Flash Crash Update

Discussion in 'Trading' started by psytrade, Mar 9, 2011.

  1. sprstpd

    sprstpd

    This notion that there can be a VPIN contract that somehow enables market makers to "stay in the market" is completely flawed. Who is going to provide liquidity for the VPIN contract when the crap hits the fan? No one. There will be a flash crash in the VPIN contract and all the market makers will call it a day. You cannot guarantee liquidity during dislocations. Anybody who thinks otherwise is delusional.
     
    #31     Mar 11, 2011
  2. nitro

    nitro

    I think the posts in this thread are missing some important points:

    1) The liquidity in the markets that appears from the DOM is very flightly and will only be there in quiet markets. It is a HUGE extrapolation to assume that it will be there under any market condition.

    2) HFTs act as surrogate MMs under quiet conditions, but have no obligation to get run over in the precense of evidence for a statistical event that predicts a strong move, measured by a took such as VPIN.

    3) Even in markets that supposedly had designated market makers, in the presence of a crash, they simply refuse to make a market. Read about the 1987 crash. So if a human being refuses to stay in front of a freight train, why would one expect an algorithm to do so?

    I never understood why people argue the obvious. If I can predict that I am going to get run over, "I" [bot] sit there and let it happen? HFTs add considerably liquidity when markets are quiet, and walk away when the statistics say they should. What could be simpler or more sane?

    The only way to prevent crashes is by having "cooling off periods" like automatic "stop-trading" triggers like they implemented. Now the hope is that those rules don't get manipulated.
     
    #32     Jun 29, 2013
  3. sprstpd

    sprstpd

    My feeling is you can't prevent crashes. The price goes to where it wants to go. If you have limit-up, limit-down days, where trading ceases, it seems like it would cause even more panic because you can't act. Are there any legitimate studies or instances that show that a cooling off period actually works?
     
    #33     Jun 29, 2013
  4. I agree crashes can't be prevented really.

    When hell breaks loose, not many want to be in front of the bulldozer whether you are a human or algo.

    Remember last time in the flash crash they canceled stock trades below a certain threshold but not futures trades. So if there is a crash again, trade the futures . I remember specifically freaking out and calling my broker making sure my trades would not be canceled in the futures market.

    Some of those stocks trading at 1 penny , just crazy.

    Did they cancel any trades in etf's like 3x leveraged funds ect?
     
    #34     Jun 29, 2013
  5. edit: (3x wasn't around at the time, because full margining of 2x instruments was still allowed).

    Proshares busted a QID print from above 16 to the next tick below 8 when the price would be higher than 24 where it should have been.

    In a short summary on Iranian capability, the scheduled launch of nuclear tipped compatible missiles in that country were appeared to have been targeted toward a US tank battalion near its Iraqi border.

    Market's work, and this is what happens in that situation.

    Iran called it testing, I called it manipulation, and they made many millions knowing that that would be the reaction by twisting the rocket in our Army's direction during their testfire.

    Market makers have already chosen to be adversely selected for these situations, so to say that another derivative could prevent the FF from happening is not going to be any different than what happened. MM's have futures, and that's what they're for. This contract sounds like automatic puts with free call options the instant they transmit their trades with an OSO OCO pegged to the spread at an exact percentage and price to the probable loss before their next trade.

    Good info to keep in mind, but as apparent as the need for these instruments are, the fact is that most of the MM's simply feel like good old boys that shouldn't have to pay us Qnerds for these kinds of protections.
     
    #35     Jun 29, 2013
  6. Other conspiracies think it was because of the vote on "Audit The Fed Bill" was in hearing on the same day. There was a more watered down version passed a while after the crash.
     
    #36     Jun 29, 2013
  7. Look, this (my theory during and thereafter was probably the threat of imminent nuclear war and not the Fed Stabilization Policy Uncertainty) was confirmed on CNBC, (with video documentation taken in Iran), but released a couple years after, and without much emphasis either because they probably were told not to make much of a fuss about it and at the time I heard that it did seem like the information should not be public.

    Thus, you can count my theory as "conspiracy", but it is not.

    To your busted trades question, that was the only conflagaration that received the blessing of the exchange to take away those trades in QID. Proshares now on the NYSE is much more liquid than it ever was at AMEX.

    The futures contract Lopez is suggesting is a micro category only for market makers who can set sub-penny prices up to 100,000 times per second now. The enhancement would allow them to basically be a SOES bandit again, just with millions of shares in inventory. TOUCHE!
     
    #37     Jun 29, 2013
  8. So Iran had nukes pointed at our military and you think that was a big cause of flash crash? I'm not saying you're full of shit but
    didn't North Korea also ? North Korea basically said in public the US was in their sites yet no flash crash.

    BTW faz fas 3x was here since 08 so I missed whether any trades on that got canceled. I don't remember the bank stocks trading at ridiculous 1 cent levels though.
     
    #38     Jun 29, 2013
  9. NK didn't actually launch in SK's direction, or so my thinking goes. (Scared me a lot to the point I wanted to bail on my trades, though).

    But if I see a financial reporter on CNBC discussing the video taken at the time in Iran, I usually take it to be legitimate.
     
    #39     Jun 29, 2013
  10. The 1 cent levels were machines, and very few stocks traded at the price.
     
    #40     Jun 29, 2013