Tuesday, April 28, 2009
TLT breakdown
The most interesting thing that happened today was TLT broke below 100. Either we've already entered slow summer trading mode or everybody is waiting on the FOMC annoucement tomorrow, but it's been like watching paint dry.
Saturday, April 18, 2009
Discretionary vs. Automated Trading
I've been neglecting this blog lately because I have started the transition to automating my trading. My long term goal has always been to develop automated trading strategies. First I needed to become familiar with the markets so I could discover firsthand what works and doesn't work. I do believe that the best automated trading will underperform the best discretionary trader in terms of win/loss ratio and expectancy. However, there is expectancy and opportunity as Van Tharp explains. If I am only able to find one or two excellent trades per day manually, and an automated system finds 10 decent trades per day, the results can be equivalent in terms of P&L. For example, 2 discretionary trades with an expectancy of 2.0 will result in 4R profit. Ten automated trades with an expectancy of .4 will also result in 4R profit. The one drawback is that the 10 trades incur a greater transaction cost in terms of commissions and slippage.
The transition to automated trading won't occur overnight. This is going to be a lengthy process that begins with translating all of my favorite setups to code so an alert will be triggered every time something sets up. I think this alone should improve my trading because the criteria will have to be clearly defined (should reduce some impulsive trades that don't completley fit my criteria) and the universe of stocks that a scanner can monitor will be much larger than my small watchlist and unusual volume scans. I have been coding in Metatrader MQL4 and will eventually port over to Tradestation's Easy Language.
The transition to automated trading won't occur overnight. This is going to be a lengthy process that begins with translating all of my favorite setups to code so an alert will be triggered every time something sets up. I think this alone should improve my trading because the criteria will have to be clearly defined (should reduce some impulsive trades that don't completley fit my criteria) and the universe of stocks that a scanner can monitor will be much larger than my small watchlist and unusual volume scans. I have been coding in Metatrader MQL4 and will eventually port over to Tradestation's Easy Language.
Subscribe to:
Posts (Atom)