LOL... I wonder if I am draining myself more, or those who read my journal and try to help! hahaha
Yes, there won't be any more journal updates until I actually have the results of some backtesting to report.
Trading NQ via Price Action
- Thread starter k p
- Start date
LOL... I wonder if I am draining myself more, or those who read my journal and try to help! hahaha
Yes, there won't be any more journal updates until I actually have the results of some backtesting to report.
I have a scalping system based on the 1-min chart. It's complex and "to-the-tick" as you put it. It requires a sort of concentration that I can only keep up for brief periods of time. I don't care if I miss most the setups based on it. I care about 5-min setups and 1-min entries. That's my meat and potatoes.
You are struggling with a simple statistical analysis of one of the commonly traded setups known (breakout, pullback to breakout and continuation hook in the direction of the breakout). When it works, a profit target target of 2x risk is hit easily way more often than not. Don't worry about "to-the-tick" until you have stats telling you how precise you need to be. Analyze several ideas like I suggested before: Place an order at the breakout level and log the result of stop/target combos - 2/2, 2/3, 2/4, 2/5, 3/3, 3/4, 3/5, and so on. Then do the same analysis based on either a) price reaching the breakout level and trailing a buy/sell stop 1 tick outside that 1-min bar or b) price pulling back gradually at least three 1-min bars, then trailing the buy/sell stop.
You will find a "Holy Grail" if you do that. You really will. And as for where to draw the trend lines, draw them across the swing highs/lows on the 5-min chart.
You admired Geez (as did I much later after I realized why his method worked so consistently over time). Geez didn't worry about little details. He didn't worry about trend lines, fibs, bollingers, stochs, phases of the moon, or his girlfriend's menstrual cycle. He didn't worry about 5 losing trades in a row. He didn't worry about how much farther price went after he took his profit. He didn't worry about price coming a single penny from his target and then stopping him out for a loss. He was working pretty much full time while trading stocks part time from a laptop; he didn't have time to wring his hands over minutae.
All he cared about was following his simple plan when a stock that fit his criteria was exhibiting relative strength or weakness. When his alert fired off telling him this was happening, he announced to the chat room that he was watching XYZ. As soon as there was a little pullback on the 5-min chart (and I mean a little pullback) he announced where his buy or sell stop was, his stop loss and his profit target. His OCO orders took care of the rest.
All he cared about was execution. The profits took care of themselves.
I didn't realize he was actually in a chat room... but I guess I was reading this stuff years later so it wouldn't have even mattered. You had talked quite a bit about chat rooms as well that you were in on over the years. Not that I have any interest in following anyone else, I essentially do have all the tools I need to make a plan myself, but I still think it would be nice to see a trader live, how quickly he puts on a trade when he sees something... how much he agonizes perhaps in the first few minutes, how quick he is out or back in again. (I know youtube has lots of videos, but so much of this is fake I'm sure)
After the fact its of course easy to know where an entry should be and where an exit should be, but its all that stuff that happens as the bars are forming where my interest lies and what one does during these critical first few minutes. As has been suggested to me before, perhaps just putting the trade on and walking away might be good initially. I certainly couldn't do any worse than I have been doing... taking 5 point losses and 2 tick profits! LOL This part of the geez strategy appeals to me.
Anyway... I think I will actually track 2 setups initially. Hinges are just so easy to spot, and if they are in the middle of nowhere, it will take only minutes per setup to analyze as I scroll through the days. I will make sure to focus on stuff that happens within the time frame that I watch (1 hour before the open to perhaps 1 1/2 to 2 hours in).
The second will be this 5min/1min setup. As you say, I will just leave my lines as they would be right at the open. I think I've figured out that once I plot the lines on a 5 min chart, and then switch to 1 min, I will nudge the lines a little so they match up properly with the swing points now on the 1 minute. Then I will just track the first trade that happens as a result each morning. I will take a screen capture of each setup so that I can always go back and see what combos work the best, but initially I think I will just stick with something simple like 3 point loss versus 6 point gain, and see which is hit first. The entry of getting in at the BO level versus following the RET is interesting, so doing both would be prudent, but I think its more important for me to just finish 100 setups first. I would much rather finish 100 setups with only one statistic than tracking 3 different ways to enter and even a few different R:R combos. I suspect that it might only take a few dozen instances to see that one entry method is already superior over another.
At any rate... I've got all day today, and although I realize this should take days, there is no reason why by the end of the day I don't already have an update. So lets see if I can actually get through at least a dozen setups and formulate an extremely preliminary result.
So I've got 20 charts done with my 5 min trendlines as they would be right at the open. Here is a snapshot of the first 20.. just to prove I did the work! 
(took about 1.5 hours, but that included prep time, so the next 20 will be much faster)
1. Some like this first one don't allow for any RET as price doesn't even come back to the line.
2. This second one sure is messy with an exit below first, and then ultimately an exit above.
3. This third one (Sep 25) doesn't even have a DL (oppps.. that upper line should be pink). It breaks the SL, but doesn't go higher, but there is not a DL to break on the way down to get me into a short based on a BOPB.
4. This 4th one (Sep 26) is just so messy. I guess that the fact that price keeps coming below the line should keep me away from taking a long, although if my trigger for this long was price breaking above, and then coming back to the breakout level and going long at that level, this one would have filled and it would be a loss.
Of course there will be losses, the secret is I guess finding out the sweet spot of getting out for the tiniest loss without sacrificing profits by getting out too soon. It does seem like getting in at the BO level will overall lead to bigger profits and smaller losses, but I gotta study this.
When the BOPB trade works, it just works, and its easy to see where to get in. But I gotta collect the stats for at least these 20 first charts... so let me get on that.
I do have a statistics related question if anyone wants to answer. So the idea is that if I find an "edge" in these 20 charts, it may not very well stand the test of time, and a sample size of 100 will be much better... this I understand.
But is it also true that if these 20 first charts don't produce a positive edge, then there would be no point in continuing with these parameters.... correct? What I mean by this is that if I choose something like hitting 10 point profit before hitting 2 points loss just doesn't work over 20 days, then I might as well throw out collecting this stat because its unlikely that over 100 instances this would turn positive if it can't even work over 20 instances... correct?
The other thing that I think I really have to track is what price does when the BOPB isn't a BO, but rather a bounce off the line. (ie. If I have a SL slopping down and price just graces the 5 min SL or is FBO and doesn't set up a long above this line, is there a way to get short and would this work as a rejection trade. ND... if you're reading... do yo think there is any value in this?)
EDIT: Oh ya... I should mention that I started this on the first week of this new Dec contract... seemed like just as good a place as any. I do realize that the market is different in the summer... and this is where characterization would come in as per Db... but I think a couple of months of this setup should be more than illuminating enough.
(took about 1.5 hours, but that included prep time, so the next 20 will be much faster)1. Some like this first one don't allow for any RET as price doesn't even come back to the line.
2. This second one sure is messy with an exit below first, and then ultimately an exit above.
3. This third one (Sep 25) doesn't even have a DL (oppps.. that upper line should be pink). It breaks the SL, but doesn't go higher, but there is not a DL to break on the way down to get me into a short based on a BOPB.
4. This 4th one (Sep 26) is just so messy. I guess that the fact that price keeps coming below the line should keep me away from taking a long, although if my trigger for this long was price breaking above, and then coming back to the breakout level and going long at that level, this one would have filled and it would be a loss.
Of course there will be losses, the secret is I guess finding out the sweet spot of getting out for the tiniest loss without sacrificing profits by getting out too soon. It does seem like getting in at the BO level will overall lead to bigger profits and smaller losses, but I gotta study this.
When the BOPB trade works, it just works, and its easy to see where to get in. But I gotta collect the stats for at least these 20 first charts... so let me get on that.
I do have a statistics related question if anyone wants to answer. So the idea is that if I find an "edge" in these 20 charts, it may not very well stand the test of time, and a sample size of 100 will be much better... this I understand.
But is it also true that if these 20 first charts don't produce a positive edge, then there would be no point in continuing with these parameters.... correct? What I mean by this is that if I choose something like hitting 10 point profit before hitting 2 points loss just doesn't work over 20 days, then I might as well throw out collecting this stat because its unlikely that over 100 instances this would turn positive if it can't even work over 20 instances... correct?
The other thing that I think I really have to track is what price does when the BOPB isn't a BO, but rather a bounce off the line. (ie. If I have a SL slopping down and price just graces the 5 min SL or is FBO and doesn't set up a long above this line, is there a way to get short and would this work as a rejection trade. ND... if you're reading... do yo think there is any value in this?)
EDIT: Oh ya... I should mention that I started this on the first week of this new Dec contract... seemed like just as good a place as any. I do realize that the market is different in the summer... and this is where characterization would come in as per Db... but I think a couple of months of this setup should be more than illuminating enough.
Attachments
So as I'm sitting here working.. I get an email about my statement for Oct. One way to look at it is that I'm down $750 for this month. This isn't a good result, but since I don't smoke or drink or do drugs, I think I can reasonably call this my "entertainment" budget for the month and absorb it just fine, or at the very least a "hobby" expense that can very well turn profitable.
But here is the real shocker. $723 of this is commissions! Am I reading this right? In terms of how many points I gained or lost, being down $25 makes this 5 ticks. Now of course commissions are a real business expense and the loss of $750 is real, but jesus... this is crazy. You'd think IB would comp me some free stuff for all the money I'm giving them!
If anyone has any feedback about these numbers, I'd love to hear. Going forward of course, once my trading plan is ready, taking less trades is what will I'm sure be one of the results of a proper plan, and of course holding a profitable trade and not a losing trade, which is the exact opposite of what I have been doing.
But here is the real shocker. $723 of this is commissions! Am I reading this right? In terms of how many points I gained or lost, being down $25 makes this 5 ticks. Now of course commissions are a real business expense and the loss of $750 is real, but jesus... this is crazy. You'd think IB would comp me some free stuff for all the money I'm giving them!
If anyone has any feedback about these numbers, I'd love to hear. Going forward of course, once my trading plan is ready, taking less trades is what will I'm sure be one of the results of a proper plan, and of course holding a profitable trade and not a losing trade, which is the exact opposite of what I have been doing.
Attachments
As I've said, repeatedly, there is absolutely no reason for you to be trading if you don't have a trading plan. It's not a matter of "taking less trades"; you shouldn't be trading at all.
If you can't stop, then you have something much more important to worry about than collecting stats.
Oh for sure... those numbers are for October, and when I said taking less trades, this would be for after the trading plan is done. I doubt my trading plan would tell me to take as many trades a day as I was on some of these days in October to rack up so much commission charges. So the natural result of the plan will hopefully be taking less trades, holding the winners, and cutting the losers quickly.