Forgot your password?
typodupeerror
Bug Businesses

Computer Problem Caused Price Errors on NASDAQ 160

Posted by Zonk
from the that's-an-oops dept.
buckthorn writes "An article running on Yahoo News states: 'A computer problem at an unidentified stock trader caused erroneous, exaggerated prices -- some as high as $950 per share -- to be posted to the Nasdaq Stock Market Friday morning for 1,680 different stocks, a spokeswoman for the Nasdaq said.'"
This discussion has been archived. No new comments can be posted.

Computer Problem Caused Price Errors on NASDAQ

Comments Filter:
  • by TripMaster Monkey (862126) * on Friday May 13, 2005 @04:44PM (#12524643)


    Good afternoon gentlemen. As you are all no doubt aware, I have perfected a method of manipulating the various stock exchanges throughout the world. You received proof of this this morning, as relatively worthless Nasdaq stocks such as Maxco, Inc. and J.W. Mays Inc. traded briefly at hundreds of times their real value. I believe my latest caper, which I've puckishly dubbed 'Operation Stocking-Stuffer', is certainly worthy of your attention...

    You see, gentleman, when 'Operation Stocking-Stuffer' is deployed in earnest, all stock exchanges will be laid waste...all trade will effectively cease, and global civilization itself will crumble...that is...unless you pay me...



    One hundred billion trillion fafillion dollahs!!!

    (cue dramatic music)



    Gentleman, you have my demands...peace out.

  • by pudding7 (584715) on Friday May 13, 2005 @04:45PM (#12524654)
    Sell! Sell!
  • Friday 13th (Score:2, Insightful)

    by LiquidCoooled (634315)
    THIS is the reason I try to stay in bed on Friday the 13th.
  • I call do-over!
  • by Anonymous Coward on Friday May 13, 2005 @04:47PM (#12524677)
    A problem where technology caused wildly erroneous stock prices? I liked it the first time around when it was called the dot com bubble. The parties were better.
    • by Anonymous Coward
      How about the second time around, when the wildly erroneous prices were caused by SCO FUD?
    • by Anonymous Coward
      This issue was actually caused by trading software provider Trading Technologies (who are currently suing all of their clients due to supposed patent violations). It was caused by a bug in their automatic spread trading module, which caused it to run amok on one of their client's trading consoles.
  • Did the little PI symbol for the Praetorians show up on the NASDAQ ticker, too?
  • by robyannetta (820243) * on Friday May 13, 2005 @04:49PM (#12524704) Homepage
    Just as long as the price of this stock [yahoo.com] wasn't artificially manipulated by $950 per share, we could have been in hot water.
    • Oh to be Warren. That'd rock. A lot.
  • by Anonymous Coward
    Those 100 SCOX shares really paid off!

    W00t!!
  • For example, Berkshire Hathaway* trades at nearly $90,000 per share. Saying that some shares were trading at $950 per share does not indicated the magnitude of the problem without knowing which shares those were. * I know Berkshire isn't listed on the Nasdaq, but others like CheckPoint trade in 3 digits so I'm sure $950 isn't unheard of on the Nasdaq.
  • "COMPUTER" error (Score:5, Insightful)

    by whoever57 (658626) on Friday May 13, 2005 @04:52PM (#12524735) Journal
    I just love those "computer errors". Did we jump intosome future world like that of of the Terminator, or The Matrix, where the robots are in control?

    It was either a programming error (human) or an operator error (human) or some other cause. OK, clearly hardware faults can cause data corruption, but this is really pretty rare and it's hard to believe that a hardware fault could cause such a widespread fault.

    • It was either a programming error (human) or an operator error (human) or some other cause. OK, clearly hardware faults can cause data corruption, but this is really pretty rare and it's hard to believe that a hardware fault could cause such a widespread fault.

      Another good reason to eliminate The Human Factor. Let the Robot Age begin!
    • by Phisbut (761268) on Friday May 13, 2005 @05:08PM (#12524905)
      It was either a programming error (human) or an operator error (human) or some other cause.

      I love this quote :

      At the source of every error which is blamed on the computer you will find at least two human errors, including the error of blaming it on the computer.

      • Thank you, sir! I now know what will new hires in my company have included in the induction process. Regular checks of their knowledge of this matter will be implemented, too.

        Yes, I am sysadmin guy, why do you ask?

      • "including the error of blaming it on the computer."

        You musn't forget Dr. Alfred Lanning's "ghost code"...
    • Re:"COMPUTER" error (Score:1, Interesting)

      by Anonymous Coward

      OK, clearly hardware faults can cause data corruption

      In the case of hardware failure, it would be the person implementing the systems who made the error, for not adequately accounting for this type of problem (e.g. RAID, clustering, etc).

    • "but this is really pretty rare and it's hard to believe that a hardware fault could cause such a widespread fault."

      It'd also be pretty rare for a fault like this causing robots to suddenly self-replicate and rebel.
    • BEHOLD, Dijkistra is posting slashdot comments from the grave.

  • SCOX? (Score:3, Funny)

    by WillRobinson (159226) on Friday May 13, 2005 @04:52PM (#12524737) Journal
    I thought this was why scox went up, but when we looked at the time it went up, it didnt conside with the error. Oh well, guess its just a pump/dump for monday.
  • Why is there a PHP icon at the top of the page, which should be a Firefox icon.
  • It is possible (Score:3, Interesting)

    by dimss (457848) on Friday May 13, 2005 @04:54PM (#12524756) Homepage
    I am working for local internet shopping centre. Most prices are generated automatically based on wholesale prices and competitor retail prices. Sometimes glitches in this process lead to funny prices on website. When this happens sales reps are ready to kill us IT guys.
    • When this happens sales reps are ready to kill us IT guys.

      Local Police are reporting tonight on the death of the entire IT staff at a local internet shopping centre. The victims were beaten senseless with several tennis rackets and golf clubs shortly before being run over by a Corvette, three Porches, and a BMW 5si.

      Police refuse to confirm or deny the use of expense reports in the binding of the victims, although a couple bloodied credit cards were found at the scene next to half a dozen Venti Latte

    • Re:It is possible (Score:3, Interesting)

      by MenTaLguY (5483)
      Given the impact of advertising funny prices, it seems like it might be wise to throw a safety net into the process -- e.g. determine "reasonable" price ranges for different classes of items, and flag any prices that exceed that range for manual approval (and perhaps adjustment of the "reasonable" range).

      Of course, how practical that is does depend on how many different categories of products you have to deal with. You can take an initial crack at "reasonable" via an arithmetic mean or something, but "rea
    • Re:It is possible (Score:2, Interesting)

      by Moofie (22272)
      "local internet shopping centre"

      A what?
    • I had that happen once. I was looking around a (small) net-based shop, and found a number of expensive items usually priced at a couple of thousand bucks for 1/1000th the actual price. Just for the fun of it, I ordered all those, but unfortunately, someone did notice the glitch and sent me a polite email, telling me that those prices were, unfortunately, not the real ones. Ah well. :)
  • I'm rich! (Score:2, Funny)

    by kihjin (866070)
    Those transactions, which were made 15 percent above or below the previous day's closing price, will be "broken" -- the buyer will get his or her money back, and the stock will revert to the seller.

    I wont let them. They can't take it back.... right? ...

    What the hell is banging on my door? They are insi#%*#&)^*!&$NO CARRIER
  • by southpolesammy (150094) on Friday May 13, 2005 @04:55PM (#12524769) Journal
    Michael Bolton: I must have put a decimal point in the wrong place or something. Shit. I always do that. I always mess up some mundane detail.
  • Profit? (Score:3, Informative)

    by xiando (770382) on Friday May 13, 2005 @04:56PM (#12524773) Homepage Journal
    "A metal heat-treating company that normally trades between $3 and $4 per share, was briefly quoted at $951.47 Friday morning. It later traded at $4.10 per share." There IS money in the stock market. Invest $40, sell for $9510. Profit: $9470! Cheers and congratulations.
  • by tktk (540564) on Friday May 13, 2005 @04:56PM (#12524780)
    Just look at SCO's share prices.
  • Damnit! (Score:5, Funny)

    by Gogogoch (663730) on Friday May 13, 2005 @04:57PM (#12524783)
    Damnit, I could have made some money! Except that my funds are tied-up in a Nigerian opportunity at the moment. But boy-o-boy is that opportunity gonna make me rich! Rich! I say!
    • I know you were probably joking, but anyone who actually sold their shares at those high prices will have to give back the money (actually, they never got the money in the first place, as settlement is 3 days after the actual trade).

      NASDAQ has a policy on "Clearly Erroneous Transactions".

  • instead of Red Hat today.

    Oh, wait, I would have made even more ...

    Darn.

  • by hsmith (818216)
    Microsoft blames IE Share loss [slashdot.org] on NASDAQ computer errors
  • its a problem (Score:3, Interesting)

    by Anonymous Coward on Friday May 13, 2005 @05:02PM (#12524847)
    working for a large bank on a program trading system (yes it runs on linux) automating orders on behalf of clients this is the sort of thing that gives us the cold shivers.

    one coding error can suspend the trading of a major stock, or worse as in this case move the price miles from where it should be.
    theoretically the exchanges shouldn't allow you to do this. some work by suspending automatically and restarting with an auction, some work by suspending it. locking you out and fining you loads of cash. this costs you big as nobody else can trade that stock through you for a while. thats really expensive - the lost business.

    when your taking in client orders and trading them automatically and you recieve orders via say FIX.
    yes you do lots of checks with vol, movement on the day, movement since the close, momentum etc.
    but how well do these checks work when your market data feed has gone down (or worse gone wrong) or even gone down, but the heartbeat process is still pinging so you think you have a good price?

    but i guess sometimes we get it wrong just like anywhere else. this is just a rather high profile and embarrassing example.
    still they won't do it again (for a while)
    • Can you say... 'in flight fills'? The number of exchanges which don't support delta quantity is insane - and FIX doesn't support it either, officially... Total bloody nightmare, if you ask me.

      We haven't had anything go too badly wrong. Yet.
    • Very insightful comment. I too worked for many years in the FX sector, and oversaw a few trading systems being build.

      I get shivers because a few mistakes we caught in beta like certain currencies where the buy and sell have been reversed, etc.

      Either way, the whole point with big money systems is to have everything worked out in PRE-PROD and tested and stream live.

      Looks like somebody missed a step somewhere, but rolled it back quick enough. Just wish I had my hand on the sell button soon enough ;)
    • Re:its a problem (Score:1, Informative)

      by Anonymous Coward
      I've worked for a number of banks on various front office trading systems and I can tell you that the risk of major error is an ever present possibility, and in fact many non-US markets will reject orders with prices beyond a prespecified high/low for the day to prevent wild and inadvertant price fluctations or runaway meltdowns.

      The quality of trading systems varies tremendously. I've seen some that can easily handle many hundreds of thousands of orders quickly and reliably with a minimum of hardware inv
    • working for a large bank on a program trading system

      I'm sure it's interesting work, but do you realize that your software skills are being used to create widespread instability and volatility in the stock markets? Banks try to constantly arbitrage whatever they can (quickly take advantage of price discrepancies, exploitable market inefficiencies) and if you are using a lot of leverage in the process, you are contributing to a stock market that has become nothing more than a speculator's playground.

      • Re:its a problem (Score:2, Insightful)

        by Anonymous Coward
        exploiting arbitrage opportunities merely has the effect of closing the gaps and ensuring fair price for investors.

        a market cannot work without people arbitraging, speculating and investing.

        his skills are being used to reduce transaction costs and inject liquidity into the market, without which pension funds would operate at a much higher cost.

      • Actually, his software skills are being used to bring stability to the markets. Arbitrage is almost a non-occurrence in today's markets, as hardly anyone has big price discrepancies. But even if banks do take advantage of these kinds of events, rare as they are, it actually brings the market more in line with what it should be. Removing price discrepancies through arbitrage removes instability, rather than creates it.

        All investing is speculation. You can take some risk out of it by doing lots of research,
  • So you're telling me that as long as you're a trading house with a direct tie-in to NASDAQ, there's no input validation or sanity checking done? I realize its probably too much to ask that there be a moderate amount of human oversight given the extreme amount of data passing through the lines. But given the importance of accurate data, you'd think NASDAQ would put better checks and balances into their systems.
    • I guess they need some validation code to alert them of a problem.

      if(new_price > old_price + 900.0)
      {
      // holy shit!!
      int fd = open("/dev/ticker", O_WRONLY);
      while(1)
      write(fd, "TILT! ", 6);
      }
    • Re:Input Validation? (Score:1, Informative)

      by Anonymous Coward
      I don't deal with the NASDAQ, but I can tell you that on electronic derivatives exchanges there are often "circuit breakers" that are tripped when certain volitility thresholds are met. The problem with most of these is that they look at a few ticks (quotes or trades) before they go off. So a single erroneous price in the system causes big problems. There is also nothing stopping me from buying a share of apple for $1000 if I so choose. Most likely these trades will be rolled back.

      This btw, is not the fi
    • "you'd think NASDAQ would put better checks and balances into their systems."

      Why would you think that? It's not like these people are going to be held accountable for their errors...
  • by G4from128k (686170) on Friday May 13, 2005 @05:13PM (#12524948)
    In the extended hours markets, some traders post bids or offers at outrageous prices, hoping someone will make a mistake. They will offer to buy some stock for $0.02 per share or sell some stock for 200.00 per share (that normally trades $20 a share) in hopes that someone screws up. The low cost of participating in an electronic market makes it easy to post these orders.

    The real lesson is that stocks don't really have "a price" in a traditional sense. (At best, the price on the last transaction serves as a proxy, but is no guarantee of getting that price in the future). In reality, stocks have both a bid and an ask price. For thinly traded stocks (especially in the off-hours), the bid-ask spread can be very very large.

    Buyer (and seller) beware.
  • by bnitsua (72438) on Friday May 13, 2005 @05:37PM (#12525147)
    after some poking around, I believe the unidentified trader was scotttrade. if you use the ticket on their website, it's the only one I've found that reports the incorrect highs (such as maxco being traded for $951.47). you can find other ones if you look hard enough...
  • 2 Revealed! (Score:2, Funny)

    by Aggrav8d (683620)
    Step 1: Buy worthless penny stock
    Step 2: Await computer error
    Step 3: Profit!!!1!

    In Soviet..erm, I'd like to see a beow...uh, I for one, welc... oh, just forget it.
  • Ha ha! Funny stuff (Score:4, Informative)

    by bigberk (547360) <bigberk@users.pc9.org> on Friday May 13, 2005 @06:09PM (#12525432)
    Believe you me, the NASDAQ is one of the only major exchanges I mildly trust precisely because it is electronic. Other major exchanges, notably the NYSE, involve human floor traders gathering around posts and barking out bid and ask prices.

    Do you have any idea how crooked stock trading through middlemen is? There are a thousand ways the retail investor and small trader gets screwed. For instance, market makers are definitely not impartial and favour their own trades ahead of clients'. You can not even catch the fraud the occurs. There are about a dozen NYSE market specialists that are charged with fraud every year.

    There is absolutely no reason to involve humans in the securities trading process any more. None! The rampant fraud can be easily avoided. When things like this are publicized, I almost wonder if it's got some bias in favor of the human trade specialists who make trading floor operations tick. They're useless middlemen, profiting from spreads and leverage.

    Electronic trading is the only way to go. When an exchange switches to electronic, you should see that as a sign of quality and a commitment to do away with the fraud that EVERY insider knows is a standard mode of operation in stock trading.
    • Humans are still involved in electronic exchanges, and there's just as much scope for fraud. In the end, the trades are between people (a computer can't own stock) and they're the ones making the decisions. The fact that they do it through a terminal rather than in open outcry is neither here nor there.
  • But did any trades actually take place at that price or was it just listed as the current price?
    If I had been interested in Maxco, Inc. stock, I would be watching the price and waiting for it to hit around $4 before buying, or perhaps be willing to buy it up to $5, if I thought it was going to keeep going up.
    If it was at $951, I wouldn't buy it, since that's orders of magnitude higher than I expect it to be in a week.
    I would assume wall street traders similarly limit the price when they buy.
    (I do pity t
    • There is a branch of investing called technical analysis. What people who follow it do is use various formulas to look for stocks that are moving. These people intentionally know nothing about the stocks they are trading.

      Technical analysis works because it assumes that others are doing their homework, and so you can follow their lead and get almost as much gain as they do. When prices jump as happened in this case, technical analysis would say buy (depending on the formula of course) assuming someon

  • by djrok212 (801670) on Friday May 13, 2005 @08:31PM (#12526222)
    I was actively involved in todays "mess." The root cause of the problem was related to the prices which were using for electronically making markets in stocks which did not have any trades yesterday. When there is no activety for a stock in a day, the price quote is always 0 x 1900. The median of these two numbers is 950. A firm who makes markets in ALOT of stocks, accidentally changed their systems to use the midpoint at yesterdays close instead of the normal final print. So they started putting up quotes at 950.00 thus causing the problem. All the trades 15% out of the market from the previous close were broken. The money wasn't "removed" from peoples accounts since the proceeds or loses from a buy or sale of stock aren't realized for atleast 3 days.
    • OK, so the people who bought at those erroneous rates didn't lose their money. But what about the people who sold at those erroneous rates and made plans based on the money they believed they had made?

      I think this is unjust, personally. I think stock exchanges should never cancel erroneous trades, because that penalises those who took advantage of errors, but does not penalise the source of the error. The source of the error should be penalised, on the "it's your fault, you pay for it" principle - which w

  • about money (Score:2, Insightful)

    by Numtek (839866)
    Are we now all ready to acknowledge the fact that money is nothing more then a set of bounding rules? It was never introduced to be taken advantage of, like the first world is doing everyday. I believe any system will break down eventually, if it's not used in harmony with the idea the founders of that system had in mind. Can't realy imagine people back in the days would be proud now, seeing people die without food/water, while others whine about their stock-value.

Work without a vision is slavery, Vision without work is a pipe dream, But vision with work is the hope of the world.

Working...