Is CBOT changing order fill preference?

Discussion in 'Financial Futures' started by cornholetrading, Nov 29, 2005.

  1. mcurto

    mcurto

    Members have no advantage over others in terms of what they see on the screen. That is why depending on how much size you trade it may not make even make sense to be a member anymore. I could understand your worries, but can assure you that members are not able to see anything different than other traders. The only advantage they may have is being in the pit and simultaneously trading on the screen. In that sense they are leaning on something different and may know where paper and stops are based on their read of order flow on the floor (which several 10-yr locals do effectively at this point in time). In my opinion you can't beat FIFO. The 1 and 2 lot Autospreaders will eventually cannibalize themselves as they all chase the same spread and larger (more well-capitalized and risk seeking) locals attempt to leg them on their spreads even though they are first in the que. Don't worry about the policing of the screen, those who cheat and steal are always caught, no matter what venue.
     
    #21     Dec 3, 2005
  2. First of all, there are very few successful electronic exchange platforms, and interestingly most of those exchanges are public (CME, ISE, DB), therefore there is no reason or otherwise for the exchanges to provide a "backdoor" for the specialists. The public fallout of such existence would destroy an exchange, since all an exchange as going for it is a clearly established venue for trading.

    Secondly, even the pro-rata algorithms in use today are limited in scope. I can only speak for ED and my knowledge of the CBOT agri futures, the pro-rata only works up to a volume cap after the first order the first order. For instance, ED is 150 after the first order, CBOT treasuries is 100 contracts. So a specialist, doesn't matter how hard they try, can not get their 5000 order that was just entered on the electronic platform filled ahead of the '000s of orders in front.

    Rufus
     
    #22     Dec 3, 2005
  3. Thank you mcurto and rufus for your commentary.

    BTW, I used "specialist" with quotation marks around it and meant it in an allegorical way pointing to a possibly disappearing species at some stock exchanges.

    I further agree with the general preferability of a FIFO algorithm without exceptions. Although I am kind of removed from the action in Chicago, I can imagine all kind of scenarios with other queue allocation methods. What about fast markets, or even worse, panics. Block size dependent allocation would give large players a chance to bail out passing over the better placed traders in the queues who might be saddled with a likely loss.

    Although I don't want to suggest that this would be applicable to futures queues, a well known strategy in managing allocation to production queues in manufacturing shops is to give high priority to the smaller orders. It can be shown that such a strategy optimizes flow in such environments - just the opposite of the large block allocation prriority!

    It would be interesting to have an unambiguous explanation as to what would be the "superior" operational performance in exchanges resulting from schemes other than FIFO.

    nononsense
     
    #23     Dec 3, 2005
  4. Note to self, don't write posts while eating breakfast ... I meant to write "agricultures" ... heh
     
    #24     Dec 3, 2005
  5. IMHO there is no huge advantage with FIFO anymore. There was a time when one could post a bid/offer at the front of the queue in the ZN or ZF and get filled without any risk of losing. If you were in the front and got filled, you could either take profit or have enough time to scratch. Now, sometimes you are lucky if you don't lose a tick trying to cover. Basically every price bids and offers so these trades dont work anymore. I suppose it would be harder to cover loses if it moves to pro-rata and thousand lots pile in ahead of your hundred lot.
     
    #25     Dec 5, 2005
  6. hi all. thanx for posting this stuff. i trade gold through the ecbot and the website says gold uses the pro-rata system.
    so,my question is:if i put a bid in of 10 contracts and someone comes after me and bids 11 contracts,does that mean he gets first crack at 11? or just the first 1 contract? any help on this would be appreciated. with the bonds it doesnt affect me much, but with the gold contract i'm getting frustrated. thanx:)
     
    #26     Dec 9, 2005
  7. chisel

    chisel

    The COBT is changing the 2 yr. T-note algorithm. I hope the 5s 10s and 30s stay FIFO since I don't trade much size.
    ______________________________

    CBOT® Pilot Program - New Trading Algorithm in 2 Year Treasury futures

    The CBOT Board of Directors has approved on a pilot basis a modification to the current algorithm in outright Two-Year U.S. Treasury Note futures to Priority Pro-Rata with minimum volume and maximum volume cap requirements from the current Price/Time algorithm. The decision to implement this change was based on feedback from various Members and Member Firms that included a broad-based representation of market participants. This change will be implemented effective January 2, 2006 (trade date January 3).

    Effective December 15, 2005, Members will be able to test their applications utilizing the Priority Pro-Rata with minimum volume and maximum volume cap requirements algorithm in the Simulation Environment.

    Priority Pro-Rata with Minimum Volume and Maximum Volume Cap Requirements

    The Priority Pro-Rata algorithm gives a priority flag to an order that betters the market and meets the minimum volume threshold. The maximum volume cap means that the order with the priority flag assigned will only take all incoming volume orders up to the maximum volume. When the volume cap is reached, all remaining volume is divided among this order and any other orders at the same price by the pro-rata algorithm. For Two-Year U.S. Treasury Note Futures, the minimum volume requirement will be 1 contract and the maximum volume cap will be 1,500 contracts, which is consistent with the minimum/maximum values for the Reduced Tick Spreads in Two-Year Treasury Note Futures.

    Simulation Environment

    The Simulation Environment is available during e-cbot productions hours. For specific information about connectivity and access, please see e-cbot Bulletin # 47 of 2005, Change to e-cbot Release 9.0 Simulation Environment at:
    e-cbot Bulletin #47 of 2005


    http://www.cbot.com/cbot/docs/66077.pdf
     
    #27     Dec 19, 2005