Everyone is telling Iron something different. Sh-t, no wonder people turn to indicators. Obviously PA isnt as simple as people say it is. hmm.
I did that for many weeks on a demo account. I had > 90% profitable days. I thought I had found the holy grail. I even posted some P/L statements with many +$800 days. And then in one day I blew the account. The only way I could make that profitable was by martingaling into positions. Everybody's buying (price going up)... sell! Around 50% of the time I would make a profit instantly. around 30% of the time people would keep buying and I would keep selling on the way up. You have to double the open position each time, tho, or "breakeven" gets to be too far away from your average cost basis. Around 5% of the time I ended up entering 5 or 6 times, which becomes 31 open contracts, which is way, way, way more than I would ever do in real life (seeing as how I trade 1 contract now). But I would still win most of the time due to statistical mean reversion. But on a black swan day, I blew the (demo) account. The problem with martingaling is you can't change your mind if the position goes against you, because you'll be in the hole too much money already. And it only takes a small move in your favor to be profitable. So really, there are only two kinds of trades with mean reversion martingaling: Profitable trades, and account-blowing trades. There are never losers (ok, occasionally there are, when price goes up and you're selling selling selling and then price goes horizontal... then you just accept your whatever loss and get out, but that doesn't happen too much, which is why I had > 90% winning days). You will make money on every trade (provided you follow your rules) until the one time when you blow your account. And since martingaling only works until it doesn't, you get a million small wins and one catastrophic loser, I am trying my best to avoid such techniques now I think it's time for the trend to be my friend.
Buying when people are selling and selling when people are buying does not also mean you have to margingale. It does mean you have to be patient and pick your spots. When you do what Redneck suggests you can't be swinging for the fence each time. Play for shorter, obtainable gains. You should know the flip side of trend trading. This was taken from a Curtis Faith post some time back: "Perhaps most important of all, trend following using futures takes a much larger account to trade than most people who want to start trading generally have, say $50,000 at the very minimum and more like $100,000 to $200,000 if you want decent diversification. It also requires one to endure drawdowns that approach the level of the returns, so if you want to make 40% returns you need to be prepared for a 40% to 50% drawdown. Most people cannot endure this therefore trend following is not a good trading style for most people."
To me, candles are more clear than OHLC bars. The only time I use OHLC bars is when I am using a study called "candle color based on slope" (of an indicator), because it paints the entire candle/bar that color, and then with candles, i can't see if it's an "up" candle or a "down" candle, but with OHLC bars I can still see whether the first tick was higher or lower than the last tick. I can see patterns better in candles. Probably cuz I read Steve Nison's candlestick course book before I ever knew what OHLC bars were. And to be honest, if you look in the Day Trading 2.0 thread by jjrvat, I used OHLC bars exclusively in there for months (because I was using the color based on slope study). It's a bit refreshing to me to be using candles again in this thread. But I will try OHLC bars again, too. It can't hurt
The only way I found to do it without martingaling was by picking tops, and I can't do that very well at all. What properties of the futures market cause that to be the case for trend trading? I had 40-50% drawdowns on a daily basis when I was counter trend trading in my demo account.
Tonight's question is in this image: (edit - sorry for the typo... "shoring" should be "shorting") So I am beginning to try to quantify this PA stuff. The rule I was testing here was: - go short in a down trend after a retracement when a red candle breaks the low of the previous bar Based on the attached image, I believe that rule should be scratched as useless. The next logical modification of that rule will be: - go short in a down trend after a retracement when a red candle breaks the previous low (swing low). But I've already heard people in this thread saying to do the opposite of that.
lol, Iron, like I said before, that first line you drew on that red candle, you shouldn't take that trade because the market is widening..it made HH's and LL's than the candle before it. on the 2nd candle, yes that was a valid short entry, so that was just unlucky. the 3rd line you have is a perfect short, price DIDNT make higher highs, so thats a good short, and your stop would be at the highs of that green bar. pullbacks usually test their last areas of support and resistance.
It wasn't higher when I took it. I'm entering trades in the middle of candles now. Based on what you said, now I would not have taken that trade if it had printed above the previous candle at the time.