Hi all, I was wondering if there are traders out there who trade based on eod systems, but place trades intraday? if so, do you have rules of thumb you utilize for when to take an intraday trade, and when to wait? In particular - I'm specifically wondering about the sell. Ex: If I check prices in the afternoon, and I'm pretty sure my eod system is going to tell me to sell at close (the price of security is above my sell trigger), is it best to sell intraday, or is it best to wait until the close? Is there a rule of thumb? I can't backtest this with my data, I can't really glean any pattern to the trades I've been doing, and I'm asking for perspective on what tends to have the edge. I'm trading a short term mean reversion system. So, do you wait for the close to sell to be as near the system trade result as possible, or take an intraday trade that exceeds your sell parameter? thanks for perspective.
If its mean reversion, I'm guessing your system (for example, with a long trade) is a low close relative to some recent metric or "range down" day or something like that? If so, u can do some stuff simulating various intraday rules with an approximation (buy with limit order if X occurs), but the only real way or best way is to use intraday data. I don't think the rules of thumb will be all that helpful.
Perhaps try to do a Google search about EOD trading. I would think EOD trading is irrelevant to intraday. However it doesn't men you can't do anything which is innovative in trading systems/methods.
Trying to improve a good EOD system by involving yourself intraday is simply going to negate the statistical probabilities of the system and cause you frustration. Set your price targets and let it run. That's been my experience.
Thanks for the feedback from everyone. Appreciated. Yes, it's long only presently: it has several entry conditions - some dependent on market state - but it buys the dip in individual stocks below short term MA, and then sells on a close when it moves back over the short term MA. It's inspired by the first model in Connors/Alvarez/Connors Research book "High Probability ETF trading". The specific reason I asked was I've been thinking about my behavioral bias in, and specific case today was a trade w/ PRGO. It was one of the buys from Monday and I sold to close out position around 3:00est today at 190.25 when I checked prices, and I would've been better to wait as it closed at 191.25 at 4:00pm. I realize behavioral bias can come into play here, but I also realize I'm probably not going to wait until 3:59pm each day to place all my sell orders either. I'm manually entering trades so I have to start my actions a "while" before trading day closes to wrap things up. In the past I've seen fades on some of these reversion type trades at the end of days, so sometimes it's good to take the mid-day spike, and sometimes it trends on upwards like PRGO did today. As I track it over several # trades I can't tell much difference. Everybody's likely right that I just won't know unless I get some intraday data and test it... Again - I appreciate the feedback. I'm still learning swing trading so the perspectives help. (I come from a longer term investment framework for most of my life).
I don't find it to be behavioral bias. The fact is one event in one day gives you very little information to make a judgement with regards to the metrics of your system or the right way to implement it. Just write the events of today off as randomness all round. Next, get to work, because what you have discovered is that you need more precise entry/exit logic with regards to how your govern this decision process of entry and exit. That is the lesson. What i have found is that so long as the basic idea is right and your emotions are not guiding you in to doing something stupid, you should be OK structuring some simple governing rules even if they don't exactly match the parameters of how you evaluated your system. This assumes you are doing something conceptually sound, which it sounds like you are. There is nothing "special" about the closing price, its the concept (rebound from oversold over a time horizon) that counts.
I just think profit performance should not be too much a concern for EOD trading. It's more a lifestyle issue, easy and relax work, still a brain challenge game. The freedom and flexibility of time availability to a trader is also important. Just my thoughts!
" http://www.cmegroup.com/trading/equity-index/fairvaluefaq.html#q6 Q6 How will Market-on-Close (MOC) orders be handled on days when end of month settlement procedures are applied? A6 Market-on-Close (MOC) orders shall be handled in the typical fashion on days when end of month settlement procedures are applied. In other words, an MOC order becomes a market order to be filled at prevailing prices during the last thirty seconds of the trading session, i.e., between 3:14:30 and 3:15:00. The filled price may or may not resemble the settlement price. "
ES Specific : 08:30-15:15 http://www.elitetrader.com/et/index.php?threads/moc-futures-orders-at-ib.189564/ Q http://www.cmegroup.com/trading_hours/equities-hours.html Open Outcry CME ClearPort US Index Futures and Options (CME) E-mini S&P 500 Futures & Options 16:00 17:00-16:15 16:45, 15:15 17:00-15:15, 15:30-16:15 S&P 500 Futures & Options 16:00 17:00-08:15 16:45, 15:00 17:00-08:15, 15:30-16:15 08:30-15:15 UQ