First 2010 trading drawdown analysis on the verge of a new record

I am close to passing my biggest monthly drawdown ever in my forex paper trading endeavour. My account drawdown is at -3.34% now. My record was set in last November at -3.44%. And it’s only the 13th! To make matters worse, this isn’t one of those transient, surprise drawdown. It is a slow and steady one. An excruciating one. See Figure 1 for my account balance graph for this month so far. Feel my pain!

My post earlier this month on 10 questions to ask yourself if you are holding a ginormous losing position and don’t know what to do was a real harbinger. I wrote that piece at precisely the top point on Figure 1. If only I can time my trades that well! But instead of a single ginormous losing position, I am faced with losing trades one after another. Thus, drastic measures need to be taken before I mark a new drawdown record. This post is a self-assessment on what I’ve done wrong and my plan to turn this wreck around.

Early bird does not get the worm in forex trading

[This post was updated on January 13, see addendum below.]

I have discussed my enthusiasm in shorting the Aussie a few times this week. First trade was shorting AUD/JPY at 84.53. After that got stopped out, I re-tried at the same level the next day. My trade history in this currency pair is shown in Figure 1 below. Evidently, I am early in my trades as the recent upward momentum is still going. But that’s just speaking from hindsight advantage. What is actually wrong with this play is that I was too eager at the first entry when the price was just at a minor resistance (short-term trendline of Fig 1). In turn, that caused me to lose my confidence and not be as aggressive as I should be when the price reached a major resistance at 85.50.

Referring to Fig. 1, I made 5 attempts on the minor resistance below 84.60 in the first round. What’s worse is that I didn’t have enough patience to wait after these shorts failed and I tried again at 84.52. Even though that resistance level is voided already.

But here comes the kicker. I actually noted the important 85.50 level before my first attempt. It was clearly on the chart and I even mentioned about its significance. I just figured that it wouldn’t be touched. How wrong I was.

As I said, another problem with being early in a trade is that it stripe you of your confidence prematurely in a good setup. My setup to short AUD/JPY is as good as ever now. I am drooling over that negative RSI divergence shown in Fig. 1… But since I’ve already taken a beating in this pair this week, my account drawdown dictates that I limit my risk. As such, I can’t back up the truck to pile in on this trade at this point. But why not?

I won’t break one rule to compensate for another. Although I am light in this trade now, I will not compromise my risk management rules to compensate for my own mistakes. That would just open a whole new can of worm. Besides, I can’t say this often enough — risk management is always the top priority in trading.

In any case, there will always be other opportunities.

The moral of the story is this. Do not be too eager to get into a trade. Secondly, adjust your position size accordingly to the significance of the setup in play.

AUD/JPY, 3-hour

Update January 13:

Figure 2 below shows an updated view of AUD/JPY a week later. The red dot on the 9th was my last stopped out entry. As you can see, I missed the smooth 200 pips ride down. This experience is proving to be better than I expected to illustrate how being early in a trade can really hurt your account.

A week later in AUD/JPY, 3-hour chart

December 2009 Trading Performance Review

I have set 2 goals back in November for December. 1) Think more, trade less; and 2) Be more aggressive on winners. With 276 trades in December, I have failed miserably for goal #1. I couldn’t meet my second goal either as I’ve missed most of the big moves from failing to recognize the bigger picture. Nonetheless, I ended up the month +2.71% in my demo account by picking pennies here and there and with the help of some decent risk management to limit my risks. Overall, it was a bad month for me because of too much time spent trading with little reward to show.

I noticed my over-trading tendency early in the month. My solution was to limit myself to the longer timeframe to reduce moving in and out of trades too frequently. Reviewing that particular post, I was spot on with identifying my problem. However, my over-trading problem lingered on for the rest of the month (i.e. “staring at the trading screen too much once I am in a position and tried to micro-manage my trades to minimize my risk”).

While this can be attributed to the more volatile market in December (discussed later), the real downside of my over-trading can be seen in Figure 1 showing my demo account balance graph for December.

December 2009 Account Balance

Notice the following characteristics in Figure 1.

1. two up-spikes in the month (around the 8th and 17th)
2. a slow, barely-upward grind in the middle of the month (from 9th to 16th)
3. a marching drawdown at the start and end

#1 was mostly luck. I was in the right position at the right time.

#2 was the grind as it should be. I intend to delegate this type of trading to my automated trading system once it’s developed (haven’t even started yet though). But for the moment, it’s all manual.

What’s most concerning is #3. Sometimes it’s just better to not trade at all. That is something which I still need to learn to do. Evidently, I still need to work on narrowing my trading setup to filter out these erroneous trades. I outlined the new steps for that in my previous post already. I just need discipline to follow my own advice.

Performance Statistics

Monthly Metric December (November)
Sarting Fund $5,469.11
End of Month Fund $5,617.44

Total Net Profit $148.33 ($166.90)
Gross Profit $591 ($598)
Gross Loss -$443 (-$431)
Profit Factor 1.34 (1.39)
Net Profit % 2.71% (3.15%)
Max. Drawdown % 1.91% (3.44%)

The above stats look consistent with the previous month’s data in brackets. My maximum drawdown actually improved significantly down to 1.91% from 3.44%. That is due to the fact that I took a lot more small trades to rake in pennies rather than bet big on big moves. Frankly, I would prefer the latter strategy as I don’t typically have so much time.

Sterling Ratio 0.41 (0.41)
Excess Return 2.66% (3.10%)
SDev Return per Trade 0.12% (0.19%)
Kurtosis 17.38 (3.45)
Skewness 3.30 (1.54)

Comparing my performance statistics using my new modified Sterling ratio, it’s coincidentally exactly the same as November. My maximum drawdown improved but its benefit in the equation is cancelled out by the increased volatility in December. My ultimate target is to achieve a consistent monthly modified Sterling Ratio of 0.50 or above.

The kurtosis and skewness statistics got a lot worse this month. These are expected from the large number of small trades and the few big profitable trades as seen in Figure 1. However, what’s alarming is the data below.

95th Percentile P&L $11.97 ($23.20)
5th Percentile P&L -$5.73 (-$10.31)
Net Profit Mid 90% Only -$46.96 ($33.82)

As you can see, it would have been a losing month if it weren’t for the outlier trades. This is definitely the most worrisome statistic this month. I need to stop picking pennies and concentrate on bigger moves.

Total # of Trades 276 (153)
Percent Profitable 40.77% (46.36%)
# of Winning Trades 95 (70)
# of Losing Trades 138 (81)
# of Break-even Trades 43 (2)

As I’ve said before, I took the most number of trades this month since my short forex career. My percent profitable drifted farther below 50% to 40.77%. I might as well throw random dart to make my pick. This has once again proven that I over-traded and didn’t spent enough time with my analysis.

The large number of break-even trades is also a sore-sight in the data.

Average Trade Profit % 0.01% (0.02%)
Average Winning % 0.11% (0.16%)
Average Losing % -0.06% (-0.10%)
Ratio Win% / Loss% 1.94 (1.61)

Max. Conseq. Winners 9 (12)
Max. Conseq. Losers 18 (14)
Largest Winning Trade $47.44 ($41.88)
Largest Losing Trade -$11.64 (-$20.34)

Avg. Time Held [hrs] 0.26 (0.56)

Goal for January

There’s only one goal for me for January and it’s the same as November‘s.

Think more and trade less! Refer to my post on Focusing on trading my timeframe to reduce over-trading to see my plan to achieve that.

November 2009 Trading Performance Review

This month has been a roller coaster in both the markets and my account balance (see Figure 1). Here’s a quick recap of my paper trades in the forex markets for November. I started out by cashing the carried over long yen and dollar positions from October. Those trades could have been better as I have lamented in a post. Nevertheless, it was a good start for the month.

Then, things went downhill for me; literally, as you can see from my account balance at the first half of the month in Figure 1. I tried my luck day trading EUR/USD with horrible results. I marked this particular period of drawdown from these EUR/USD day trades on Figure 1 with a red band.

The fear lingered on and I gave up all the initial gain of the month to scrap just below the opening balance. Fortunately, I was able to realize my problem and clean up my act. I recuperated by documenting each of my entry like I used to to force myself to think more and trade less. The turnaround didn’t happen immediately. Not until I became even more stringent on myself and stuck with using only my best setup, the FTC setup. I marked this period on Figure 1 with the green bar and arrow.

As you can see, I happened to be in the right markets and at the right times to make a full recovery back to the top. In fact, I made my first 400+ pips trade and my best EUR/USD day trade during this comeback.

I thought I could finish the month above the critical $5500 level (10% gain). But a couple small losers on the last day pulled my account below it yet again. However, one lesson I became even more aware of from the draw-down is to forget about your account balance. Focus on making good trades, manage your risks, and the money will take care of itself.

November 2009 trades graph

Performance Statistics

Attached below are my performance statistics for November. The first number is this month’s and the second number in bracket is from October for comparison.

I will highlight and comment on the few discrepancies that I find important.

Sarting Fund $5,089.06
End of Month Fund $5,302.21

Total Net Profit $166.90 ($213.15)
Gross Profit $598 ($499)
Gross Loss -$431 (-$285)
Profit Factor 1.39 (1.75)
Net Profit % 3.15% (4.19%)
Max. Drawdown % 3.44% (2.32%)

This draw-down is horrible to say the least.

Kurtosis 3.45 (8.53)
Skewness 1.54 (2.49)
Sharpe Ratio 16.64 (5.21)

The kurtosis and skewness are surprisingly good. A low, but not near-zero numbers are preferable. The Sharpe ratio is also surprisingly good too. This isn’t because my return is good, is just that the standard deviation return of my trades have been very consistent. Ironically, that’s mostly because of steady small losses (see Figure 1).

95th Percentile P&L $23.20 ($70.11)
5th Percentile P&L -$10.31 (-$33.60)
Net Profit Mid 90% Only $33.82 ($46.34)

I failed to meet my goal from last month to increase my mid-90% return. It would have been easily met if I didn’t go crazy with the day trades and loose a bundle.

Total # of Trades 153 (34)
Percent Profitable 47.06% (44.12%)
# of Winning Trades 72 (15)
# of Losing Trades 81 (19)

I started to make use of scaling in and out of a trade. So sometimes one trade would be counted as 2 or 3 small ones. Still, I made too many trades this month. Even if I consider a 2 for 1 ratio of trades (i.e. 153 / 2 = 76), it is still more than twice as many as last month. My net profit is lower too. Wasted time and effort for lower return. Lesson for the month is clearly to think more and trade less.

Average Trade Profit % 0.02% (0.12%)
Average Winning % 0.16% ( 0.66%)
Average Losing % -0.10% (-0.30%)
Ratio Win% / Loss% 1.57 (2.21)

This could be the most worrisome statistics for me this month. My average winning percentage from a trade is significantly less. I need to be more aggressive when I am right to reap the most reward from it. Stash in the cash as the position gets greener!

Max. Conseq. Winners 12 (4)
Max. Conseq. Losers 14 (6)
Largest Winning Trade $41.88 ($171.79)
Largest Losing Trade -$20.34 (-$44.17)

Avg. Time Held [hrs] 0.56 (2.18)

Goal for Next Month

Let’s start by saying what I did well this month. I think my comeback was phenomenal, but definitely with luck on my side. I approached my huge draw-down systematically and rationally by reducing position size, trading less, and restricted myself by using my best trading setup only. Then when the time was right, I was able to pound on the winning positions. I will need to remember this the next time I’m experiencing a draw-down (hopefully not as much as this month’s).

On the other hand, I made many trading mistakes this month. Most were just short-term problems and I have already written about them in my lessons. Upon further review of my performance this month, two major problems are evident as discussed.

1. Think more and trade less. Limit my trades to approximately one a day and know when to NOT trade at all. Not even with tiny positions to “test the water” as I like to do.
2. Be more aggressive when the position is right. Achieve this by improving on my scaling in and out strategy by accelerating the scaling in and decelerating on the scaling out.

EUR/USD +95 pips in 2 hours: An example of swing and day trade convergence

I had my best day-trading day this quarter on November 27. I made 95 pips in two hours at the open with only two trades. This post is a review of my trades that day for my future reference. To save you and myself from reading, the two reasons for this good run are:

1. Convergence of intermediate term signals with day trading signals
2. Ignoring my bearish sentiment (Dubai news) and traded what I saw from the prices

Before moving on, let’s start by setting the stage with some background information to summarize what conditions were like that day.

Background

The news on November 25 after NY close of Dubai World delaying their debt repayments sent the dollar and yen soaring. I made my first 400+ pips trade by being at the right place and time on this panic wave. The U.S. market was closed that day but FTSE sold off more than 3% the day after the news. So the expectation was for US to do a catch-up sell off when it opens again on Friday 27th. Risk aversion (equities down, yen and dollar up) was the talk of the day.

Figure 1 below shows the charts of ES futures (e-mini S&P500). Left is a 1-min. from midnight of 27th to that morning’s NYSE opening. Right is an hourly showing the move overnight from 26th to 27th.

ES Futures (e-mini S&P 500)

Intermediate term signal

Figure 2 shows my intermediate term signal. EURUSD has been printing an uptrend on the daily (not shown). Figure 2 shows that EURUSD was testing the bottom of my channel so I should consider going long. Notice that this is the same channel that called for the 1.5150 top on 25th. At least it has proven itself useful once.

EURUSD, 3-hour

Short term signals

For the short term signal, we need to go back to Figure 1 for the ES charts. ES tanked 40 points over the holiday with very low volume. It made it all the way to 1067 just before 3am on 27th. Then it lifted all the way to 1080 at 9am. 1080 is the line in the sand for the bulls and bears. So retracing all the way to that level before opening is significant.

This pre-open strength in ES was the sign of oversold for me. This is the short-term confirmation I needed. I entered my first position to long EURUSD @ 9:09 am to test the water.

Then I watched both EURUSD and ES closely for the next half hour. Once NYSE opened. I noticed that the TICK was relatively mellow with readings around -250 to -500. It’s unusually good for a -20 points open drop on S&P 500. One would expect at least -800 readings for a strong negative sentiment. So after watching EURUSD, ES, and TICK for about 10 minutes, I entered a second position on a weak retracement at 9:39 am.

The exits

For my exit, I took profit at 1.5180 because it’s the 50% Fibonacci level and it’s also the top of another of my channel (Figures 2 and 3). After about 10 minutes of watching the 1M bars, EURUSD failed to clear that resistance, so I took profit on the second half to go flat in this pair just before 11 am because it’s an early-closing Friday (NYSE closing at 1pm).

The gains were +95 and +93 pips for +0.27% and +0.28% in my demo account total. Not bad for holding for 2 hours.

Update: Thanks to Jeff on Twitter for the question. I updated Figure 3 with the Fib. levels and showed the top and bottom used to draw it.

EURUSD, 30 min.

Conclusion

In summary, I started with an intermediate term setup that aligns with the major trend. I observed the reaction of the markets on an important news. Identified an oversold condition from the short-term charts for a short-term positive divergence signal. Scaled in a position. Then added more with another confirmation. I scaled out when the target was hit.

Recommendation

Once again, I could have been more aggressive on this trade. As shown in Figure 3, 1.4920-4930 was a short-term resistance. I could have added more to my position when that level broke, with a stop at 1.4900.