Just how liquid is the ES!?!

Discussion in 'Index Futures' started by RangeTrader, Jul 1, 2012.

  1. I was doing some analysis with numbers bars...

    On events like the Merkel spike, and others...

    Those large blocks on either side of the market are an illusion... If this data is correct... They are just algorithmically placed and when the market actually moves toward them they disappear before they get filled.

    I mean seriously... Get a few points outside the strike and price is only hitting 400-500 contracts on it's way upward through each tick??? In the last hour of trading where there "appears" to be 3k+ contracts at pretty much every single bid or ask in either direction?
     
  2. The market just skims through places where there was previously 2000-3000 contracts waiting at the bid or ask merely hitting around 250-750 contracts on it's way through the prices...

    If my data feed is correct...


    Sometimes the market even spikes through areas hitting only 50-100 contracts...


    Interesting that the liquidity that "appears" to be on either side is nonexistent.
    The real liquidity is all market orders and the limit orders being pushed in at the bid or ask right around price. The rest of the book is mostly fake...

    The bottom line of all this is... There isn't much reason to have a L2 up anymore unless it's your favorite way to enter orders. You can do the exact same thing by dragging and order up/down on a chart without a L2.
     
  3. hft_boy

    hft_boy

    Yeah, I've seen that. My guess is that those 'fake' orders are placed by market makers who are jockeying to be first in line should the price get there. Of course, if the price actually does get there, they (algorithmically) reevaluate if they actually should be in the book at that price level depending on market conditions.
     
  4. And this is a surprise???? Oh yea, a very big news event just happened, but I think I'm going to keep my 5k in assorted lots just where they are because I want to provide liquidity irrespective of the fact that I'm about to get ass raped by the market. Compared to other instruments like nasdaq and dow, ES is still the clear winner in liquidity.
     
  5. It's not that much of a surprise. It's just I never had a way to accurately measure it. I was playing with numbers bars for fun and discovered they could be used to accurately analyze true liquidity levels on each side of the market by how many contracts are hit 3+- handles up/down during spike news events.

    It appears that liquidity levels on each side are like this...

    Midday... 1000-1500 contracts appear on each tick on L2 ten ticks in each direction.

    Actual orders(Non fake):
    Within two ticks: 500-750 contracts per strike.
    Within two-four ticks: 250-500 contracts per strike
    Within four-eight ticks: 150-250 contracts per strike


    First 30 minutes/Last 30 minutes... 3000-4500 contracts appear on each tick on L2 ten ticks in each direction.

    Actual orders(Non fake):
    Within two ticks: 1500-2000 contracts per strike.
    Within two-four ticks: 750-1500 contracts per strike
    Within four-eight ticks: 500-750 contracts per strike


    Just a rough estimation of what I see... Nothing for traders around here to worry about... But anyone tossing in any big blocks (5000+ ES) at market is going to shoot right though the liquidity and get bad fills.

    The L2 deceives them into thinking their is liquidity there, but as soon as they toss their order in it vaporizes in a split second and shoots the market upward. Then the alogo's come in and profit.

    These are not human market makers. The blocks are moved around far too fast.

    Anyway, nobody in here should give a sh*t about this unless your trading 2k+ blocks of ES... So nobody com into this thread and b*tch. No reason to.


     
  6. hft_boy

    hft_boy

    Nice. So half the orders within 2 ticks are real, or for every real order there is a MM order.
     
  7. It's actually a published fact that half of futures trading is automated at this point.

    Might be 75% automated within a couple years... Who knows. :/

    That really only effects very short term traders and scalpers. I know from personal experience that the alogos move in fast... When you get the proper price you either have to have a limit order in place waiting or ten seconds to place that market order before the alogo's strike or you miss the trade.


    I'm actually recently moving more into swing trading as it's less hassle and much easier/less stress than short term trading.

    There are only around 2 good intraday swing trade setups per day on average... And the big swing setups only come around a few times a month.

     
  8. TheBlackHand

    TheBlackHand Guest

    Nice. But dont expect any Nobel prizes for this work.

    It has nothing to do with HFT BTW. Simply people provide real liquidity and trade where they perceive value to be, not the points in between. As people dont want to trade at non value prices, the market goes through them like a proverbial dose.

    What s/w package is that btw?
     
  9. When the blocks at each bid/ask are 3-4k... And the market has a news spike less than human reaction time, and the blocks at the bid/ask are gone before price hits them... It's automated.

    Sierrachart L5 Numbers Bars with a custom Point of Control Indicator.

    Nobody can click fast enough to remove ten 2.4-3.5k blocks out of the way of price within milliseconds to insure they don't get filled... Maximum human reaction time is greater than 180ms... Add in screen lag and connection delays.

    It's all automated alogos fishing around price. Everyone knows most futures trading is automated anyway... I just never saw any rock solid proof until now. There was an article a while back that described how these bots work...


    The co-located machines must detect the big blocks coming down the pipe fast enough to move them out of the way of news events... There is probably also automated analysis of news reactions too at this point. They have started writing sports news and other simple things with automated algorithms so why not analyze market news the same way?


     
    #10     Jul 2, 2012