Dec 282010
 

I’ve been enjoying my time away from trading real time.  For one it’s much more relaxed, I can step away from my PC to help my daughter figure out her wii game or held the younger one get dressed.  Second, I’ve been taking a step back to re-evaluate everything and look at things from a fresh perspective.  I’ve also spent time reading blogs and researching other methods and view points.  It has all been very helpful for me.  The last time I did this was during our summer vacation (about 4 months ago) and I think it would be a good idea to take a break and step back every four months.  In the past I’ve made progress each time I’ve done this.

This time I’ve been working on a few ideas.  The first is that lately I’ve been identifying levels and just fading them, without much regards to what could happen for the day or where the market was going over the next 30+ minutes.  I’ve been trading with 2 contracts, going for +6 ticks and +12 ticks.  The problem is often I get +6 (or slightly less) and then I put my stop to entry – 6 (or whatever my first target got me) and stop out on the 2nd and end up with nothing to show for it.  The theory is the bigger winners will make up for the breakevens.

There are different ways to use scaling out (or not use it at all):

  1. Some use the first scale to pay for costs and then try to be there for the big wins.  Joe Ross advocates this method and says that he may only win 20% of his trades and most of the rest will end up breakeven, with a few being a full stop out.
  2. Some scale out gradually (FT71 comes to mind).  His scale outs are something like 12.5% each scale out.  That means he can scale out 8 times before being totally out.
  3. Some don’t scale out at all

When trading 2 contracts, the 2nd method isn’t really available.  When the first is taken off that’s 50%.  This is why FT71 has advocated trading 3 contracts minimum.  When you take the first at say 6-8 ticks, that’s 1/3 of your position.  I guess I have proved to myself that trading 2 is not optimal.  So I’ve been trading more like the 1st.  But to do that one needs some big winners to make up for the breakevens and losses and I haven’t proven to myself that I’m capable of big winners.

I think I’ve been wanting to trade like FT71 a bit too much.  I’m not going to get his 2.5 pts/contract.  I don’t even see how that’s possible for me.  I think a high expectancy like that would take years of practice and experience, which FT71 has and I do not.  So it’s a lofty goal.

So I’m thinking I have skipped a step which would just be the 3rd option of not scaling out.  I’d trade 1 contract and go for 6-8 ticks.  I think this first target should be profitable before adding on.  There are two cases:

  1. It’s not profitable.  In this case it doesn’t make any sense for the 2nd contract to bail out the unprofitable first contract.  Better to just trade the 2nd and skip the first.
  2. The first is profitable.  In this case do I need a second?  Why not just ramp up size on the first and be out of the market?  More testing would have to be done to see if an extended second target really adds profits or just reduces the profits from the first target.

The other thing is I’ve studied some statistics from other profitable traders and found that many average just 2-4 ticks.  I’ve seen many traders making $750+ per day just averaging 2-3 ticks.  I’ve seen a prop firm average just 2 ticks.  This idea was brought to my attention by Markus and I’ve been doing a lot of thinking on this point.  My November average was 2 ticks and I think that’s not so bad.

In the sake of completeness, there are also big traders averaging over 2 pts/contract.  But I don’t know what they’re doing.  They could be swing trading or anything else.  But with the 2-3 tick average I know they’re not swing trading (at least I hope not).

To be clear, this doesn’t mean these traders are using 2 tick targets.  It means they’re averaging 2 ticks per contract at the end of the day.  They might have 4 tick winners and 4 tick losers and win 75% of their trades.  There are a number of ways to arrive at 2 ticks average.  I’m convinced that my way wasn’t optimal as I was using loose stops often getting out for +2 ticks, so that’s what I need to work and not the +2.  I hope that makes sense.  In other words, +2 avg is ok but I need to reduce my risk.

I’ve been debating about scaling out for a while now and my current thought is to trade 1 contract and use a max 6 tick stop and target the HVN ensuring that I have at least 1:1 R:R ratio or I skip the trade.  This is a very simple risk management and I need a 50% win rate to be profitable.  Ideally my risk would be 4-5 ticks and my target would be 6-8 ticks (often my levels are 6-8 ticks away from an HVN) and that’d give give a 2:3 R:R (risk 2 to make 3).  With that my win rate would have to be 40%.  Best case, risk 4 to make 8 and that’s 2:1 which requires a 33% win rate.  So averaging all those would give a win rate around 40% required win rate to be profitable.

So my current thinking is that I will give this a shot.  I will keep stats on how often price goes to the other side of the HVN and that will tell me if a second target would make sense or not.

In order to increase the chances of getting a 40% win rate, I’ll need to be more selective with trade areas and that means looking for more confluence and not just fading every level I see.  This will take some more patience but trading two markets makes it much easier to be patient.  Overtrading has been a big problem so this is one way to reduce overtrading.

So this is my current plan.  What I’m doing now is using playback & market replay to practice a bit on historical data.  This will help build confidence and give me some more data to work with.

This is a work in progress and could change over the next few days, but I thought I’d write out my thoughts.  It really helps me in my thinking and I hope you find it useful as well.  If you feel like sharing your view on scaling out, let me know how you’re doing it (or not doing it).

  19 Responses to “Working on the trading plan – Scaling out revisited”

  1. thanks for sharing your thoughts again. it’s really helpfull to see your developing process in order to master this game. Good Luck!

  2. Michael, at least 50% of my targets should lie within a “normal rotation” (5-9 ticks for the FESX at the moment). In order to get the normal rotation I look at a3 minute chart at write down the swings. I know this is discretionary and no rocket science. Then I put the frequency in a histogramm and analyze it. The other scales are for a bigger swing and the SL is around the low of the last swing for a long. 3-4 units are enough imho, don’t need 8.

    Cheers,
    Markus

  3. I plan to do this as you have described. What do you use to determine swings? The fractal indicator?

  4. As a discretionary trader, I use discrectionary money management. Lots of people don’t even realize this is an option, for some reason.

    I think of scaling like insurance. If I’m nervous enough to want less money on the line, I scale out some. If the trade turns out to be a disaster I’m glad I scaled out (just like I’m glad I paid for medical insurance when I need surgery). If the trade is not a disaster, I have to accept lost profit in exchange for peace of mind.

    So, most of the time, no matter how many contracts I’ve got on, I try to treat the trade the same as a 1-contract trade… because I learned to trade futures on 1 contract, and I don’t care to fix what isn’t broken.

  5. Very good analogy Richard and I think this is an excellent way to scale out – to reduce risk and lock in some profits. It makes a lot more sense to do it your way rather than just have a hard target at say +4 ticks. if the trade is going perfectly and not pulling back at all then why get out at +4?

    In the past I’ve “went back to 1” contract and done better (on a per contract basis). so I’m looking forward to it. And then when I add two I’d like to do like you said and only take one off when I sense the trade is no longer looking good and may come back.

  6. Michael, I play swing intraday using Fibs levels to fix entries (0.618), stops (0.786) & targets (0.382-12 ticks)…I scale out with 2 contracts at 6 & 12 ticks using 6 ticks as stop loss…When first target is hit I move stop to breakeven. I only need 40% win rate to be profitable…This trader helps me a lot…http://www.ratiotrading.com/

  7. I think it all depends on how you calculate expectancy, if you include all traded contracts (winner/loser/scratch) then I think having consistently positive expectancy of 2 ticks is fantastic! It puts you in absolute top in this industry. Actually you would be roughly 4 times better than one of the best fully documented day trader in the world – Don Miller http://donmillerjournal.blogspot.com/2009/01/special-post-final-cme-pricing-thoughts.html

    There is so much hype and high expectation in this business that most of retail traders really hope for big wins in day trading and nobody is telling that consistent big net incomes come from either (1) having big number of very small profits per contract using fairly big accounts paying extremely low commission which by itself requires total commitment to actively trade all day long or (2) swinging extremely (for retail traders) big positions with even bigger accounts, same commitment and low commissions. Both styles assume that there is positive expectancy methodology and a trader capable of executing it.

    But that option is unattractive to the public for it requires high capitalization/discipline/structure in everyday slow grinding work.

    Can it be achieved outside of the best prop firms?

    So far I’ve found only D. Miller has done it but am still looking for other provable records, meanwhile I’m modeling my approach after him with recent addition of FT71 input. My expectancy (last 11 months/almost 3000 contracts) is just above 1 tick with no losing month.

    I think your direction is right and you should soon increase either frequency of trades or size but not both at the same time. Go slowly but gradually push yourself up.

    Good luck!

  8. Jan – Excellent post thanks for sharing. I wrote up a lengthy reply but I thought it deserved it’s own post:

    http://www.tradewiththeflow.com/2010/12/29/comparing-my-performance-with-don-miller/

    I also wanted to ask you how you’re learning from Don Miller. Have you purchased his Jellie Webinar?

    Congrats on averaging 1 tick for 11 months, that is consistent. What’s your focus going forward? Are you increasing size or increasing your profit?

  9. no, I didn’t buy Jellie – just following him.

    my plan is to increase size before increasing targets, I want to be very good and comfortable with losing full increased position and making $ with all in/out before trying scale out technique,

    scaling out is a fantastic approach but very dangerous for novice (like me) and I think I will use it only in conjuncture with scaling in (i.e. admitting I’m still not that good at pinpointing my entries),

    I like D. Miller’s position sizing instead of hard stops. Here is how I use it:

    my account is 40k, so I should be ok with max 7-8 cars, for my comfort level (and skills) I don’t allow myself to be in with more than 4 cars ever and knowing that if my trade premise is still valid (for example even bigger divergence, absence of OTF against my position or quick opps for scalps) I still can do it,

    but my base size is 1 car which mentally I always treat as 1/4 of my ideal position, so each time I’m “wrong” meaning I pinpointed my entry perfectly (with no more than 1-2 ticks of heat) and trade works right away, I most of the time take qiuck 2-3 ticks profit, simply because I don’t believe I was that good and it will retrace. Most of the times it does and if trade premise is still valid I reenter, having now better cushion (or FT71 theo avg),

    on the other hand if I’m really right (by using only 1/4 position) and price goes right away against me (without giving me even 2 ticks profit) then I have 3 comfortable choices (all win/win): no more trade premise – take a modest loss on 1 car; premise in but not too strong – wait; valid, strong premise, more diverg – add 1 car,

    usually I go up to 3 cars and then try to exit the last 2 as soon as possible mentally scoring small avg loss on first 2 cars and now being in the $ with only one in case price goes even more against me or I hold that one for little bit longer, kinda like runner, especially when sequence was prolonged and there is more chance of pullback or it’s just like all over again – add, scale out then in and out…and so on..

    notice how I turn almost everything upside down for the purpose of cheating my subconscious mind to think that I’m always right, that I’m always in win/win situation – this little trick helped me the most to overcome fear of loss which I think is basically fear of being wrong.

    by doing this way I forced myself to generate some income (albeit small) with just 1 car at the time and by adding (scaling in) I take even more trades and have even more real battle experience,

    this year it started to be working quite well, hope I wont screw anything with adding VP, I see the value but have to adapt it to my strategy, mainly in terms of risk/money mngmt, I just don’t see myself being 100% in to start my trade

    good illustration of adding/scaling in was today first trade were I was premature but then had premise valid (more div), added 1 car, took 2 ticks profit and waited few minutes for market to show me its hands, “unfortunately” it didn’t give me 2nd chance to add and decided to go up with my original idea (do you see my win/win care free state of mind?) and I took another 2 ticks. One might think too bad I didn’t wait for more but I know there will be another little exhaustion soon and possibly small div…and there was (80% of the time there is) – so it doesn’t matter to me if I take some $ on long or short in this little swing.
    (got in the end 1.75 pt on this swing)
    then I waited whole eternity, like 40 min (slow day) for market to prove me wrong and allow me to add, no luck – took 3 ticks, again felt like too early but saw little positive div, got in and was too early, waited, waited, saw div still in but waited for price to go bit deeper (don’t like to add at less than 4-5 ticks better than original entry), took my 3 ticks just in case this little micro swing turns down (mentally scoring 2 ticks loss on first car) and then waited for VP target which at that point still was 1254 vpoc – it took forever but congratulated myself for being patient.

    http://screencast.com/t/iLs9dHmw

    below some samples how I see my entries, mind you I’m coming from indicator based times (before FT71) but recognize them as only pointers of something being not right (divergence),

    and no indicator is 100% right (not even VP levels!) but most oscillators work well over 80% of the time when on short leash (small targets):

    http://screencast.com/t/9cLLbae67

    http://screencast.com/t/deLyLubsQL

    http://screencast.com/t/NWwE5G3UYa

    http://screencast.com/t/9FGQjD22

    http://screencast.com/t/0W9sy7IEuv

    http://screencast.com/t/glarPZWkMS

    http://screencast.com/t/4x6aOz9kYY

    http://screencast.com/t/dh2qx6cTKA

  10. some more samples

    http://screencast.com/t/H5rU5KxZCG

    http://screencast.com/t/PMGsuz5i7

    on non-trend or slow days I’ve lots of success fading + – 500 Nyse TICK

    http://screencast.com/t/CYRhLHXrP7vP

    also in TF:

    http://screencast.com/t/eZfiqqBfz

    tomorrow will show you something even more interesting.

  11. Jan – great examples. I think it’s your trade management which gives you an edge. I like your thinking on that with the win/win. I think averaging down can be ok if done on a tight leash (as you say). My problem was I didn’t get out when my premise was proven wrong and then took a big loss (5 pts per contract on my biggest loss in December). It seems you will “scale in” but it’s still within 6-8 ticks which keeps your risk lower.

    You have several different indicators and I’m curious what you prefer. On the example from today the indicator isn’t labeled so I don’t know what it is. I’d like to take a look at this today & this weekend. I can use Ninja’s market replay to practice.

  12. Michael, did you check what happened after you close that position, was that 5 pts w/o any rotation, were you scaling, if yes, why didnt you scale something out, I know it’s after the fact that we see everything clear but do analyze it,

    the indicator is simple TRIX w/length 2, if properly used it can be great tool to visualize state of most recent volatility in oscillating market, for me volatility (increase of the length of my bars) is very important, it tells me that the end, peak is near. I mainly use it when price tries second or third attempt at high/low but TRIX notices decrease of volatility between those peaks. On the other hand when it fails it tells me even more. By fails I mean price wont rotate 4-6 ticks on 1 min chart. Then if I’m not in a trade I consider going with the trend at first TRIX dip below/above 0 (it has to actually register – + reading).

    I employ similar tactic w/TICK, after seeing big lots (but that has to result in increase of the length of at least one bar higher than my timeframe, I like to see on 3 or 5 min timeframe bar visually being bigger than previous ones) – if I don’t see that, I continue to treat the market as oscilating. Reading of 500 in trendless day – fade, TICK returns to 0 in strong trend or after OTF – go w/trend…and so on…

    This is all my visual interpretations, only now from FT71 I learnt that it is OTF or LVN or and understanding of auction market.

    Another thing which I use and are scorned by VP is MA’s but I use them for saving room on my screens, quick check, I have few other timeframe and markets in very narrow windows (like showing only few bars) and using 2 MA’s I quickly know where are we (going up/down/flat) and I don’t like to go w/trend on pullbacks on my timeframe if there is no trend higher timeframe …like here:

    http://screencast.com/t/MAwYW6Jt8noC

    I rather wait for an excess and fade it, after all that happens most of the time, something like 70% or so.

    I also use RSI to see excess, TRIX has no boundaries.

  13. Micheal, I was today all day shorting, not knowing if the price ever gonna go down but my method says take next signal no matter what and take profits when you can (usually I also look for some opposite div to get out but not a must for me) plus all indicators (also MACD) were saying there will be some pullback, at least 2-3 pts – keep doing what you know 90% of the time works.

    at the end we got strong down where I exited all (yes I could profit even on my first entries but that is obvious after, so I took 1 loss today – big deal!) and then noticed TRIX is telling “not so fast, tempo is falling” – good opp to buy, see here:

    http://screencast.com/t/MAwYW6Jt8noC

  14. btw, if the price never came down, I would be very slightly down…thats the price of sticking to my plan.

    but chances were more than 50% that there will be 2 pointer down…so stay put and take as many little profits to offset loss if the other 50% is right

    just my outloud thinking and sharing with you.

    my other reason for sharing is that just maybe there is somewhere somebody like me and we can help each other stay tune in…..and you know it is not easy to be focused all day.

    Happy New Year to all !!

  15. I have a question to everyone here, how do you determine your entries, what triggers them, mechanical way, reading order flow (DOM, prints), or something else?

    Regardless what system you follow (MP, VP or anything else), what money/risk management you use?

  16. I use the FT71’s homework process to pre-define the hypothesis and “areas to do business”, when price moved to the “a.t.d.b” and matched a hypothesis, I use MD Footprint readings, DOM, and Time&Sale combined to trigger my entries. I have specific rules for each of these 3 parts, I want all 3 to agree, or I delay to cancel my entry. It’s not perfect, I missed good trades and sometimes ended up entered late at worse price than I just simply enter on limit order. I guess this is because my DOM reading skill is still lacking. (I don’t use any indicators and don’t use market breadth).

    I use 8-tick hard stop as automatic emergency eject bottom, but my stop loss is 4-6 ticks. I scratched too often (at B/E or +/- 1-tick) which is something I need to work on to stop doing it now. I only trade one contract now and my target is 6 ticks but again I often cut it short at 4 ticks or less. My plan for this year is to not scratching or cutting short my winning trades, this is easier said than done though, I am sure all of us had encountered this demon.

  17. Jan – I was using the DOM but now I’m trying the footprint again. I’ve used footprint off & on for a year now. FT71 keeps talking about the ” quality of prints ” and that’s hard if not impossible to see on the ninja DOM (the X_Trader DOM is much better as it shows the cumulative volume at each price). So the footprint helps make up for ninja DOM limitations. Sometimes I still think putting on half with a limit at the level would be best and then enter the other half. That way I’m guaranteed not to miss any trades and if it blows through the level I’m in with only half. That’s just an untested idea though. I think there is no easy way here. Many do it with just time & sales (SMB capital comes to mind) so I think it’s requires a lot of practice whatever tool is used.

    ky_ny – we’re on the same path. I too use 4-6 tick stop now, I used to scratch way too much (often to see my original target would have been hit), and my current focus is not managing the trade. I want either stop or target to hit. I’m not afraid of stops any more. I know in the long run I’ll make more this way.

  18. Jan – Thanks for describing your method and providing all the examples. I looked at it a bit and I think that you’re trade management and intuition account for the majority of your edge. I think it’s not as easy as I originally thought (yes I’m still a sucker for an easy setup). While I’m very curious to learn more, I have to be very careful as this is totally different than what I’m doing now and I don’t want to interfere with my normal trading. In my normal trading I only trade at levels (pre-determined before the open) so in most of the cases where there may be a micro-divergence – I’m sitting out there. To make up for the lack of opportunities & time spent waiting, I trade two markets (Bund/Stoxx in morning and ES/ZN in afternoon). So then I thought ok I could look for micro-divergences at my levels but they are almost always present as price will almost always stall at a level.

    I will keep looking at it and practicing on market replay but it’ll have to be outside market hours. During market hours I may observe as well but not trade it, not even on sim.

 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>

(required)

(required)