Ok - I was nice and relaxed Sunday evening, popped open a few websites to check for evening business news, and came across the latest shitstorm that's circulating with ZeroHedge in the eye of the storm.
Last week Zerohedge wrote a post about Goldman Sach's continued dominance of the principal program trading statistics that the NYSE publishes weekly, and about the overall ramp up in program trading volume. Anyone who knows anything about program trading knows that this streetwide increase was due to expiration: large baskets of stock are traded against expiring index futures and options, and the S&P 500 also does a quarterly rebalance on the same day. Thus, the program trading numbers for the 3rd week in June are always large. I took the time to comment on Tyler's post, "and yes - last week's large program trading numbers are from June expiration - and yes, this weeks will also be huge due to Russell Rebal." I even added "and there will be a massive jump next week in the "customer facilitation" numbers on the weekly program trading statistics. There is no conspiracy theory to be derived from it - it's the Russell data."
See, I know what I'm talking about when it comes to program trading, and this happens every year. The Russell Rebalance is the biggest event of the year, resulting in massive volumes. Of course, the weekly report came out, and the customer faciliation numbers were up massively. Although this is completely normal, and I'd even mentioned it on Zerohedge itself, Tyler still chose to characterize the data as follows: "This week's NYSE Program Trading report was very odd: not only because program trading hit 48.6% of all NYSE trading, a record high at least since the NYSE has kept tabs on this data, and a datapoint which in itself was startling enough to cause some serious red flags," before getting into the real oddity, which I agree with, and will get to in a minute. I can't tolerate the fact that he chooses to mislead people with hype of "startling" data which should raise "serious red flags," when I had even taken the time to explain to him a week ago that this data would be coming! There is nothing surprising, conspiratorial, or doomsday about it!
There was something strange about the report - which I again took the time to mention as a comment on another Zerohedge post four days ago, as I think this is actually interesting: Goldman Sachs is completely missing from the report. This is a virtual impossibility for Russell week.
Sadly, the story quickly morphed into a self-feeding conspiracy theory, with Tyler @ Zerohedge and Matt Goldstein at Reuters somehow each quoting the other's article in their own story. Naturally, hype-first-ask-questions-and-do-proper-research-later savant Karl Denninger hopped right on the paranoia bandwagon.
There is definitely a story regarding Goldman's sudden falling off the program trading league tables, but it is most certainly not, as Reuters' Goldstein suggests
"On the week ending June 19, Goldman was ranked first on the NYSE program trading list. But on the week of June 22, Goldman mysteriously didn’t appear on the list of the top 15 firms at all. It simply vanished without any explanation. Then the NYSE stopped reporting the numbers. The Zerohedge blog was all over this controversy a week ago."
We know, if we simply read the report from the NYSE, which I explained last week, that 1) The NYSE did NOT stop reporting the data. The report will continue, using automated, more accurate data, and that 2) the "NYSE stopped reporting the numbers" foil hat explanation is especially ridiculous considering that the changes haven't gone into effect yet! I would have expected more journalistic integrity from Reuters. Anyway.
Perhaps Goldman also misread the rule change (unlikely) and stopped reporting their DPTR numbers a few weeks ahead of time - this is extremely unlikely. Perhaps they totally pulled their quant models from the marketplace, as a result of the theft of some intellectual property, which Tyler covers extensively in his post. Still, I am very confident that GS would have still had significant program trading volumes to report. Perhaps there was a simple error in the report and Goldman's data got wiped off the grid portion - this seems most likely to me, as the numbers in the report appear to be inconsistent, with the "Total for all member firms" volume being lower than the "total for top 15 reporting member firms" volume, which shouldn't be possible.
There is an interesting coincident story which the Zerohedge post focuses on - that of a former employee of what appears to be the Goldman Sachs quant desk, who basically stole the code from his former firm and brought it to his new firm. He was stopped by the FBI on his way back from Chicago. This is a standard intellectual property case, although Tyler Durden is asking why the FBI is involved. Well jeez - for a guy who has a PhD in Goldman Sachs conspiracy theories, TD should have been able to answer his own question: you don't mess with Goldman. If you do, they release the hounds on you.
Unfortunately, the two stories get combined into a clusterfuck of a conspiracy theory that is sure to light up the blogosphere tomorrow. I would have really liked it if someone with the will and the means to get to the bottom of the mysterious program trading statistics without Goldman Sachs, such as Zerohedge, had stuck to that question first, instead of morphing the story into a global financial espionage conspiracy that they claim should require a dicslosure from both Goldman and the NYSE (no doubt about the pending immolation of the stock market) as a result of a former employee stealing quant code.
oy vey.
-KD
Last week Zerohedge wrote a post about Goldman Sach's continued dominance of the principal program trading statistics that the NYSE publishes weekly, and about the overall ramp up in program trading volume. Anyone who knows anything about program trading knows that this streetwide increase was due to expiration: large baskets of stock are traded against expiring index futures and options, and the S&P 500 also does a quarterly rebalance on the same day. Thus, the program trading numbers for the 3rd week in June are always large. I took the time to comment on Tyler's post, "and yes - last week's large program trading numbers are from June expiration - and yes, this weeks will also be huge due to Russell Rebal." I even added "and there will be a massive jump next week in the "customer facilitation" numbers on the weekly program trading statistics. There is no conspiracy theory to be derived from it - it's the Russell data."
See, I know what I'm talking about when it comes to program trading, and this happens every year. The Russell Rebalance is the biggest event of the year, resulting in massive volumes. Of course, the weekly report came out, and the customer faciliation numbers were up massively. Although this is completely normal, and I'd even mentioned it on Zerohedge itself, Tyler still chose to characterize the data as follows: "This week's NYSE Program Trading report was very odd: not only because program trading hit 48.6% of all NYSE trading, a record high at least since the NYSE has kept tabs on this data, and a datapoint which in itself was startling enough to cause some serious red flags," before getting into the real oddity, which I agree with, and will get to in a minute. I can't tolerate the fact that he chooses to mislead people with hype of "startling" data which should raise "serious red flags," when I had even taken the time to explain to him a week ago that this data would be coming! There is nothing surprising, conspiratorial, or doomsday about it!
There was something strange about the report - which I again took the time to mention as a comment on another Zerohedge post four days ago, as I think this is actually interesting: Goldman Sachs is completely missing from the report. This is a virtual impossibility for Russell week.
Sadly, the story quickly morphed into a self-feeding conspiracy theory, with Tyler @ Zerohedge and Matt Goldstein at Reuters somehow each quoting the other's article in their own story. Naturally, hype-first-ask-questions-and-do-proper-research-later savant Karl Denninger hopped right on the paranoia bandwagon.
There is definitely a story regarding Goldman's sudden falling off the program trading league tables, but it is most certainly not, as Reuters' Goldstein suggests
"On the week ending June 19, Goldman was ranked first on the NYSE program trading list. But on the week of June 22, Goldman mysteriously didn’t appear on the list of the top 15 firms at all. It simply vanished without any explanation. Then the NYSE stopped reporting the numbers. The Zerohedge blog was all over this controversy a week ago."
We know, if we simply read the report from the NYSE, which I explained last week, that 1) The NYSE did NOT stop reporting the data. The report will continue, using automated, more accurate data, and that 2) the "NYSE stopped reporting the numbers" foil hat explanation is especially ridiculous considering that the changes haven't gone into effect yet! I would have expected more journalistic integrity from Reuters. Anyway.
Perhaps Goldman also misread the rule change (unlikely) and stopped reporting their DPTR numbers a few weeks ahead of time - this is extremely unlikely. Perhaps they totally pulled their quant models from the marketplace, as a result of the theft of some intellectual property, which Tyler covers extensively in his post. Still, I am very confident that GS would have still had significant program trading volumes to report. Perhaps there was a simple error in the report and Goldman's data got wiped off the grid portion - this seems most likely to me, as the numbers in the report appear to be inconsistent, with the "Total for all member firms" volume being lower than the "total for top 15 reporting member firms" volume, which shouldn't be possible.
There is an interesting coincident story which the Zerohedge post focuses on - that of a former employee of what appears to be the Goldman Sachs quant desk, who basically stole the code from his former firm and brought it to his new firm. He was stopped by the FBI on his way back from Chicago. This is a standard intellectual property case, although Tyler Durden is asking why the FBI is involved. Well jeez - for a guy who has a PhD in Goldman Sachs conspiracy theories, TD should have been able to answer his own question: you don't mess with Goldman. If you do, they release the hounds on you.
Unfortunately, the two stories get combined into a clusterfuck of a conspiracy theory that is sure to light up the blogosphere tomorrow. I would have really liked it if someone with the will and the means to get to the bottom of the mysterious program trading statistics without Goldman Sachs, such as Zerohedge, had stuck to that question first, instead of morphing the story into a global financial espionage conspiracy that they claim should require a dicslosure from both Goldman and the NYSE (no doubt about the pending immolation of the stock market) as a result of a former employee stealing quant code.
oy vey.
-KD
6 comments:
Ah, if only people were as objective and level-headed as you. But you don't get ZH/Denniger level traffic, do you?
Anyway, I think an explation of why GS was omitted would be in order from GS and/or NYSE. Should be fun to watch how GS squirms out of the theft story though!
yes - there is certainly an explanation needed. The proper method should be to get the answer and THEN write the story, don't you think? rather than hyping a global financial espionage conspiracy theory, the key tenets of which are CLEARLY incorrect (1 - the NYSE is not stopping the report, 2 - this week's volumes were NOT unusual)
Thank you for your posts! I'm a mom and pop investor trying to sheppard our nest egg through the storm. I stumbled on ZH via Seeking Alpha. It's interesting stuff. But, I'm also a scientist - Occam's Razor and all that. I also know something about Erlang and the kind of infrastructure GS needs to use it. The idea that someone could steal their code, take it home and make a profit is laughable. I was getting uncomfortable with the tin foil hattedness of ZH. Your posts confirmed my uneasiness. Thanks again!
well, it's likely that someone DID steal the code - and if they make it profitable, SO WHAT?!?! that doesn't hurt the rest of the world. it just hurts Goldman. And no - it doesn't destroy the firm.
Sure, they could have stolen the code but that would be the equivalent of stealing the plans for a Vegas casino. Could you do some damage with the code/plans? Yep. Could you use them to set up your own trade desk/casino in Chicago? Not likely and to what end? I don't know enough about casinos to extend the analogy but there's no way someone's going to use that code to front run GS or any of the other wild speculations by the commenters on Tyler's post.
Nice post you got here. I'd like to read something more concerning this theme. Thanks for sharing this material.
Sexy Lady
Female escorts
Post a Comment