change in one minute price of the Cash S&P 500, with a running total kept for the ES and SPY (similar to Effective Volume and Larry Williams A/D calculation). The first minute of each trading day was not used so as to eliminate the effect of opening gaps. Only SPY trading hours were used for the ES (9:30 am to 4:00 pm EDT). The gap opening phenomenon associated with the ES futures is important, but a subject for another post. A further caveat–S&P 500 pit futures exceeded the ES futures in total dollar amount traded until a few years ago and were more significant in influencing the cash S&P 500; however, we don’t have intraday volume data for it, and it was not considered. This preliminary study does not attempt to weigh the effects of the actual stocks that comprise the S&P 500 and implicitly assumes it is possible to lead the cash S&P with ES futures and/or SPY.
Relative effect of ES and SPY volume on Cash S&P 500
The weekly chart, above, demonstrates that SPY volume (scaled left) has been basically on permabid since the beginning of 2002, with the ES only aiding the bulls (at least during the day session that we are measuring) from mid-2005. The ES was showing chinks in the rally’s armor in Q4 2007 and led nearly the entire bear leg, with SPY relenting and sharing the lead down only during February 2009. It may be worth remembering this when we get another down leg.
In the daily chart, we’ve highlighted some interesting inflection points. After the ES led most of the early rally from the March 2009 low, it led the June correction, after which SPY took the reins into the end of August. Curiously, since then it’s been nearly flat, while the ES has led most of the down ward legs. This is due to the fact that SPY’s upward and downward contributions have been netting more intraday, which suggests that either (1) the bullish firepower has been temporarily directed to the ES (though the fact that the ES’ net contribution is trending down is ominous), or (2) that there is institutional selling in SPY that is counteracting the upside firepower.
Fortunately, there may be some trading insight to be gained here. In future posts, we will attempt to determine if interim bottoms and tops in the S&P 500 can be anticipated and confirmed by these measures and their derivatives, especially when taking into consideration divergences between the two and by delving intraday. Until then, it looks like SPY picked up the slack yesterday where the ES left off, so we should be safe until the opening bell.
»
- EB's blog
- 12431 reads
- Printer-friendly version
- Send to friend
- advertisements -
And just look at today, Far East markets mixed, Euro area flat but US futures storming ahead on thin out of hours trading...
Dow future has made up half of yesterdays losses just like that!
What a complete con
ES trades 23.75 hrs a day SPY trades 10.5 hrs a day. Seeing that there is a 12.75 hr difference this cannot be normalized and the fact that 1 share ES is not equivalent to 1 share SPY is also a big factor.
You can take this study and throw it out the window. It is an apples to watermelons comparison.
I've been watching and posting about the /ES-spy volume discrepancies for months now over on TF.
It is an excellent way to predict a continuation of a trend. If /ES leads on the way up, most likely that trend will be dominate through the day. If it leads down, that will dominate.
The best way I've found to track this is to check the volume of each for a given time period (I usually use an hour) and compare it to the 30d average for that same time period. Lets say /ES is at 120% of it's 30dma and spy is at 100%. There's a 20% difference between the two, when there shouldn't be.
Under normal conditions, they run with less than 10% difference between them. When it gets to 20 I start paying attention. At one point today, /ES was 135% higher than spy for one 5 minute time period. and averaged 40+% most of the day. To the downside for the most part.
That's INSANE and was indicative of the comments ben was making about really heavy institutional action in the pits.
Excellent tip Asimov.
cash, spx, emini arb programs explains this
If, last fall, you were down 61 dollars, and now you are only down 14 dollars, then how many dollars have you made since last fall? That's "X".
It's always struck me that as the numbers involved are less than a few million or billion, it usefully/unfortunately/conveniently prevents GS etc from contributing to this blog. In the interests of fairness perhaps we should have the likes of: "If you are paid a $10M bonus and an nightly entertainment package costs $5000, how many can you afford?"
Ah, but there's no simple answer to that question. It would depend on the bonus recipient's tax bracket and how creative the recipient can be in writing off the packages as a business expense.
Ah, but I find I can wriggle out of that, I think you'll find that by arguing you are doing "God's work" you are no longer taxable? For instance
All praise be to the Church of GS.
Of course I suppose that there might be opposing quotes ("render unto Ceasar" or whatever), but hey, this is a free market, you can choose which horse you back. Although perhaps some might favour this one too:
Isn't bible study so confusing, quite like the markets really; as you can pick what ever quote you like to support your point of view that God-like association is quite realistic pehaps.
Can't we also look at NQ? I don't know how many times ES wanted to sell off and with just a few hundred contracts NQ stops the move lower in its tracks.
I would also argue that a single stock can stop NQ which in turn can stop ES.
The implication that SPY trades are used to drive the index to a greater extent than ES perplexes me because margin requirements for ES, being low, would provide ES with much higher leverage. Or an I wrong to assume retail margin requirements where the traders are the Broker/Dealers themselves?
(PS: "-61 + X = -14"? Those CAPTCHA math questions are getting kind of hard!)
Join the club on these CAPTCHA. Have to actually think which is a bummer when just wanting to comment
Actually, he would love nothing more than to see GS blow up, but there are limits fighting the government.
You don't have to believe me. I am just telling you, there's a player out there who dwarfes all the other players including the squid . And considering the damage he has inflicted over the last two year, I am surprised that they let him still play. Although, as I mentioned earlier, he's been restricted on the S&P to 200-300,000 lots a day.
He tried to corner the gold market two years ago and he was "advised" not to take delivery.
It shows one thing though: financial journalists or just journalists aren't doing their job properly as to this day I am astounded that nobody has exposed them.
explain please? who, why, where firepower from?
http://zerohedge.blogspot.com/2009/04/visible-hand.html
SPY is arbitraged against both the cash market as well as the futures (ES), and there is a complex 3-way relationship between them.
To determine which is leading and which is following of the 3 (ETF, futures, cash market), you need to look not only at volume, but the timing with a granularity measured in micro-seconds (this is one of the most crowded arbitrages, so the opportunities are quite fleeting).
There is sometimes a skew between the cash market value and the ETF because the ETF is being driven (at times) by futures/ETF arbitrage. A good MS Thesis paper from Ronald Tarantino on the subject can be found here: http://w4.stern.nyu.edu/emplibrary/Ron_Tarantino_honors_2004.pdf
Beyond the lead-lag relationship analysis (which needs the much finer time resolution to be meaningful), it is also important to understand which futures contract is being traded, and the number of days until it rolls.... as you would expect a convergence of the cash market to the futures market (and therefore the ETF as well since it should only diverge from the cash market due to futures arbitrage) as the time period to the futures roll-over approaches 0.
You may find what you *believe* to be correlations between the change in volume on the ETF and futures and their corresponding future price direction, but I suspect that it will be erroneous (i.e. no such signal exists) on time scales measured in units greater than milliseconds.
Thanks for your comment and the link, which I will read. I'm aware of the arbitrage that occurs in microseconds, the futures premium, increased volume on the contract roll, etc. However, by aggregating small minute by minute biases over time, there may be worthwhile information to be gleaned. A similar methodology has been successfully applied to determine accumulation/distribution in stocks and futures.
There is really a crazy player out there with almost unlimited firepower ( although he's been restricted lately by the authorities to only 200-300,000 lots a day on S&P and he is hiding his hand pretty efficiently). One of his goals is to crash markets (so you can understand that he's a bit frustated these days).
I know this is hard to believe but it's unfortunately true. You can see his prints on the market when there's suddenly a huge sell order out of nowhere and the SP tanks. He was at it yesterday on S&P and also on the oil market.
restricted lately? huh?
that player is the vampire squid
if it were anyone else, they would be shut down inside of a day because the manipulative purposes are so obvious.
trying to crash things? are you some sort of disinformation agent for the monster? we're up at new highs, 8th day in a row of sickeningly steep rally AS THE ECONOMY IS DYING.
I think all 4 of you "anonymouses"...(or is that "anonymice"?)... must be 1 guy with 4 voices in your head...
Schizophrenic - clozapine = 4 voices in your head...
If you reply please let me know if it is anonymous voice 1,2,3 or 4... chronilogical order will do...
FOLLOW THE YELLOW BRICK ROAD TO RESTON TO FIND THE '6'
http://fofoa.blogspot.com/2009/04/covert-market-operations.html
the vampire squid conducts fed financed blasting of es in order to MANIPULATE the market higher.....duh
How exactly is 'volume effect' calculated?
I believe you have uncovered one of hundreds of proprietary algos directed at outfoxing the hens.
Nice post EB, the subject is worth further analysis and let's see if there are connections/ causations as opposed to correlations (which seem tenuious so far).