Nov 232010

Brainstorming here on position sizing & stop loss:

Let’s say you have a $20k account and you are experimenting with stop size.  You want to risk 1% max per trade which is $200.  With a 3 tick stop, you could do 5 contracts.  With a 8 tick stop you could only do 2.  Now the 3 tick stop will be hit more often so let’s say in you do 1 pt/contract with a 3 tick stop and 2 pts/contract with 8 tick stop.  The 3 tick stop would make 5 contracts * 1 pt = 5 pts.  With the 8 pt stop it’d make 2 contracts * 2 pts = 4 pts.  So even though the 3 tick stop method may have a lower expectancy, it can still make more money.

Interesting isn’t it?  The 3 tick stop method would make $250/day.  At that rate another contract could be added after just 15 days and a second contract even less than 15 days later.

Now in between a 3 tick stop and 8 tick stop there are lots of possibilities.  Not only in the stop but also in expectancy.  A tighter stop may lower the expectancy.  So there are a few variables here.  Everyone has to work it out for themselves.  And that’s my current focus.

I don’t want to write too much about my trading today.  I was “off” but I made a good trade going long 1175.75 and ended up positive.  I didn’t trade well and that’s the main issue.  I took a 6 pt drawdown on a trade.  I’ve written about that before.  I decided a month ago that my trading plan would not allow such things.  And it’s happened several times lately.  I often just take the next setup and somehow figure out the best way to get out of the entire position.  I know that’s not optimal but the problem is it usually works.  “Works” as it I end up with a small profit or small loss.  And that’s why I did the math above, to show that total profit isn’t just about avoiding losses or maximizing profit per trade.  It’s about finding the best combination of variables to product the most money. I have a feeling what I’m doing now isn’t efficient.  For one I’m not trading the maximum size for my account.

It’s a bit difficult to open up & write about this publically.  Most traders would just pretend it never happened, at least publically.  I can’t count how many times I see someone tweet a trade entry and then go silent as the market goes against them.  It’s common to see someone tweet how they exited with a few points, it’s rare that I see someone tweet how they exited with a loss.  It’s just human nature to try and cover it up.

But covering it up doesn’t help us grow.  Today someone told me it looked like I was revenge trading and I should “take a breather”.  He was right.  I don’t know if it was revenge trading but it certainly was trading outside of my plan.  And when that happens one should take a break.  He did me a service and for that I thanked him.  My point is covering it up doesn’t help.  Looking out for each other definitely helps.

It’s good that there’s a holiday coming up.  I have some work to do.  October I spent most of the month working some issues out on simulator.  Now I think I have to work out this stop issue.  When I stopped trading today I took some sim trades.  Actually took one of them while I was in that big drawdown.  My ninja DOM is set for a 4 tick stop.  And guess what?  I did quite well:

6 pts on sim with 6 contracts.  That’s 1 pt/contract.  Compare that to real money:

Real money I got 2.75 pts on 13 contracts!  Talk about night & day difference.  The sim trades never went against me more than 2 ticks.

So my point is I know I can trade with tight stops.  I’ve done it in the past on simulator.  But I have trouble doing it with real money.  And I need to fix that before I increase size.  So two ways to work on this:  With real money and risk losing money, or on simulator and risk not making any real money.

To be honest I thought I had this worked out when I was trading with Ninjatrader.  I set the ATM with the stop loss and it was automatic.  With X_Trader I don’t have the OCO plugin so I don’t have an automatic stop loss.  I don’t want to make excuses because I could (and should) set the stop manually with X_Trader (or use the OCO plugin), but I haven’t.  I slacked off, time to fix it.

I really sharing all this as it’s very helpful & motivating for me.  I hope you find it as useful as I have.

  6 Responses to “Some work to do”

  1. Hello.

    Setting stop sizes: I’ve been round the houses with this over the years…never found the right answer. However IMO stop size is a function of how the market is behaving and it’s volatility. Assigning a 3-tick stop and working out contracts , profit expectancy etc is interesting, but not sure how realistic it is? If the ES character changes to be 30,40 pt days then a 3 tick stop is going to get hammered ? Just out of interest, how to do set your stop ? it based on candle you entered from or market profile areas ? I am probbaly being a bit dense here, but if you base your stop on market areas then as you say, you can work out your position size based on 1% of your capital. Not sure how the 3-tick, 8-tick math calculations is relevant.

    As part of your trading plan, do you have a limit on the number of trades you will enter ? personally, when I incorporated this, it made things a bit easier. I knew how much max I could lose in a day, which made it psychologically easier to handle. It forced me to look more critically for the ‘best’ setups rather than sub-par ones. And it helped me to avoid overtrading, chasing losing days, and also lowered commissions. I don’t think you have many problems to be honest, so just a general comment rather than aimed at your trading specifically.

    On a more general note, you trade ES, Bund and CL (occasionally). Just out of interest, how did you arrive at these instruments ? Were other futures like YM, NQ, ZN not fitting into the way you trade?

    Thanks….keep up the good work with the blog

  2. Very interesting comment Amarjit, thank you for sharing.

    I have two entries: at a level or a specific order flow setup. My idea is that if a level holds it holds and if it doesn’t then it doesn’t. But how much leeway? This is the big question. Sometimes markets poke beyond 2 ticks and sometimes they stop to the tick, sometimes go 1 pt past. So this is something I have to experiment with. I don’t know if there is an optimal fixed amount. But one could calculate the number of contracts for various stops, say:

    3 ticks: 5 contracts
    4 ticks: 4 contracts
    6 ticks: 3 contracts

    and then pick. These are all just my ideas. I don’t have much feedback here from professional traders. I know one guy who uses a 6 pt stop and it’s rarely hit. I think that’s equivalent to what I’m doing now. I don’t have a set hard stop but when it goes against me I just look to get out on a pullback. 6 pts works well for him and sometimes I think trying to get 1 pt or 1.5 pts is just a wild goose chase.

    I don’t base anything on candles because I’m not basing any trade decision on bar types. However if I can put it a tick or two under a low or above a high I will do that. It’s not based on a candle but rather a recent high or low.

    To me stops have to be either really tight (6 ticks max) or really big. The reason is if it’s tight then you can stop out and re-enter at a better price. If I use a 3 pt stop and stop out then it’s not easy to get a better price as that’s often at the high/low. and 6 pts is rarely hit. 6 pts is perfectly valid, just results in a small position size.

    I do not currently have a limit to the number of trades. I know my limits and I’m pretty good about quitting when I’m “off”. I haven’t had a big losing day in a long time. Thank goodness I have most of that under control. Ultimately my goal is that my daily loss limit be no bigger than my average daily win. That way I can have 2 losing days and still be positive for the week. But really my win rate is very high so I haven’t had to worry about that. But I see your point. If I limit my trades I force myself to pick better setups and prevent overtrading and that is something I need to work on.

    I’ve traded ES for 3 years now. I started with ES and have a good feel for it, at least on a multiday perspective. Bund is the most similar to ES during European hours. Stoxx doesn’t rotate as much. Bund is more volatile and has some nice moves. I really love it.

    Crude & Dax are similar. Very volatile and I’ve never been successful with them long-term. I haven’t given up on them but I want to focus on ES & Bund first.

    Thanks for your compliments. I’m learning a lot so I think it’s a win/win for everyone.

  3. Haha, me like you, software engineer find it hard to be wrong, actually it is impossible Im wrong and that’s why this trade need some more “room” before it goes to my target, ye! then it goes to my entry I get out BE or +1 tick because “some lie” I will tell to myself…

    Anyway, I am not sure your hypothesis that a bigger stop-loss is a bigger win expectancy is true, I think bigger stop-loss means lousy timing. I like what you said:
    “To me stops have to be either really tight (6 ticks max) or really big…. if it’s tight then you can stop out and re-enter at a better price” – once I’ll solve the problem of “being wrong” I will practice on re-entering at better price.

    thx for sharing, I’m also interested at the bund (actually anything to trade at Eurpoe hours).

  4. Michael,

    “All I have to do is fix these issues and let the trades take care of themselves.
    And that’s why I think yesterday was a turning point. Despite these issues, I was still able to take 10 pts out of the market. I see there is potential to be making a lot more.”

    Sometimes the desire to be right or question what went wrong leads us to continually change our plan. New trade size, stop, indicator, pattern. Especially for those of us who like to twiddle with settings, indicators, etc. Most important is to be accountable, stick to the plan, and if you are profitable, ride out the losses without making changes or questioning the plan at each downturn. Analyze what works and doesn’t away from the market. And give your plan time to prove itself. Like your 10,000 hours to become an expert, do you have an idea in mind for how many trades you will take b4 adjusting the plan to give you the proof that it works and doesn’t need to be changed?

    My fear monkey sometimes looks at recent losses and says “change smthg”, or ” that loss was a mistake, something must be wrong.” He says that because he does not want to be responsible. LOL Like all that tweet chatter today regarding people with datafeed issues, people are looking for the culprit on the outside. I am in Asia with crappy servers and its just a fact there will be some moments of latency. But chasing better software, and datafeeds just add to the distraction. I’ve been reluctant to switch to Investor RT for this reason. But inevitably I like to watch the FT71 investor RT screen via Teamviewer on a mates screen, LOL.

    My biggest issue has been changing the plan with new information. It took me from periods of profitable, to not profitable, to profitable, to not profitable. That’s brought me back to Sim for a month also. To digest new information, look at what worked best and then move forward. The way for me to get back on track is to say “these are the setups and this is the plan” for the next 100-200 trades I choose. Full stop. Second most important is accountability. Journaling is very important & your blog is great for accountability. I’m seriously thinking about which blog to post a month or so of trades onto and making it part of the plan.

    PS I like the idea of changing stop loss and targets with mkt volatility, as long as its part of the plan.

  5. TheMults – In regards to your comment:

    “I am not sure your hypothesis that a bigger stop-loss is a bigger win expectancy is true”

    Say you want your target to be 2x risk. With a 3 tick stop you could have T1 at +6. With a 2 pt stop you’d want that at +4 pts. Figuring winning 2/3 of trades I came up with expectancies of roughly 1 & 2 pts.

    When I use the ninja ATM with automatic stop loss, I don’t have a problem taking stops. I don’t like it but I take them. For some reason when I went to X_Trader I got into a bad habit of not having a hard stop.

    MarkH – I think it’s important to say I’m not changing my plan to have tighter stops, tighter stops were in the plan and I deviated from that. I’ve only had one losing day this month so I look at this as fine-tuning my plan in preparation for the next step which is to increase size.

    I got into the DOM and spent weeks learning to scalp on the DOM and then I tried to integrate that into my trading. And I kind of stopping doing that. So need to bring the scalping aspect back into my plan. This time I’m going to be more descriptive in writing.

    I agree with you that one shouldn’t change too frequently. 100 trades sounds about right. When I trade smart I make around 5 for ES and 100/5 = 20 days. Could be less for a smaller change.

    I sometimes keep a more detailed private journal. I capture screen shots when I enter a trade and I write my feelings and why I did what I did. I stopped doing that about when my results were good & consistent. I started it again this week to help me work on tighter stops.

    Market volatility is very important. When the market is more volatile or I’m not sure, I trade 1/2 size.

  6. Yes losses are not easy to reveal. Given I have hidden my identity, I stopped my business, sold my property, and sat down every day for 15 months and studied, then traded with a drawdown of 80k. After that I realized that it is not the skill but the emotions that needed tremedous focus…. the hardest area. Stops are not the issue I believe. It is inner confidence using both sides of the brain, the willingness to do the work even when down and out. To me trading is a reflection of who you are. There is no one there but me. I am greatful for what I have and am learning. I can see how it is so easy to give up. I have for while now. And that is a reflection on how I have run my life.

    Rude awakenings do have its benefit.

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>