25,000 Roundtrips a day?!

Discussion in 'Prop Firms' started by pinetboltz, Sep 12, 2016.

  1. I was interested in the cost of doing this kind of volume....
    I am estimating Mstier is doing at least 4-5k R/T's per day...so even at 10 hours per day, that comes out to 6 to 8 R/T's per MINUTE !
    So in looking at the CME's rate schedule here :http://www.cmegroup.com/company/files/CME_Fee_Schedule.pdf , assuming an AMP Futures-like commision rate of 0.15 and then the CME fees of .35, that's 50 cents total per side or $1 per R/T per contract.
    So Mstier, is your daily commision cost about $10-30k per DAY ?
    Or is the commision fixed for the month and not per contract ?
     
    #121     Oct 7, 2016
  2. tommo

    tommo

    Exactly!! Spoofing hurts HFT momentum algos that jump in front of peoples orders when theres a directional move.
    Iceberg orders are encouraged by the exchange and they are just as deceiving as spoof orders (icebergs showing less size than you really want, spoofs showing more size than you really want). One of them helps HFT's the other one makes life harder for them. No surprise which way the exchanges decide to enforce.

    A world where spoofing is allowed is a world where HFT's are squeezed and have to reduce their activities
     
    #122     Oct 7, 2016
    i960 likes this.
  3. mstier

    mstier

    That's pretty funny actually. I generate plenty of stats and have more tracking tools than I can reasonably justify spending time using. How would knowing my # of round trips improve my trading, given that I already know how many contracts I trade daily, my $/contract/product/strategy, descriptive stats for each running strategy, etc? Nobody in the industry asks about your round trips / day.

    Example of a useful tool at the firm- A script that will take all your trades in a given product during a date range, and then run each of them against historical tick data to show you edge after given amounts of time. Each line will show execution time/price, and then WMM relative to that execution after 100ms, 1s, 10s,30s,60s,300s, etc. A bit of a crude way to track alpha decay.

    My general feeling is that "tracking" is useful to the point where it's helping you make trading decisions - where to size up, size down, adjust aggressiveness of params, focus research efforts...and that's about it. Spending inordinate amounts of time in excel creating some chart-porn is time that could've been spent elsewhere.
     
    #123     Oct 7, 2016
    propwarrior and algonoise like this.
  4. mstier

    mstier

    I am fortunate to work at a firm that is colo'd at exchanges and pays member rates at CME. The short answer is yes of course, many thousands a day are paid in commissions. Just under $9k so far today.

    If I found myself paying retail rates my style of trading would of course change radically and volume would plummet. I'll of course also feel the fee increases at CME in January and will adjust as necessary.

    I posted thinking it would be fun to share typical volume numbers from the perspective of paying member rates and trading at scale. The lower $/trade, push volume motif.
     
    #124     Oct 7, 2016
    algonoise likes this.
  5. So Mstier, if you're average net win per contract is just a quarter of a tick, you are grossing (before fees) about $40k-60k per day ?
     
    #125     Oct 7, 2016
  6. you monumental muppet. You are using retail round trip costs for a trading firm lol. They will have membership or clear via an entity that has membership at all the major exchanges. On top of this they will likely be qualifying for the maximum published discount volume tiers. You then have further discounts which are not published are invite only. I would also hazard a guess that Mstier is likely also providing liquidty on some of the challenger exchanges, these exchanges will typically pay a stipend/rebate.

    Yes there are trading firms who run strategies where they are grossing 10%-20% of a tick and making good money. Retail pays $3 to $4 to per round trip on ES, the largest players are 10c to 20c all in.

    It's rare anyone from a firm posts here. It's not hard to see why.
     
    #126     Oct 7, 2016
    tommo, helgen_1 and mstier like this.
  7. mstier

    mstier

    I would love to average (net) quarter of a tick per contract. You are using net and gross in same phrase - net quarter of a tick would make me a happy man. I could do it on lower volumes, but that doesn't seem to be optimal point on total revenue curve.

    In any case, higher volumes are definitely possible given the right setup and many member firms wouldn't consider the volume numbers I mentioned particularly impressive. Again, if I were to find myself in the retail space the profile would be totally different. What I mean to say is that given the right setup you don't need to be anyone special to push through a decent contract count. Our most conservative trader has pushed through 3008 contracts so far today.
     
    Last edited: Oct 7, 2016
    #127     Oct 7, 2016
  8. just21

    just21

    #128     Oct 7, 2016
  9. Every other business has a retail/wholesale difference of between 2 to 4:1.
    In this case, it's 20 to 30:1 ?
    No wonder Wallstreet is in such disdain.
    Where's the CFTC on this ? They should impose exchange restrictions to cap the advantage for the big boys at 5:1 of the average retail R/T. So, if that's $2 then 40c, if $3, then 60c.
     
    #129     Oct 7, 2016
  10. Sadly it doesn't work like that. The fed and exchanges are owned or controlled by the large US banks. The politicians in charge for decades are funded by the banks. They have it sown up. It's not a level playing field so if you want to compete you have to choose your battles carefully. For most retail this means they should trade longer time frames. The industry promotes day trading by allowing crazy low day only margins, this encourages tight stops and small accounts. It's easy money for them.
     
    #130     Oct 8, 2016
    i960 likes this.