poker player tries his hand at futures trading

Discussion in 'Journals' started by fletch2, Dec 6, 2005.

  1. fletch2


    Hi folks,

    I figured I'd start one of these journals from the very beginning of my trading activity.

    I am an engineer by training but consider my poker experience at least as relevant. I have a gambler's mindset, and I don't mean that in any kind of derogatory way. I like to know the odds and I play to win. Maybe I will learn this is the wrong way to think about trading, but this is where I'm coming from at the outset.

    I have been researching system ideas, but made my first few trades this week on discretionary basis, just to get my feet wet. I'm up a couple of hundred dollars, but it is kind of irrelevant, I had no formal strategy and was mostly guessing. The point was simply to get a few trades under my belt to dispel that "first trade" anxiety.

    I plan on trading ER2 using some system ideas I have been playing with. My tradestation account should be open any day now, at which point I will re-implement my ideas in easylanguge and take it from there.

    Ok, that's enough for a first post. See ya around.
  2. fletch2


    I got Tradestation 8.1 a couple of days ago, expecting it to make my backtesting life a lot easier.

    Well, it kind of sucks. It's slow, the UI is pretty awful, and EasyLanguage is pretty rigid and not very expressive or general.

    I'm doing some simple testing of back data right now, and it's taking 10-15 minutes to process a year's 5min ticks.

    I coded up the same thing in C++ and it took 15 secs to process the same amount of data. Very disappointing.

    This would be more acceptable if the UI gave me all sorts of power to process results, but it doesn't. I go to the trade list, ok "that trade looks interesting, how do I go there in the chart?" Well, you can't, other than going blind manually scrolling to it. You should be able to double click on the trade and have it jump you to that place in the chart. That seems so obvious and basic to me, that it's clear that these guys aren't actually using their own software.

    Overall I give it a C so far.
  3. Hey fletch.. just wanted to wish you good luck in your futures trading. Look forward reading about your experiences. I have had some mild success in futures lately, but certainly not confident I can make a living at it right now.

    But just started playing poker online at party poker. I can truly see how the 2 are related (emotional discipline for instance).. although I have much to learn about poker ;)
  4. Wulfrede


    If you can code in C++, forget TradeStation and enjoy the edge you have on all other non-programmer traders :)

    Plus, you can code for free.

    May you always have pocket aces!

  5. fletch2


    Ok, I've been a system trader for two days now. The results (as if it matters):

    Day 1: 1 short trade, -$70
    Day 2: 1 short trade, +$230

    This is trading my best strategy only so far, which trades at most once a day. I have one more (which also trades at most once a day) which needs a little more work before launching. I have a few more ideas I've researched which need a lot of work or may not be fundamentally viable. I still have a good half dozen or so ideas I haven't had time to research yet.

    I am starting with $10k capital. This is going to be at least several months before any of this gets particularly interesting in the sense of statistically significant. Feeling good so far. I am finding the system research quite fun and interesting. Definitely in the "honeymoon period" still with this whole idea. We'll see how long that lasts!

  6. Hamlet


    Tradestation should be embarrassed that this product, which has been in development for over ten years and is on the umpteenth revision or release, is so innefficient and lacks so much. Heck, you've only been using it a few days and you can already see its shortcomings. Don't waste any time calling the company to complain or make suggestions because they just don't care.
  7. Shazbatz


    Tradestation is a terrible intraday/scalping platform. Not bad for testing and hands free swing trading though :)
  8. Fletch,

    Welcome to the world of legal gambling. Trading futures is fun and enjoyable if your system is working and your broker has no issues.

    This former Poker Champion is also a good trader, read his article on trading posted last night, it might give you an idea or two on designing your system.

    I trade ER2 and ES, I use the same idea, buy on weakness in uptrend and sell on strength on downtrend, using fib extensions/retracements, making about 6 trades a day. See the chart below, double click to expand it:

    I'll visit your journal often, please feel free to visit my journal :

    The ER2 Trading Journal:

    Good luck in your new journey!

  9. Fletch,

    From what you've posted, you seem to be approaching this in the right way. Start small, do plenty of backtesting, keep goals realistic. I think exploring systems is a good way to begin, even though many who do that end up using a semi-systematic, semi-discretionary approach.

    You might want to explore No PM's journal and updated thread for some interesting ideas on a classic approach, divergences. Harris tech's journal could also be a good source of direction for you.

    Good luck and keep us appraised of your progress.
  10. fletch2


    Thanks for the positive comments.

    I'm at Day 4 of my trading career. The record so far (as if it matters):

    ER2, 1 contract
    Day 1: short, -$70
    Day 2: short, +$230
    Day 3: short, +$240
    Day 4: long, +$260 but still open

    I've got two things on my mind at this point. The first is to start working out for myself how to reconcile multiple simultaneous strategies when they indicate trading against one another.

    The second issue is how to systematize my position sizing. At the moment I'm just trading a constant 1 contract on my initial thin margin of $10k, so there's no issue yet. But within a few months to a year, knock on wood, I hope to be in shape to start trading more than 1 contract.

    My initial "common sense" approach is this:

    Take the most recent 500 trades from the backtest, and, starting from some initial capital, start choosing them randomly, and determine a percent risk of ruin over, say, another 1000 trades, via a monte carlo approach for 1 contract, 2 contracts, etc.

    Then simply choose the # of contracts based on a chosen risk of ruin (say, 1%) for a given level of capitalization as implied by the backtest trade history. This approach has obvious imperfections but seems like a reasonable first pass.

    So, this is about a half day's programming work, I would guess.

    Adios for now.

    #10     Dec 20, 2005