Naked November tradelog (scalping)

Trading Systems
84

Wednesday 1/12/11
Quick stats:
3 trades
Trend bias taken from the 5min PSAR (0.03, 0.3)
Trades with 5min trend bias: 3 wins, 0 losses
Trades against 5min trend bias: 0 wins, 0 losses
1.8142% account change

Got stuck in another scaled trade today. Somewhat of an execution error, but still these things can happen and I need to be prepared to deal with them in realtime. The bigger issue remains: how can I learn to just avoid these crazy scaled trades? That is my focus and I need to continue searching for a non-arbitrary way to take losses. I can dig myself out of a lot of bad situations, but I know the day of reckoning is just around the corner…

800×559

800×561

85

Good idea! I need to pay better attention to the chart patterns, that is something I’m definitely guilty of. I hope these will become more natural to me with screentime. Btw one thing I’ve done recently that has been helping I think – is to reduce the grid spacing on the 5min and 30sec charts. Previously my rule-of-thumb was to target about 10 pips per trendline on the 30sec chart. Now I’m targeting somewhere between 5-10 pips. This has the effect of drawing more resistance areas. To me the grid is a fractal representation, so it can go on forever. Too many lines can become messy or confusing, but too few lines can make scalp levels difficult, so it’s a balance for me.

I agree with you 100% on this. Please share any ideas you have about “training” as you say. I am currently searching for a way that I can know when a setup is going bad. I’m hesitant to use an indicator because I don’t feel that it should be necessary. However, if one did exist that would spotlight a problem, I’d be glad to study it. I haven’t had much luck using the PSAR as a stop because it seems to be whipping around too frequently in recent markets. I do like 5min PSAR for a dynamic trend indicator and it allows me to trade in both directions with confidence. I might need to adjust the settings more; I’m still studying it’s behavior.

One other thing I wanted to point out… I need to go back and re-read, but pip-siphon made a comment at one point that, in the beginning of his scalping journey, he found that he was taking a lot of scaled trades. Over time he said these basically went away because he gained the ability to identify price action better. However, he never completely eliminated them. It is clear, though, that he does know how to take a loss, and there were a few of these shown in his most recent charts posted.

So the challenge for me is to figure out when price is overextended, to the point that it makes reasonable sense to get in another leg of the trade idea. But also, I must know when momentum has shifted enough that I cut losses.

THANKS for your input, fresh ideas are greatly appreciated!

86

I have trained myself the hard way, by using hard TP and SL of 2.0 and 5.0 respectively.

Then I used soft (verbal) SL to give myself some room and avoid stop hunting (usually done by professionals rather than your broker). A hard TP is necessary when the market is moving rapidly - you never seem to be quick enough - and when on ‘autopilot’.

For direction, why don’t you use 20 ema? That is well respected by EURUSD.
Also used by Pip-siphon together with 2 and 3 BB.

Also you can use the 200 BB 1,2,3,4 to indicate overextension. I had a scaling trade last week that went from BB-4 to BB+4, now what’s the chance of that? That’s forex for you, never a dull moment.

87

Thursday 1/13/11
Quick stats:
5 trades
Trend bias taken from the 5min PSAR (0.03, 0.3)
Trades with 5min trend bias: 5 wins, 0 losses
Trades against 5min trend bias: 0 wins, 0 losses
1.9258% account change

It’s interesting to me how you can see arguments for both an uptrend and a downtrend at the same time. On one hand, this is great when you’re in a trade and you need a little confidence. On the other hand, if you have problems executing then you could come up with a million reasons not to (analysis paralysis). But mostly I see it as a positive because I think with proper management you could make scalping work in any conditions.

I am still using the 5min PSAR to dictate my trend bias. What I’ve found is that it tends to be somewhat elastic. Price will swing waaaaay far to one high/low, enough to breach the PSAR, then snap back. What I try to do is identify these swing channels ahead of time. I like to scalp with the PSAR when it’s near the middle or nearing the snap-back edge of the channel. Price action is pretty predictable as it approaches the channel boundary. I like to see it slow and out of breath (the only other thing I see is a forceful breakout usually). Often the PSAR will be breached after I’ve already taken one or two legs. I don’t worry about this for the current trade. Instead I adjust my bias after the current trade concludes (sidenote: this is why I’m unable to use PSAR as a stop indicator, at least in its current form).

I may continue playing with PSAR to see if I can find a less-whipsawing group of settings. But, on some level I like the whipsaws because they seem to catch the mini-trend bias pretty good.

800×591

800×592

88

Thanks again for your insight ikdenk.

I will definitely take your ideas into account. I’m unsure if 2.0 hard TP is a good idea. My average TP is higher than that. I am generally targeting about 3, or sometimes a logical level if the price is moving gracefully (like some cases today for example).

I will play around with adding some BB’s and EMA. But I’m hesitant to put these on my primary trading chart (ie 30sec) because I’m afraid they will start dictating entries for me. I’m trying hard to avoid a highly mechanical system like that. I might try putting them on a separate chart in the background though, like 5min. It would be nice to have a dynamic way to identify over-extensions.

89

Friday 1/14/11
Quick stats:
5 trades
Trend bias taken from the 5min PSAR (0.03, 0.3)
Trades with 5min trend bias: 5 wins, 0 losses
Trades against 5min trend bias: 0 wins, 0 losses
1.3537% account change

Today felt pretty normal, albeit a little slow at times. I’ve started watching the 5sec candle charts from Oanda (it’s the smallest timeframe available). I think there could be serious clues in this, I’m trying to listen.

800×559

800×562

90

Monday 1/17/11
Quick stats:
3 trades
Trend bias taken from the 5min PSAR (0.03, 0.3)
Trades with 5min trend bias: 2 wins, 1 loss
Trades against 5min trend bias: 0 wins, 0 losses
-9.8308% account change

Today turned out to be a big learning experience. Or at least, I’m planning to make it one. Today is a day I probably shouldn’t have traded (being that it is a USD Bank Holiday and all), but in retrospect I’m actually glad that I did.

I’ve known for awhile that the stop controls on this strategy are messed up. I feel that I cannot move forward if I don’t get this one part under control. So I’m taking this opportunity to re-focus my efforts. The naked trading seems to have merit, but without a plan for handling stops, I don’t think this could work long-term.

I’ve actually had this same problem – trying to control stops – in my previous trading life as well. It is the one problem that has kept me from moving forward, getting to that next level.

I want to take a step back and make note of the specific things that ARE working with this strategy. In many cases these are goals that I’ve accomplished:

  • GOAL: extremely simple, no complicated indicators or rules
  • GOAL: for the most part it’s 100% math-free, frees brain to study crowd psychology
  • GOAL: fits in a specific trading window, no obligations after I conclude (note: with stops under control, there should never be a need for autopilot or limit orders)
  • GOAL: ability to produce a positive return each day consistency

Now the things that AREN’T working:

  • ability to take stop loss when trade not working
  • ability to stop trading at end of trading window without need for limit orders

My objective with this study is not to see how much I can ramp up my account balance (though that would be nice). I’m reminded that the MOST IMPORTANT thing is for me is to learn how to master consistency. This must be all-encompassing for a complete trading plan: consistent setup planning, consistent interaction with platform/broker, consistent rules application, consistent executions, consistent profits and consistent journaling.

I’m planning to do some exploring at this stage. I want to play around with some ideas to help with stop controls and see if I can find a breakthrough. During this time, I’ll scale back my trade units to minimize the bleeding. But I’ll keep an eye on the percentages. My objective should be to produce a consistent net positive return of any amount, while constraining risk. Once this could be accomplished, I can then go back and recalculate the risk thresholds.

Here are some training ideas I had that might help me grow through this:

PHASE 1:

  • no scaling
  • hard stop
  • hard takeprofit

PHASE 2:

  • no scaling
  • mental stop
  • hard takeprofit

PHASE 3:

  • no scaling
  • mental stop
  • mental takeprofit

PHASE 4:

  • scaling
  • mental stop
  • mental takeprofit

800×559

800×562

91

Dusktrader,

I feel your pain, even looking at your -9% loss brings back memories. I have been looking in on your thread since you started and you were on a good roll for a while there and I was crossing my fingers for you. But I have had the same luck previously with consistent winning days (with this scaling based trading plan) and BOOM one bad day. Setting in place my mental hard SL have been my life line. Also I noticed you mentioned a few times you have raised your position sizes. I have played around with my sizing for awhile and I try to get a scope for the days trading, the range of how far it may or may not travel and then adjust my position sizes accordingly (all on personally preference). I don’t really change it much but smaller position sizes yield longer scaled position flexibility and vise versa. But I still maintain my overall hard mental SL. Just some food for thought.

92

Commiserations on your loss. Unfortunately losses, even big ones are just a part of trading forex.

I believe that scaling is a crucial part of this technique, especially at the beginning. Scaling determines most of the wins, but unfortunately also the big losses - the skill is to make them rare or better non-existent.

I limit myself now to five identical sized trades, the original plus four scales.
When the loss exceeds 2% of my account, I take it on the chin. 10% is way too much.

Without scaling and a hard stoploss and take profit, you are reducing the trading to a form of gambling where you have to achieve a certain percentage of wins, usually around 70% to break even, let alone show a profit.

BTW, what hours do you trade in GMT? Some hours like the Frankfurt, London and New York opens are much more tricky because of frequent, violent break-outs that would ruin any scaled trades.

93

Tuesday 1/18/11
No live-trading today, as I am exploring options for controlling stops better.

Today I was working on a theory (described below) that felt productive. Unfortunately it took me close to an hour to get my demo platform up and running properly, due to all the trendline drawing (I don’t have to start from scratch each day on the live platform since I’m only updating the new Daily, 3hour, 1hour and 15min trendlines). I will play with this again tomorrow and see if it tells me anything new.

I’m looking at a couple different indicators that Oanda offers. My objective was to find a way to have a dynamic stop value rather than a hard-stop. Specifically, I want an indicator that is relevant to current volatility, as this seems to be the most logical way to determine a mechanical stop level. Today I chose to look at ATR (average true range) which seems to correspond nicely with the height of the candles (ie, when they start getting huge, ATR increases).

I am hell-bent on keeping my trading chart free of indicators. I believe it’s possible and also is an exercise that promotes mastery. So for these reasons, I try to go out of my way to keep the 30sec chart pure. The only lines you see on them are: trendlines and psych levels. In the future I MIGHT consider adding fibs and/or wave counts, but for now I think there are plenty enough lines.

As such, I am ok with putting indicators on the 5min chart, which is what I look at in the background. This chart is never used directly for entering or exiting trades. I am still using PSAR(0.03, 0.3) on the 5min chart to tell me which direction to open new trades in, and so far I’m happy with this. Now I’ve added ATR(14) also to the 5min chart.

To keep things simple during this experimentation phase, I am setting the default takeprofit to 2 pips (note, as can be seen below, there is a a fair amount of slippage going on. The few winning trades had actual pip gains of 3.3, 3.8 and 3.1. A similar degree of slippage seemed evident with the stoploss orders as well.)

For the default stoploss value, I have used this: I glance at the ATR value (in pips) and take half of that rounded up. In most cases today the value was 5 or 6, fluctuated a couple times.

On the chartshot, I’m trying to figure out what information would be relevant or useful to me in retrospect, after I’ve collected some trading data. I’ve shown here the estimated max drawdown pips and also the max potential pips on winning trades. This info could be useful later on to optimize a strategy.

The hit rate was 50% exactly, but not too many trades due to the limited time. I’m curious if this will remain near 50% over time, or if it is possible to eek-out a higher success rate (without scaling).

800×591

94

Hi crashtriple, just to clarify… the way that I figure units is 250 units per $50 of equity. I have recently started adjusting this each day as an exercise to keep me focused on percentages and not dollars or pips. I will probably need to revisit this calculation if/when I come up with a better stop-loss strategy. I don’t want to continue looking at 5% or 10% as realistic levels. I’d rather be out of the trade much sooner and just make up the loss with more profit trades. If a tighter stop can be proven profitable, I see that as a way to increase leverage on the trade, because I calculate the units based on the potential max loss.

Yep thanks and no problems here… I want to look at losses as “wakeup calls” and opportunities. I am looking for real (wisdom) growth and not fly-by-night results. I would rather stumble and bloody my knees now so I can create a well-oiled profit machine down the line.

I believe that scaling is a crucial part of this technique, especially at the beginning. Scaling determines most of the wins, but unfortunately also the big losses - the skill is to make them rare or better non-existent.

Glad to hear someone else say this. I’m pretty sure pip-siphon would also agree, as he never dropped scaling altogether. If I recall correctly, he stated that in the beginning he took a lot of scale trades but over time he began to recognize the prime setups more intuitively. I hope that will happen for me as well. I’d like to think that 50% accuracy is not acceptable. MyFXBook shows accuracy around 70% but it’s hard to gauge that because they look at individual trades whereas I consider a group of scaled trades to be a single/averaged trade idea.

I limit myself now to five identical sized trades, the original plus four scales.
When the loss exceeds 2% of my account, I take it on the chin. 10% is way too much.

I agree 10% or even 5% is more than I want to risk. I would be thrilled to death if I could make a decent profit only ever risking 2%. Interesting that you will scale up to 5. I probably need to take another look at that, because the limit of 3 came from an older study. I prefer to create more dynamic rules though, than hardcoded things like this. A dynamic rule for scaling might be looking at BollingerBands, STARC or StdDev lines. Seems like the likelihood of price reaching each successive outer level becomes more and more slim.

One question I do have for you regarding scaling: with your max 5 legs, have you explored scaling unequal weights? I have thought about the idea of scaling heavier or lighter, but I’m unsure what the impact would be. For simplicity, I have kept all the legs the same. I do know that with 2 legs scaled, I need price to move back near to my 1st leg entry price. With 3 legs scaled, I need price to move back just beyond the 2nd leg entry price. Either of these situations is a huge profit (usually 5-6 trades worth). If I were to reduce weighting on each successive scale, it would allow price to run further before hitting the stop level, but it would also have less impact on the average-position line, which is how I base my takeprofit level.

Without scaling and a hard stoploss and take profit, you are reducing the trading to a form of gambling where you have to achieve a certain percentage of wins, usually around 70% to break even, let alone show a profit.

Agreed. I am working with your idea of hard-stop and hard-takeprofit now as an exercise in discipline only, and to see what it may tell me.

BTW, what hours do you trade in GMT? Some hours like the Frankfurt, London and New York opens are much more tricky because of frequent, violent break-outs that would ruin any scaled trades.

I trade 4:30am-7am Eastern on most days (note: setup generally not complete until after the first half hour).

GMT 09:30 - 12:00

95

Wednesday 1/19/11
No live trading today. Still playing around with ideas for stop-controls.

800×589

800×592

96

I use only one lot size for all trades to simplify matters for now. Sometimes I will scale two trades at the same time at a major S/R.
When I scale, whether it be 1 or 4 additional trades, I am no longer looking for 3 pips, but will exit all trades as soon as the p/l turns green. I will even close at a (hopefully small) loss if there is no further movement.
Scaling is risky and needs to be tightly managed.

As far as your trading hours are concerned, you are more of a dawntrader than a dusktrader, lol.

I think scaling is even riskier in the London session because of the high volume and volatility. Maybe one scaled trade at the most and a quick exit if it doesn’t work.

In the end pip-siphon only traded the ‘quiet’ Asian session and only for a short period too. Good on him (or her)!

97

Thursday 1/20/11
No live trading today. Still playing around with ideas for stop-controls.

Oops. I just realized mid-trade today that I have had my default takeprofit set incorrectly since I began this testing. The other day I commented on the slippage factor which seemed high, but in reality my platform was defaulted to a takeprofit of 3.0 instead of 2.0.

800×589

800×590

98

Agreed it needs to be watched like a hawk. During scaled trades my eyes stay glued to the screen nonstop. I’m afraid to get up even for a cup of coffee! But, I am not satisfied with a wash-trade or a loss, after bearing such risk. If the price is going to move all the way back to my average position line, then I want 2-3 pips of leveraged profit out of it. This generally works very well.

As far as your trading hours are concerned, you are more of a dawntrader than a dusktrader, lol.

LOL – yeah well, the history behind that is that when I started trading my goal has always been to trade outside of the dayjob and/or daylight hours. When I was a swing-trader, I did my setups at night. I am more of a morning person these days and the schedule doesn’t bother me at all. In fact since I started this schedule, I jump out of bed immediately without snoozing my alarm. (My wife thinks this is abnormal, but she’s not obsessed like I am either.)

I think scaling is even riskier in the London session because of the high volume and volatility. Maybe one scaled trade at the most and a quick exit if it doesn’t work.

I have nothing to compare to, as the time I listed is the only time I have available for trading. So I have to make it work.

In the end pip-siphon only traded the ‘quiet’ Asian session and only for a short period too. Good on him (or her)!

LOL … yeah it occurred to me the other day, maybe pip-siphon is a female? Haha – we may never know.

I’ve found the time I trade to be moderate volatility, generally speaking. There was one day that I did trade for several hours into the New York session, but it didn’t feel too much different really. I think if there are not hot news releases coming out then the trading should be fine.

99

Friday 1/21/11
No live trading today. Still playing around with ideas for stop-controls.

Now that I’ve got a good handful of trades to work with, I’m going to focus on analyzing the data. In addition to the raw spreadsheet data, I’ve captured the max drawdown for each trade, as well as the max potential gain. I think this should be enough information to get started with.

Trading exactly like this is not profitable because the winners are too small and the losers too big. But I’d like to play with some scenarios and see if there are any hidden gems to be found. It would be fantastically awesome if I could eek-out a consistent gain of ANY percentage trading like this. That would enable me to crank up the trade units in a MAJOR way due to the extremely tight stops and lack of scaling risk.

By the way, as a sidenote: I was noticing today that this “style” of trading (if I could make it work), is even less stressful than what I have done recently. The reason is because it is more mechanical and less discretionary. The only discretion used would be the entry points. After that, I can sit back, go get a snack, whatever. But in the original strategy I was following, I had to use discretion in both entries and exits, and also determine when or if I should scale.

As a random sidenote #2… I noticed something really cool this week about using my iPhone with Oanda’s app. I have never setup my live trading account with the Oanda app, but I do use the demo account with it. Generally speaking, when I have to put my account on autopilot, I would enter a 1unit trade in the app so that I could follow the progress. I like the app because it notifies me immediately if the trade hits my stop or target.

But what I figured out this week on accident… is that once your account is programmed into the Oanda app, it will automatically notify you of ALL stop and targets executed, even if they did not originate on the app. So this means that my phone buzzes every time a trade closes. This enables me to wander off to the cookie jar, for example, and to be notified immediately when I need to return to the computer. NIFTY!!

I’m hoping I can find a gem in the study I’ve done this week… I’m just itching to get back to live trading next week.

800×561

800×561

100

Dusktrader, do you use tymens CBL [I]count back line[/I] method for your entries? I like this method, its very accurate and easy to understand.

101

Hi tokyotrader. I have not heard of this CBL count back line method, but I’m trying to find out more about it. I see there is a thread here on BP from tymen where it is referenced so I will dig around and see if I can figure out what that means.

Personally, I don’t think ENTRIES are the issue, it’s more the exits… more specifically, when to take a loss. I have real problems abandoning a trading idea and calling it a loss. My mind would rather support the original conviction and keep scaling, it does that naturally I think.

Meanwhile, I’m still here. I traded only a half-session on Monday and it was really bumming me out. On Tuesday for the first time I didn’t even feel like waking up to trade. That’s not good, it means I’m losing momentum. I don’t want that…

I start getting depressed if I am not continually fueling a “light at the end of the tunnel” proposition. In other words, if I can’t see that I’m going anywhere, then I will lose interest.

I crunched and crunched numbers all day Friday, trying to make heads or tails of the trade stat info gathered from last week. At best, it seems that trading without scaling can be about 50/50. After all those trades, the best I could hope to do is breakeven. So that doesn’t really help me at all I think.

One thing that WOULD help is if I could find a way to detect when scaling is not appropriate, having a non-arbitrary way to take a loss. This is a big problem for me, I must find a way to determine this. There must be some characteristic of a trade gone awry that shows up in the price action pattern. If I could spot this, I could train myself to take action and cut the loss. One problem I see is that the likelihood of this behavior is very infrequent, meaning I don’t have very many opportunities to play with it. It’s less like playing with fire and more like playing with dynamite.

So taking a step back is helpful to me in times like this. I do have other projects to work on, and perhaps during this time my mind will percolate a little richer on this thorny issue. One other project is my mind movie, which I’m excited to get completed. Another is continued reading – I have a growing reading list but very limited time to devote.

One brainstorm I had, not really developed yet, is this: what if I were to more clearly define levels for scaling. These could be defined not in hard pips per se, but based on the discretionary trendlines I draw each day. This means they could change with each unfolding market. If I set a trading rule that says (for example)… only scale at the predefined levels, and only scale max 2 times… then on the 3rd level I must cut the trade as a loss. Doing it like this creates a definite loss point which makes it easier for me to then work backwards and calculate a risk leverage. I would like if the stop level were moderately tight, because this would enable me to take a higher units on the trades themselves.

One thing that might be helpful to know is how frequently price would move strongly, and how many levels it would cover. I could probably gather some of this from the history chartshots. For example, if it’s very unlikely that price moves 3 or more levels in a single thrust, then that supports the idea that I could scale up to this level (and that beyond that level might be a lost cause).

102

I´ve bin scalping on a demo account (Oanda) with a good result for a while now, are they a good broker to take this live with a account of 15-25k eur since they are a marketmaker?

Do they work against you if you are doing 2% or more per day?

cheers

103

Dusk,

You trade NY, right? Have you considered trading the Asian session?

NY session has a lot more momentum. Seems to me you’re gonna have a lot of times where you need to scale because you’re fading moves that have so much momentum. Pipsiphon mentioned he trades the Asian session because it lacks “follow through”. I feel that you’d have better results trading this method when it’s ranging, which happens much more during the Asian. Either that, or get extremely good at identifying solid ranging markets during NY.

You had also asked what Pipsiphon was referring to when he mentioned tick data and reading the tape. I believe he was talking about the level 2 data you see with ECN’s, not actual tick charts. You can clearly see the MBT level 2 data in some of his screenshots. On the right side (black area) you can see what people refer to as “the tape”. It shows all the trades taking place. I’ve included a screenshot.

Hope that helps.