Weekly results of a wanna B algo trader - 2011

Discussion in 'Journals' started by Engine99, Jan 7, 2011.

  1. [​IMG]

    Long story short, I’ve had two very bad trading days, blowing almost 11 grand. (without doing impulsive or discretionary trades)
    The reason the last two days went wrong so badly wasn’t that the market didn't cooperate, that’s something a trader has to be prepared for, the reason is that after 10 years of trading I still “bend” the rules and it comes back to hurt me (and I know I shouldn’t be doing this in the first place).

    What you see in this screenshot are my S2s on TOS. You see that the current batch is up about 6 grand. What you also see is the capital in each/every trade (right column). While most of my trades are about 10 grand capital per trade (which results in about 1% risk of my account per trade) you see two rows with 20 grand and one with 40. The 20 grand ones are:
    NETL (up 1k9) and RDC (breakeven) but the 40k position one is down almost 3 grand. The trade is PCLN and I entered after earnings. Nothing wrong with the setup, however greed has struck once again and I went in much larger than on the other positions (also chasing the entry big time to make sure I don’t lose out), and needless to say it doesn’t work and the market is teaching me a lesson.

    While S2 is great and an awesome hedge during volatile times, it doesn’t protect from trader stupidity and greed.

    As a 10 year trader I can’t believe:
    1. I gave all my 2011 profits back and then some the last couple days
    2. I forget about the variance in any trading system
    3. I still fall for the same mistakes (less often but still) I fell when I started

    Anyway, just thought I’d share cause being cocky doesn’t pay, only making money does, and following the rules is definitely a more consistent approach than playing a big swinging cowboy. Good luck all, it’s dangerous out there.

    engine
    aka theclosetgambler.com
     
    #21     Mar 2, 2011
  2. Ouch, that sucks, I'm sorry to hear that. I've had a rough 3 days as well, although mainly because my system seems to really be suffering now. Sticking with the automated trades is something that I still have a really hard time with as well. But I've gotten my ass handed to me enough such that I'm learning pretty well not to futz with things, although I've started to dabble in it again, trying to squeeze out a few tens of dollars. I guess it's my latent gambling personality kicking in as well.

    The thing that really sticks into my head is that Mark Douglas says in order to be the casino and not the gambler, you need to take all the trades of a particular setup exactly the same. Picking and choosing between particular trades immediately turns you into the gambler and not the casino. It's gotten to the point where I've written a set of rules on a piece of paper and taped it to my wall, and every morning, I read them like a prayer, in hopes that I remember. It's weird and primitive behavior, but it kind of reinforces the behavior in me, and it helps a little.
     
    #22     Mar 2, 2011
  3. thanks for your input, it's right on the money.
    I fully agree with the statement and I've read the book, so I should have remembered ;)

    You'd think having it all automated makes things easier but in reality the hardest thing to do is to keep my hands off the keyboard and "adjust" the strategies.

    Thanks again, I definitely appreciate the advise
     
    #23     Mar 3, 2011
  4. Awesome trading week (up a grand). This has been the biggest emotional roller coaster I've had in my trading in a long time. Monday Tuesday were very bad but Wednesday and Thursday were amazing. Thursday was the biggest one day gain I've ever had. Here are my daily nbrs. for this week (-2034, -8315, 4634, 6946, -42). I'm still on a high from yesterday's gain, amazing luck that the positions turned around so well. Here are some observations. S21 has had two bad trades (out of about 20 so far) but those two trades more than account for my losses. I've made some adjustments to the strategy this week, so I'm hoping to reduce the risk a little more. I still have faith in the strategy though. S11 gained a grand, that worked out well. S10 also gained from time decay, however the increasing volatility hurt a little. S2 was a great hedge and profited from the volatility increase. S81 sucked big time. I've entered ok setups but I entered with way too much leverage and they mostly went against me. I'm trying to get out with minimal damage and I hope going forward this strategy is still as promising as the backtesting results. I got back from Cabo Tuesday night, work has kicked in fully. I'll be on the road again next week, going to San Diego. Systems will be on auto-pilot and I won't have much time checking my performance other than in the evenings.
    I hope everyone who did bad Monday/Tuesday got as lucky as I did and recovered their losses. For all the ones following this journal, have a great weekend !!!



    Here are the results in detail
    S1:-2837
    S2: 8342
    S10: 3027
    S12: 1126
    S11: 1449
    S21:-510
    S81: -2928
    Weekly Performance:
    1189
    S&P 500 Weekly Performance:
    283
    Overall Performance since launch:
    7669
    S&P 500 Performance since launch:
    10508
    Risk Free Checking Account Performance since launch:
    397
     
    #24     Mar 4, 2011
  5. Strong trading week (Up 6k3). I enjoyed a ton of seafood (and Mardi Gras) in San Diego this week. Trip was good, especially since on Tuesday I've had a +6k day. Then Wednesday I gave some money back and Thursday I was where I started for the week. Needless to say I was very frustrated. Then today the big up day finally arrived and I banked some $$$. Ever since DG's reply I've been wondering if I've alpha or beta. I'd like to think it's alpha, however trading options and seeing my daily swings that went from high hundreds to several grand a day after upgrading my accounts to portfolio margin, I'm not sure. It's too soon to celebrate, that's for sure. I did kill the S&P this week and I'm actually finally leading the race again going back to when the comparison started last June http://www.elitetrader.com/vb/showthread.php?s=&threadid=200655&perpage=6&pagenumber=1
    S11 finished very well this week, S12 didn't trade, S1 made a handful of trades, nothing exciting and S10 got hit by the S&P 500 losses for the week. Not the sweet spot for delta neutral S10 right now. The good news was S21 which finally turned a profit, S2 which kills it during higher volatility environments and S81 which came back. S81 is my newest strategy (Pairs Trading) and I'm happy that I closed all my highly leveraged positions after coming to senses and now just have a small trade on. Overall a great week and a great year so far. I won't be on the road next week so I may trade some S1 if it works out. For all the ones following this journal, I hope you did well. I saw neke's post http://www.elitetrader.com/vb/showthread.php?s=&threadid=213286&perpage=6&pagenumber=17 , he did very well this week. Finally this one sided up action is coming to an end and the fun returns.
    Have a good weekend !!

    Here are the results in detail
    S1:-2697
    S2: 10803
    S10: 1680
    S12: 1126
    S11: 1996
    S21:19
    S81: 1069
    Weekly Performance:
    6327
    S&P 500 Weekly Performance:
    -3299
    Overall Performance since launch:
    13996
    S&P 500 Performance since launch:
    7209
    Risk Free Checking Account Performance since launch:
    443
     
    #25     Mar 11, 2011
  6. Well, I just started with my taxes for last year and one thing I can say for sure is that I'll owe about 8-9 grand. So to avoid going into debt again, I'll take the 6 grand out that I've put in last month. That's about 46 SPY shares, so my performance comparison benchmark drops down to 1978 SPY shares.

    Today I've lost pretty much as much as the S&P 500, I am down about 1300 for the day, the S&P was down about 1k6, so nothing to brag about on either end. One other thing that nags quite a bit lately is this whole alpha vs. beta comparison that opened another question as to how much I'm correlated with the S&P 500. Over the last 9 month the S&P has been going up, so has my performance, however I'd like to think my performance is similar even if the S&P doesn't go up. So I'll post a correlation comparison over the next couple weeks to see if there's any good input out there.

    Hope you guys did better than me today !!
     
    #26     Mar 14, 2011
  7. Can you get a daily % gain/loss for your system vs the daily % gain/loss of the SPY and then run a correlation calculation on the two? It should be relatively straightforward using excel.
     
    #27     Mar 14, 2011
  8. Yes, that's exactly what I'm thinking.
    It won't be % but $$$ based but I'll use that to derive the comparison.
    When I did that the last time I didn't like the result ;), my problem is that I've daily data going back until June of last year but not further.

    The problem was last time (and this may still be the case) I was about 85% correlated with the S&P 500 performance since both went up over the last 9 month.
     
    #28     Mar 14, 2011
  9. [​IMG]

    Here's the correlation chart. Based on Excel's correlation function (1 being correlated, -1 not being correlated) my trading performance is 0.14 correlated with the S&P 500 (that's actually pretty good). That's on 190 data points/trading days. A couple of month ago it was almost .81 correlated but that was when the market and my trading went mostly up. This week has somewhat changed the correlation, however from my perspective it's still there. I lost less in the early part of the week but I didn't make as much as the S&P today either (and if that's an indication I'll probably also loose tomorrow's comparison cause the futures are already up by 8).

    Anyway, blue line is my account, red line is the S&P 500.
    Good luck tomorrow !!
     
    #29     Mar 17, 2011
  10. Actually 0 means not correlated at all, -1 means 100% inversely correlated, so 0.14 is actually great!

    Just out of curiosity, the chart looks like the daily cumulative PnLs, is that what you correlated, or did you correlate the daily % PnL? The daily % PnL is likely what you want to correlate because it gives the raw correlation that you're looking for, and won't have any dependencies on the previous days.
     
    #30     Mar 18, 2011