IN THIS THREAD: IronFist learns (the elusive) PRICE ACTION

because price broke previous candle highs., look at some of the other charts i posted here and you'll see.

 

Me too, I will occasionally drop down to 3 min bars but 15 min is more or less my favorite.

 

Let's talk about exits.

I've made a few decent entries but I don't know how to know when to exit (other than obvious things like seeing a hammer while short or something).

For the sake of this being my thread, let's not include fibonacci-related things in the discussion of exits.

I have an exit method but it's based on my old indicator based method. Since I've removed all the indicators from my chart (except volume), I can no longer use it.

Is an exit just the opposite of an entry? Do you just take the opposite of an entry signal? Say you see 1, 2, 3 or whatever and go long. Does 3, 2, 1 mean it's time to exit?



I'm going to preemptively assume there are two types of exits:

1. Profit exits

2. Oops I made a mistake and am taking a loss exits


I am going to assume that type 1. is based on PA and type 2 is based on a fixed number of ticks, however that might not be right because I already just came up with a scenario in my head where that's not going to work.

Within the context of PA trading in this thread, I am not interested in fixed amount target profits because I do not believe those are profitable over time based on my backtesting, and they prevent winners from being allowed to run, which based on my backtesting is required in order for a trend-following system (such as this) to be profitable.

So the process is:

- PA tells me it's time to enter
- I enter
- PA tells me it's time to exit
- I exit

Not:

- PA tells me it's time to enter
- I enter
- I hit a target of a fixed number of ticks (this isn't PA)
- I exit

 

your exits are a little complex because they can and should be flexible. however when i first get in a trade, my exit is always the previous candles low - 1 pip.... once i have a profit on my hands i either move my stop to BE or below last candle's low to protect gains, sometimes it isnt much but better than nothing, or i might place an oco order where i think price will move too, and just do it that way..

check your pm's

 

IMHO

Think consolidation areas after a move


What is consolidation – to me

Price moving (up or down) then it stops to rest – why

1.) Taking a break before resuming in the same direction of its last move / direction

2.) Taking a break to stop and reverse its direction

3.) Ran out of steam and can’t go either way (this one is highly unlikely over an extended intraday period of time)

They give you a good place to revaluate your position and exit / stay in



I stated in an earlier post long (speed) candles are hard to trade unless you’re already in a position and it’s going with you – this is also why consolidation points are important (gives entry points with very definable risk)




The problem to get a handle on – are the speed candles occurring in both directions one right after the other – these are a pain in the butt




Aside – speed candles can be traded it just takes more experience to trade em successfully – on a consistent basis

Take Care
Redneck

 

Firstly this is an excellent thread and has provided me far more informative info than many books i have read on trading, however having just started out daytrading with a prop firm that focuses principally on stocks I would like to enquire if the rules that apply to prce action are generic across all instruments and if they are is there anything that should be considered when dealing with slower moving, lower volume equities like the ones I have been started on. Thanks again,

Tom

 

Reading a chart is reading a chart, whether it be stock, futures, etc. That said every market has it's own personality. I trade futures. The ES is different than the TF or NQ. You get the point. Plan on spending hundreds if not thousands of hours watching your stocks to see how they act.

When the market is moving fast on high velocity and then pulling back on less velocity you have the general idea on you want to buy a pullback , especially if you have seen two HH's and HL's and there is nice support under where you are buying.

 

I'm glad you found this thread helpful.

I have taken a break from posting in it because my progress stalled as I was not able to quantify price action the way I hoped to. I have a lot of questions to which I still haven't been able to get answers and it's taking me a long time to figure out things on my own. Mostly this is because every time I think I'm about to grasp a concept, I discover an exception or corollary to the "rule" I was working with which invalidates it and I have to go back and redo everything I thought I had learned. It's frustrating, so I'm taking a break and letting my subconscious work through some of this data while I pursue other projects.

 

lol learn something new everyday

 

Someone PMed me and asked me to post specifics of the problems I'm having in this thread.

So... (I'm typing this from memory as I don't have access to my notes right now, so sorry if it's a bit spacey)

I am having trouble "trend following."

In order to follow a trend I need a definition of a trend. The most commonly given definition I've seen that is based on price action is a sequence of HH and HLs or LL and LHs. This will subsequently be referred to as "definition 1."

Definition 1 eliminates trading from the following situations:

- When LL follows an HH (by definition 1 you are no longer in an uptrend and are essentially in "limbo" until a LH forms immediately following the LL)

- When HH follows an LL (by definition 1 you are no longer in a down trend and are essentially in "limbo" until a HL forms)


Definition 1 leads to the following questions:

- how do you know when to enter the trend?

- how do you know when to exit the trend?

I have been given a few answers for these, and I have played around a bit with my own formulas, but I have not been able to quantify any results into anything usable. Sometimes it ("it" being "whatever method I'm trying") works, sometimes it doesn't. Just to name a few, some of the methods I've tried have been:

(assume we're in an uptrend as defined by definition 1)

- Entering after a pullback when the current candle breaks the high of the previous candle (this happens all the time and then price continues downward)

- Enter after a pullback when the price closes above the high of the previous candle (this happens all the time and then price continues downward)

- Enter after a pullback when the price closes above a given, relatively fast MA (this happens all the time and if the MA isn't in sync with the current price action, which is impossible to set ahead of time, even using adaptive MAs and other crazy shit most people have never even heard of, it won't work)

- Enter after a pullback when the price closes above the previous HH (hey this one sounded dumb to me, too, but I had to test it because maybe it would work; well it only works in huge uptrends: ie. not reliable because you never know if it's going to be a huge uptrend ahead of time or not)

- Enter after a pullback on decreasing volume (unreliable as volume is a random distribution histogram)

- Enter after a pullback on increasing volume (unreliable as volume is a random distribution histogram)

Alright so that right there are 6 of the methods I've tried that are unreliable and therefore scrapped. To be honest I'm really having a hard time coming up with things I haven't tried.


So that deals with entries, now lets talk about exits:

Obvoiusly without exits your system is worthless. Using the above methods, I was unable to enter an not suffer moderate to severe drawdowns.

Let me stop right here and talk about that for a sec. Everyone talks about "use tight stops, limit your losses."

I'm not sure what instrument they're trading, but you cannot trade YM, NQ, or ES with "tight stops" unless you can pick the bottom. Enter long at 800, "tight stop" of 1.5 points, well the ES moves more than 1.5 points in noise constantly, so chances are it's gonna hit 798.5 before it goes in your direction. And even if it doesn't, it will probably go up to 802, then drop back down to 797, then go up to 810 or something. So people on ET will say "well duh just wait until that final shakeout down to 797 and enter there and get on board for the ride lol i'm a trading guru!!!" Well that final dip down to 797 breaks the HH HL structure as defined by definition 1 and therefore we're no longer in an uptrend and therefore we can no longer be taking long positions.

Side note here: I'm thinking the key to trading lies in not being bound by HH/HL or LL/LH constraints of trend definition.

Back to exits.

Target profits first:

With a trend following system, I cannot understand how fixed target profits can be used. Maybe my understanding is flawed, but based on months of backtesting AND forward testing in demo accounts, the only way I'm able to get trend following to be profitable (some of the time) is by letting every trade ride as long as it will go. Backtesting with fixed target profit stops of 3, 5, 8, 10, 12, and 15 (i think) was UNPROFITABLE OVER TIME. You get a higher win ratio, but overall your wins are less. This leads me to believe that with trend following systems you HAVE to shoot for home runs every time.

If fixed target profits on trend following systems were feasible, I suspect it would require computing power that I don't posses to determine the optimal locations. But even hypothetically I can't understand that concept because you don't know how far at rend is going to go and therefore putting an artificial limit on it is dumb. Plus remember you need home runs to eliminate your million small losses from using your "tight stops."

Now stop losses:

Similarly, in backtesting just for fun I wouldn't exit a position until the market told me to, which means dealing with INSANE drawdowns at times, which frankly I don't have the balls or the account to handle. Remember that "tight stops" will get stopped out unless you pick tops or bottoms, so holding positions through draw downs is necessary to avoid having a million small losses.

In other words, I was getting entry signals, and THEN hitting HUGE drawdown, which means that my entry signals were bad. But by the time price had drawn down to the bottom of the pull back, HH/HL structure had been violated again supporting my notion that HH/HL structure is perhaps a crutch to successful trading.


Alright enough trend following action for a while. Let's talk about COUNTERTREND.

Countertrend is a whole separate beast. Unlike with trendfollowing where you're riding price into the unknown, counter trend trading EASILY lends itself to predetermined target profits. In fact it's a bit of the opposite because the only thing that isn't immediately known with CT trading is where to put your stop loss.

I will add that while I have been insanely profitable (in demo accounts) with CT trading, it quickly becomes martingaling (or some variant of martingaling) if you're not careful, and that's probably not good in the long run.


But I'm not sure how I feel mentally about this.

Psychologically, I would like to be a trend trader. I would like to have this thought: "I am entering a trade now, I don't know how far it will go, but I know it will probably be profitable". That would be low stress, low risk, high gain. In fact, that would be traders paradise for me. But 3 months after creating this thread, which was made many months AFTER following jjrvat's "Day Trading 2.0 for small traders" thread and mechanically backtesting the hell out of everything in that thread (a brilliant thread, btw, except for all the spam in it) I still cannot apply trend following in real time. And I have no idea how jjrvat manages to scalp using that method because like I said before, backtesting with 3, 5, 8, 10, and 12 tick target profits yielded a huge negative equity curve. I sort of got that method to be profitable but it required holding for home runs. So jjrvat must have some brilliant stop loss method in place or be picking bottoms like a champ.

Now, regarding CT trading, mentally I don't like small trades (even tho it's nice to see a bunch of small winners add up, kind of the way I have a ton of small losers that add up when I try trend trading). It seems insanely stressful. But it's the only thing I can make work consistently in demo accounts.

I don't want to talk too much about CT trading because it's not really the subject of this thread, so I'll stop here.

I will say that I suspect that if I am ever to become a trader, it will not be trend following because months down the road all I've gained is a bunch of definitions for mostly abstract concepts and no quantified method of application. I've talked to people who say they are successful trend traders, yet none of them can quantify what they do so being a trend trader must require some sort of intuition that I do not possess. Nothing against any of them, because most of them are super nice people (except the ones trying to sell me stuff through PM... really now, guys), I just must be too dense to get what they do.

If I am ever to become a trader, it will probably be using some method that is more mathematical, almost assuredly counter trend, but using some sort of hedging instrument to protect against the occasional catastrophic loss. This way there's no reliance on PA. If price goes up I win. If price goes down I win. If price chops I win (once it stops chopping and picks a direction). I've been tossing around a few ideas in my head like this and running some computations but I'm severely underfunded when it comes to implementation so right now it doesn't matter. And due to the complexity of them, they cannot be "scaled down" to facilitate a smaller account.

This is a long post, hope the internet doesn't hang when I push "submit reply."