Stocks Plunge, Quickly Recover, on Fake Tweet
blogs.wsj.com
blogs.wsj.com
There are a few problems with this.
1) Forget execution and settlement risk, assuming you can get away with it, can you really make enough to make it worth your while.
You only get one shot at this, how many times can you anonymously crater the market?, and if you get caught you'll never trade again so you need to make retirement levels of money on this one "trade".
2) Execution risk. Assuming you've got your short butterfly setup so that no one will know you now need to unwind your trade in a matter of milliseconds while making it look like you are reacting, rather than knowing this event was coming.
Part of this is that like you said you'd do this over a basket of stocks. Looking at a butterfly option layout in wikipedia it seems pretty simple. Trying to get out of 30 or so of these trades in under 100 milliseconds isn't.
3) Settlement risk... You'd' be trading across multiple exchanges all with different timings and liquidity. There is a good chance that your trade gets busted by some but not all exchanges which means you go partial fills all over the place and you might end up partially short and long when all the accounting is done:(
In fact, isn't it fairly safe to assume that the SEC will spend a good deal of resources looking into any unusual trading patterns around this dip?
You are, of course, right if you know how far down the bottom is and that the market will rebound.
Maybe I'm just a coward, but all I see is risk:) Here are the big questions to ask yourself before trying this.
What if the market doesn't rebound, ie if you set your buy at a 1% drop and it drops 3% and then closes down 2.5%. That's enough to shut down alot of prop shops:)
What if it doesn't drop low enough to hit your bid, ie the market just ignores your buy and shoots way up. Then your short position crushes you.
Or if you use a short butterfly, what if the market just yawns, you've lost your one chance at moving the market but atleast you live to trade another day:)
A strange game. The only winning move is not to play.
Someone(s), made a LOT of money today. More than you or I will ever make in a lifetime.
[1] Butterfly - a limited risk investment that exploits volatility see [http://en.wikipedia.org/wiki/Butterfly_(options)]
The only way to figure out who was making trades on the expectation of this hoax would be to find the hacker, which probably means finding who controlled a botnet, which probably means convincing Russia to extradite someone. Good luck.
Eh, the percentage loss for DJI(https://www.google.com/finance?q=INDEXDJX:.DJI ) and the NASDAQ composite(https://www.google.com/finance?q=INDEXNASDAQ%3A.IXIC ) is around 1%, though. Unless you knew the exact stocks that would be disproportionally effected by this news, you're more likely to end up with around 1% net proceeds.
Which is a lot, but it's also what you could have gained just from being in the market this morning, or if you bought at yesterday's low and sold near close.
Don't even worry about the basket thing. You would just use SPY (S&P 500 ETF) or ES (e-mini S&P futures). If you can predict how far it will drop ahead of time (a big if) then the strategy is, as you have said, to leave resting buy orders at the bottom of the dip. You don't need to short though, you could just unwind when the position when market bounces back.
The tweet, which said that there had been two explosions at the
White House and President Barack Obama was injured, came
after hackers made repeated attempts to steal the passwords of
AP journalists.
But even financial websites pretty much provide a dumb questionnaire list with challenges like "what was your first dog's favorite color?" to which you can choose any kind of text string as a response ("bark" for example).
Does anyone honestly believe that these things provide additional security? It's like the TSA frisky-crotch-grope of authentication.
Yes. Especially as time-based passwords, a-la an RSA token or Google's two-factor auth, since they require something you know (your password) and something you have (the token). They expire and regenerate every minute or so, can't really be remembered or predicted, require access to a physical device that displays the password, and near impossible to predict without information about the seed.
I am a security zealot, and even I can't be bothered to have unique passwords on some - unimportant - websites.
It is so easy to set up for a user, and I might as well just have it, if it means all my sites are represented in Google's TFA app on my phone. It's no hassle to use at all, as long as the sessions last for a few weeks.
Two-factor auth involves entering a temporary code along with your password. The temporary code comes via SMS or a special mobile phone app. Google, Facebook, and Dropbox are examples of free websites that offer this.
What is with people and wanting people's jobs over every little thing these days? In the land of Twitter and Facebook, you're not allowed to make an occasional honest mistake any more.
I'm glad to know that you've never made an assumption based on authority, a minor mistake, or even taken a shortcut in your career, though.
IMO, there is not even theoretical public utility in immediately sharing a rumour, no matter how juicy, rather than confirming it and breaking it 10 minutes later. If you must, share it with a fat disclaimer that you're currently confirming it.
If a rule was passed that made it a firing offense to run with an unconfirmed tip, the entire news industry would collapse overnight.
I'm not saying that's a bad thing. :)
So the tweet would be integrated into the system, as would any reporting of it. :-/
My suspicion is that by the time other people start re-tweeting this the and you can get any sort of twitter sentiment determination the market will have corrected.
Are you taking the other side of this and hypothesizing that you can determine twitter sentiment before the market reacts? If so please mail me, I write in house trading systems for a living and I'd love to chat.
(1) In this instance, I'm pretty sure I don't know how to solve the problem naively. It went onto a theoretically legit account. I would have to cross-correlate with a couple other sources of information. At any rate, it points out to me that I need to encode a 'reliability' measure of the information.
(2) Not really. I don't have the capital to operate at low-latency levels. My hypothesis is the obvious one: markets are emotional and I should be able to construct a predictor for certain industries based on emotions emitted on accounts & news snippets relating to those industries.
Concerning, sure, but not nearly as huge as the graphs make it out to be.
For comparison, here are the top 5 that I found since 4/23/12:
(1) 4/23/13 @ 13:10: -0.56% (2) 8/1/12 @ 14:14: -0.45% (3) 11/29/12 @ 11:41: -0.44% (4) 6/20/12 @ 12:33: -0.38% (5) 12/31/12 @ 13:47: -0.35%
As best as I can tell from the news on those days, (2) and (4) were related to Fed announcements and (3) and (5) were related to the fiscal cliff.
In which case, hacking multiple relatively smaller information sources, rather than larger govt. or banking sites could provide very high upsides with much lower risks.
What is more likely is that human traders reacted to the tweet and caused microstructure changes that were amplified by algorithmic trading activity causing the rapid sell off. But then on the flip side those same algorithmic trades brought the price back to it's correct level just as quickly.
This is essentially a no harm/no foul event where the only people impacted are the game players.
Yes - Some are. Absolutely.
1. Acquire a large hedging position that cashes in if the market should fall.
2. Hack AP's Twitter account and post an alarming bogus news item.
3. Wait 15 minutes.
4. Sell your position.
Now -- who can say this isn't the real reason behind the hack?
Conveniently, you wouldn't have to hold any positions in advance, so you'd appear to just be a player who figured out the non-veracity of the news quickly and took advantage of it to beat the market. You'd also not have to sell right away--you could hold onto the positions indefinitely, if you liked. The only problem is predicting the exact inflection point of the drop (i.e. the best time to place the buy order)--but I have a feeling people might be "predictably irrational" in this, and fake-news stories about easy-to-verify facts might have a regular/well-known half-life. This is classic "insider trading", just applied to a whole market instead of an individual stock.
The only advantage of shorting in this case, I think, is if you didn't have the capital to actually make the money you wanted--you could do the shorts naked.
It's the modern equities market -- lightning fast, and not to the benefit of market stability or the average investor.
I was of the opinion that an individual hacker would still be classified as an average investor. He/she might not have had the tools to benefit from the end result. not sure
A big organisation, on the other hand, would be much better placed to exit a position based on custom algos and deeper integration into the markets to buy and sell positions.
Reasonable people may differ, but to me, someone who shorts the market, then hacks AP and plants a false story, is not an average investor. :)
It seems to show a pretty stable market to me. Think of it this way, what would have happened before electronic trading if a fabricated wire release had made it into newspapers? It would have taken weeks to sort out and "average investors" would definitely taken a hit.
The average investor, of which there are millions, can't react quickly to market changes. They're in a mutual fund or another similar investment that moves glacially if at all to market changes, and that's generally a good thing (on the premise that buy & hold is a good thing).
So there really is an average investor. And he really can't exploit fast market changes.
> It seems to show a pretty stable market to me.
Yes, but if this scenario is true, it represents an exploitation of a weakness in market information (some people have information that others don't), so it's an undermining of the premise of an efficient market.
I'm not sure what "scenario" you are talking about but the linked wsj article mentions that people reacted to a public tweet and just as quickly corrected itself after public statements from the white house. No insider information has been proven to be part of that.
The rampant and irresponsible speculation about this being some big short sell plot on this forum doesn't mean that the market behaved incorrectly. In fact, in reality it behaved exactly correctly. As soon as the hoax was uncovered the dow returned to it's correct level. Showing stability.
No one said that. I said it could be used that way, not that is was.
> doesn't mean that the market behaved incorrectly.
Any time someone exploits knowledge that isn't public, it's an exploitation of a market weakness. If someone takes a position in a stock with the intent of bad-mouthing its competitors or releasing false information, he's exploiting a market weakness.
Some exploitations of private knowledge are merely bad behavior, and some, like insider trading, are illegal. All of them represent weaknesses in the market because all investors don't have the same chance to benefit.
> In fact, in reality it behaved exactly correctly. As soon as the hoax was uncovered the dow returned to it's correct level.
By your reasoning, pump & dump proves that the efficient market hypothesis is at work in the market. I don't think so. It only proves that some investors can game other investors.
Taken to its extreme, if someone was able to game the market consistently like this, day in and day out, businesses would recognize that they were being cheated and would refuse to raise capital using equities. This is why the SEC is so aggressive about locating and punishing examples of this class of behavior.
> Showing stability.
By that reasoning, if we have the same number of bank robberies on Wednesday as on Monday, we've proven that things are stable.
The only thing we know is that a news source incorrectly reported bad news. The market responded to that bad news in the way we would expect and then when it was obvious that it was a hoax corrected itself. That is a stable system.
It would have been proof of an instable system if either A) the market moved for some incomprehensible way or B) did not correct for obviously bad behavior.
The market took all public information into account and corrected itself quickly. Precisely as we would expect in an efficient market.
The fact that the Tweet's content was false was non-public information for a short time, just long enough to see the market begin a plunge. Obviously whoever posted the false Tweet knew it was false, and they could have exploited that fact (not to say anyone actually did this), while other people were reacting to it as though it were true.
> Where was the gaming?
See above, and use your head. A pump and dump always pivots on the fact that the perpetrator knows what he's saying is false.
> Without proof of those things all speculation about insider information is moot.
Feel free to change the subject. We have already established that this is a hypothetical discussion.
> It would have been proof of an instable system if either A) the market moved for some incomprehensible way or B) did not correct for obviously bad behavior.
What does stability have to do with it? The topic is exploitation of predictable market moves, not unpredictable ones. Obviously a hacker could (in principle) anticipate, and act on, a coming market drop on news of an attack on the White House.
> Precisely as we would expect in an efficient market.
According your thesis, insider trading is impossible because the market will magically adjust to differences in information. If this were true, the SEC wouldn't care about insider trading. The fact that the SEC does care, demonstrates that your position is wrong, and that the market can be gamed.
A market cannot be efficient unless everyone has access to the same information. This is how the efficient market hypothesis is defined. And again, the SEC does what it can to assure that people can't exploit non-public information. The reason? It undermines confidence in the market's fairness and would ultimately cause businesses to avoid equities as a funding source.
A hedge fund manager who has decided to unload a huge position but has not acted on it yet would cause your definition of the hypothesis to no longer hold. This was never the intent and if it were it could only exist in some perverted Platonic cave of a market.
In the real world, this showed a very efficient market. 2 pieces of public information were consumed nearly as fast as they could be produced and the correct prices in the market were reflected. This is what the real application of an efficient market would look like.
As far as what does stability have to do with it, your central premise was that the modern equities market was not stable and was not to the benefit of an "average investor". None of this was proven by today's event and was in fact largely disproven. Stability was not jeopardized and the average investor was not impacted by what could have been a malicious attempt to game the system or could have been a juvenile prank, either way it largely didn't matter.
Are you trolling? The efficient market hypothesis is a hypothesis, and it's based on the premise that everyone has the same information. Which word didn't you understand?
No one knows whether the EMH is either valid or that it in any way underlies the behavior of the real market. It's ... wait for it ... a hypothesis.
http://en.wikipedia.org/wiki/Efficient-market_hypothesis
Quote: "In finance, the efficient-market hypothesis (EMH) asserts that financial markets are "informationally efficient". In consequence of this, one cannot consistently achieve returns in excess of average market returns on a risk-adjusted basis, given the information available at the time the investment is made."
After reading the above, has it escaped your attention that the EMH, as defined, assumes that everyone has the same information? Common knowledge is a working hypothesis, just like the EMH itself.
> ... your central premise was that the modern equities market was not stable and was not to the benefit of an "average investor".
Only when private information informs trades, and I made this perfectly clear. This is why the SEC rigorously prosecutes inside traders, because it's true -- when private information informs trades, the market isn't fair.
> None of this was proven by today's event and was in fact largely disproven.
Nonsense. The market fell based on a lie -- on private information. This proves that information must be both shared and public for the market to work as it should.
> Stability was not jeopardized ...
The market commenced to plummet based on false information. The only reason it didn't melt down entirely is because those who knew it was a lie, very quickly said so. Exactly which part of that fact do you find to be mysterious?
> 2 pieces of public information were consumed nearly as fast as they could be produced and the correct prices in the market were reflected.
Nonsense, and you are trolling. Your position is that the lie about an attack caused the market to react appropriately. But the market's move was based on false information, therefore it was not a "correct price" -- someone was gaming the system, and whether or not it was for private gain is irrelevant.
> ... and the average investor was not impacted by what could have been a malicious attempt to game the system ...
False. average investors, and all straight players, and "impacted" by insider trading -- it represents a threat to public confidence in the market. That's why the SEC prosecutes inside traders, throws them in jail.
Exactly which part of this everyday reality is causing you the most confusion?
The market can be manipulated by inside traders and people who put out misleading or false information for private gain. The SEC does all it can to address this very real threat to public confidence in the essential fairness of the market.
Circle the world you you don't understand and raise your hand.
There is no mention in any of the forms of the EMH that I know of, that short term information asymmetry cannot be used for short term gain. Only that information asymmetry is unsustainable in an efficient market. If anyone had proposed otherwise they would be laughed out of the trading floor.
Further, no one claimed that market confidence isn't eroded by insider trading, only that market confidence should not, and will not be eroded by the events of today. An oddity, that corrected itself and left the vast majority of portfolios un-impacted.
Further, I posit that the fact that the market corrected itself without outside intervention, no circuit breakers, no governmental control, should be seen as a sign of stability, not as a sign of weakness. If you are a long term investor, nothing that happened today should concern you, rather you should be happy that hackers pulling pranks cannot impact your portfolio for more than minutes.
http://digital.library.unt.edu/ark:/67531/metadc5130/m2/1/hi...
When you buy a stock, and it goes down in price a few months later, the experts tell you to buy more of it as it goes down, because it'll probably go up in the future.
When you gamble $5 on a blackjack hand, Martingale experts will tell you to gamble $10 on the next hand, and then $20 on the next hand again if you lose the $10 hand.
Both are gambling.
That's not to say there aren't plenty of other transactions that may blur the line a little more. Hell, entire options markets exist to make a commodity out of risk.