Unique swing trading signals and stock market commentary.
Friday, August 12, 2011
Hold
3:45pm: Hold signal keeps the system out of the market. One thing to be cautious about. VIX down, VXX up is usually a bad sign in the short term (1-3 days).
Ok, for some reason I am wasting my time this Saturday morning looking and thinking about JT’s model and the results over the last few weeks. Sheesh.
Conclusion - JT, get back up, dust yourself off, lick your wounds and get going again.
Yes it was a huge hit but the model has been very profitable going into this last trade and no reason it should not have similar success with further refinement (stop losses) as mentioned earlier.
I went back and looked at the model results with an emphasis on the recent results and trade signals. Of course I had the benefit of hindsight to help me but that is also what we do when we backtest any model. To date the model results have been spectacular with only 2 negative months (-6.1% in Dec and -3.8% in Apr). The loss in December was based upon an older model and the model has been refined twice since then (as per post on Jun 23). So the April loss is probably more of a “normal” loss month.
So this led me to trying to figure out what would be a “reasonable” stop loss. Something probably 50-100% greater than the April loss would be “reasonable.” Which would put it in the 6-8% loss range. Given that the model trades TNA which is a 3x ETF, this is somewhat consistent with the general rule of thumb of 2% trailing stop loss. Hey, it’s a model and something needs to be used and I am sure JT can model the optimum stop loss for his system.
Now going back to the results, the model bought on Jul 14th ($80.45) and 18th ($78.20) . The July 22nd close was $85.46 which had the model at a 7.7% profit for the current position. Then the market started tanking. By the 26th we had a 6.2% loss from the high seen on the 22nd and a 14% loss by the 27th. By using the stop loss mentioned above (or even a 10% stop loss) we would have closed out the position and had still been net positive gain for the month of July.
"VIX down, VXX up is usually a bad sign in the short term"
Really? I'd be curious to know why you say that. I've found that a VIX peak is a pretty good predictor that the market will move higher in the next few days.
And the VIX put in a huge bearish shooting star weekly candle last week.
The interested reader may wish to take a peek at a short analysis of this I did in a blog post back in February (http://nightowltrader.blogspot.com/2011/02/can-vix-really-predict-market.html).
Oh and I should add that I made a mistake last week when I called for TNA to hit 50 by the end of Monday. I should have said two days later, the end of Wednesday. Esignal shows one bar per trading day in the past on daily charts (obviously) but for days in the future for which no bars exist yet, weekend days are included in the chart. So my extrapolation was off by two days. In any case, I'm reasonably certain we're going higher on Monday.
jtthomp, I'm working on getting back up, but I have a lot of dust to brush off. :) I'm working on 2 stop scenarios... a 100% stop and a 50% stop as suggested. Keeping a 50% position sounds ideal for the many times we get a pop after the flip, which provides an excellent selling opportunity. I'm still dorking with it though, but I will keep everyone updated on this front. Interestingly, we may flip back to HEALTHY as early as tomorrow, as hard as that is to believe...
Michele, thanks for the comments. As for the VIX/VXX, I remember doing a quick study a few months back. When daily VXX was up and VIX was down, there was a lower (SPX) close within the next 4 days 81% of the time. The biggest average loss occurred by day 3 with an average drop of 0.7% SPX. There was only 1 occasion where SPX was up as much as it is today and continued higher into day 2. So odds favor some weakness the next few days.
Not sure what you use for your direction calls, but the candlestick had shown a massive sell on TNA from the 25th on when it closed below the 8 day EMA. The 8 day EMA trigger line for it is now at 48.64. If it closes above it should rally for a bit until it closes below the 8 day EMA again. The candlestick were telling the story the whole time. I have felt bad for you and finally figured out how to post a comment. Also, full stoch just broke above the 20 level which means it is no longer oversold. any thoughts on this.
Ok, for some reason I am wasting my time this Saturday morning looking and thinking about JT’s model and the results over the last few weeks. Sheesh.
ReplyDeleteConclusion - JT, get back up, dust yourself off, lick your wounds and get going again.
Yes it was a huge hit but the model has been very profitable going into this last trade and no reason it should not have similar success with further refinement (stop losses) as mentioned earlier.
I went back and looked at the model results with an emphasis on the recent results and trade signals. Of course I had the benefit of hindsight to help me but that is also what we do when we backtest any model. To date the model results have been spectacular with only 2 negative months (-6.1% in Dec and -3.8% in Apr). The loss in December was based upon an older model and the model has been refined twice since then (as per post on Jun 23). So the April loss is probably more of a “normal” loss month.
So this led me to trying to figure out what would be a “reasonable” stop loss. Something probably 50-100% greater than the April loss would be “reasonable.” Which would put it in the 6-8% loss range. Given that the model trades TNA which is a 3x ETF, this is somewhat consistent with the general rule of thumb of 2% trailing stop loss. Hey, it’s a model and something needs to be used and I am sure JT can model the optimum stop loss for his system.
Now going back to the results, the model bought on Jul 14th ($80.45) and 18th ($78.20) . The July 22nd close was $85.46 which had the model at a 7.7% profit for the current position. Then the market started tanking. By the 26th we had a 6.2% loss from the high seen on the 22nd and a 14% loss by the 27th. By using the stop loss mentioned above (or even a 10% stop loss) we would have closed out the position and had still been net positive gain for the month of July.
Anyway, just food for thought. Happy trading……….
"VIX down, VXX up is usually a bad sign in the short term"
ReplyDeleteReally? I'd be curious to know why you say that. I've found that a VIX peak is a pretty good predictor that the market will move higher in the next few days.
And the VIX put in a huge bearish shooting star weekly candle last week.
The interested reader may wish to take a peek at a short analysis of this I did in a blog post back in February (http://nightowltrader.blogspot.com/2011/02/can-vix-really-predict-market.html).
Oh and I should add that I made a mistake last week when I called for TNA to hit 50 by the end of Monday. I should have said two days later, the end of Wednesday. Esignal shows one bar per trading day in the past on daily charts (obviously) but for days in the future for which no bars exist yet, weekend days are included in the chart. So my extrapolation was off by two days. In any case, I'm reasonably certain we're going higher on Monday.
ReplyDeletejtthomp, I'm working on getting back up, but I have a lot of dust to brush off. :) I'm working on 2 stop scenarios... a 100% stop and a 50% stop as suggested. Keeping a 50% position sounds ideal for the many times we get a pop after the flip, which provides an excellent selling opportunity. I'm still dorking with it though, but I will keep everyone updated on this front. Interestingly, we may flip back to HEALTHY as early as tomorrow, as hard as that is to believe...
ReplyDeleteMichele, thanks for the comments. As for the VIX/VXX, I remember doing a quick study a few months back. When daily VXX was up and VIX was down, there was a lower (SPX) close within the next 4 days 81% of the time. The biggest average loss occurred by day 3 with an average drop of 0.7% SPX. There was only 1 occasion where SPX was up as much as it is today and continued higher into day 2. So odds favor some weakness the next few days.
This comment has been removed by the author.
ReplyDeleteNot sure what you use for your direction calls, but the candlestick had shown a massive sell on TNA from the 25th on when it closed below the 8 day EMA. The 8 day EMA trigger line for it is now at 48.64. If it closes above it should rally for a bit until it closes below the 8 day EMA again. The candlestick were telling the story the whole time. I have felt bad for you and finally figured out how to post a comment. Also, full stoch just broke above the 20 level which means it is no longer oversold. any thoughts on this.
ReplyDelete