Pyramiding Trades Intraday

Dr. Toad

Active member
Messages
169
Likes
33
Recently @Jason101 posted in a thread that a good way to maximize profits is to continually build a position during an uptrend over several days until the size is quite large relative to your account size.

http://www.trade2win.com/boards/general-trading-chat/224368-maximum-profit.html#post2922176

I have never really given much thought to pyramiding trades until now since obviously getting in with maximum position size at the earliest point in the trade will maximize the profit potential assuming you get in at near the optimal point. The major flaw with this thinking though is that by doing this, you must set your stop very close to your entry early on, or you must accept very large drawdowns.

So the large entry into a trade early on makes two "common knowledge" money/trade management principles mutually exclusive: the principals of keeping your losses small (can't do that if you give your trade room to breath some) and giving your trade some breathing room (can't do that if you keep your losses small).

Up until now I have just accepted this and tried to find good entry points that would lead to overall higher % winners than losers. I have for the most part failed to find any way to discern with better than a 50/50 chance a good entry point vs a bad entry point. So I started thinking...maybe pyramiding similar to the way @Jason101 suggested can be an answer to this problem.

The basic idea being: start with a small position --> if the trade moves against you, enter once more at a better price --> if trade continues moving against you, exit for small loss. If however the trade moves in your direction, add to it on pullbacks or add to it when price has increased to a point that you have a reasonable profit and can take adverse movement with a larger sized trade.

Certainly not a revolutionary idea, but just something I haven't really given much thought to or investigated much.


I had a bit of time today, so tried my hand at it. I try to avoid discretionary trading like the plague, but I do it on occasion to try new things and see how I react to the new situations so that I can develop systems around them. About half my initial entries resulted in scratch trades or slight losses. The other half however led to excellent profit opportunities, as well as a handful of question.

The main one being --> when do you stop pyramiding? The obvious answer is to stop and exit when the trend breaks. But I am curious if anyone else does this style of trading, how do you decide when to stop? Every adverse movement represents a potential entry point. This trade illustrates a clear failure to capitalize on my large position size due to failing to exit when the trend broke, but it did make me realize the absurdly lucrative potential of doing this. Although my last entry wasn't necessarily bad (probably a bit to large really, but I wouldn't say "bad"), I definitely should have exited the trade faster.

Pyramiding UAA 20170731.png

UAA 20170731 1 min.png

UAA 20170731 5 min.png

In hindsight the position should have been closed completely at ~ $20.68 since the trend on the 5 minute chart was broken. I waited to long on my exits and closed at ~$20.59 and ~20.49. Still profitable overall, just a small fraction of the potential. But I was able to increase my position size to a fairly absurd (for me) number of shares of 1,500 with very little initial trade risk.

The other two that I took that worked fairly well:

Pyramiding PAH 20170731.png

PAH 20170731 5 min.png

Reason for closing --> I have learned to be extra vigilant as whole numbers approach. It seems a disproportionately large number of trends reverse or level off at these points. I don't have data backing this up, just an observation from the limited amount of discretionary trading I have done. It worked out for this trade.

Pyramiding SCG 20170731.png

SCG 20170731 1 min.png

SCG 20170731 5 min.png

Started scaling this one back as the whole number approached. I re-entered once it looked like it would trade through the whole number. I don't see any problems with this trade, just probably should have held at the whole number with a tighter stop rather than exiting and re-entering.


So, I am thinking it is time for me to develop an intraday scaling strategy. If I can have these results on a half-assed first attempt at it, surely there is something to it. Does anyone trade in a similar way, and do you have any suggestions? How do you determine when and how much to scale in? Do you scale out or close it all at once? Any suggestions on good times to stop scaling and/or start closing these trades are welcome. Or any other thoughts.
 
Scale in is dumb. It turns perfectly good winners into losers. brewski1984 have some fresh samples. This is the reason the merchants tells you it's a good idea. It allow them to make smaller reverses to knock out positions.

"Authoritative" public info in majority parts are misinformation, do the opposite of everything they say and you are on the path to win.

Scale out if you don't know if the price is coming or going. It's no worst or better than any other random exit strategy.
 
Looks like brewski1984 has just added a new sample data.

In any case, it's perfectly ok to thrash around trying this and trying that. Once you have tried them all and none works, what you have remaining is what works. Elementary, my dear toadson.
 
Nice idea Dr. Toad.

I rarely day-trade but by coincidence have just started working a plan for pyramiding multi-day trend-following positions. Involves opening another same size trade when unrealised profit reaches twice the initial risk, +2r. Move stop for original trade to entry +1r, set stop for new trade to same price, so that worst possible outcome is now break-even. Ideal outcome would be for price to go to +3r, when I could repeat the process, always assuming the trend's TA is still good. If trend fails at +3r, I exit with +4r realised gain, for original risk r.
 
So, I am thinking it is time for me to develop an intraday scaling strategy. If I can have these results on a half-assed first attempt at it, surely there is something to it. Does anyone trade in a similar way, and do you have any suggestions? How do you determine when and how much to scale in? Do you scale out or close it all at once? Any suggestions on good times to stop scaling and/or start closing these trades are welcome. Or any other thoughts.

The following is from my book and may be helpful. It's been saved from one format to another, but from having just skimmed it, it looks okay.

Note that this is just trading price and volume so it may be of no use to you. But one never knows.

As for all the arguments about scaling in vs not, remember that professionals are continually lightening up their positions and "buying the dips", i.e., scaling back in on better prices. This is how one builds a position.

Note that this is a pdf, so it should be all right.
 

Attachments

  • Mngng Trnd Dy-Sclng Out.pdf
    731 KB · Views: 393
The following is from my book and may be helpful. It's been saved from one format to another, but from having just skimmed it, it looks okay.

Note that this is just trading price and volume so it may be of no use to you. But one never knows.

As for all the arguments about scaling in vs not, remember that professionals are continually lightening up their positions and "buying the dips", i.e., scaling back in on better prices. This is how one builds a position.

Note that this is a pdf, so it should be all right.

Appreciate the PDF and comment. I never have been able to make heads or tails of volume, I only really trade price action, so should be of some use to me.
 
Nice idea Dr. Toad.

I rarely day-trade but by coincidence have just started working a plan for pyramiding multi-day trend-following positions. Involves opening another same size trade when unrealised profit reaches twice the initial risk, +2r. Move stop for original trade to entry +1r, set stop for new trade to same price, so that worst possible outcome is now break-even. Ideal outcome would be for price to go to +3r, when I could repeat the process, always assuming the trend's TA is still good. If trend fails at +3r, I exit with +4r realised gain, for original risk r.

Good thoughts. This is kind of the route I am headed with this. For my original seat of the pants experiment I didn't really have anything ironed out, so no real set risk when I did that (very bad idea, I know).

Looking back at the successful trades, the low of the previous 5 min bar seems to be a good initial point to set as the first stop. Probably going to use that as a starting point. Looks to be a reasonable way to track the trades as well.

This is all very odd for me since the only reasonable way to trade this way (I think) is with even lot sizes. So I will have to find the trade risk and then filter the trades out by a max trade risk I set...backwards from the way I usually do with sizing position from the stop, but if what you are doing isn't working...

My thinking right now is allow for 2 entries at each level. One when the level is hit and another if price retreats halfway to the previous level. That way the position is being built in a strong uptrend as well as on small pullbacks. This still would have built a position size of around 1000 with the UAA trade...I need to figure out when/how to scale back still...time for some investigation I think.
 
I was able to get the basic concept coded up and functioning. Haven't really debugged it yet but it appears to be doing what I am telling it to. I likely need to add more filters and definitely need to do some more testing on it to make sure it is robust enough and investigate different ways to manage the trade/scale in and out, but the skeletons of the code are in now.

Pyramiding Strategy UAA 20170731-1.png

Pyramiding Strategy UAA 20170731-2.png

Pyramiding Strategy UAA 20170731-3.png

Now THAT is one sexy trade.
 
Top