Bot Battle!

I appreciate the bizarre side of price action when it occurs…as long as it doesn’t cost me money. Today at the end of my session I witnessed a bot battle where 2 (or more) programs got really mad at each other for about 30 seconds. It was a ferocious battle and there was no clear winner. The 6E traded about 7000 contracts in that time, usually spikes only get up to half that (and hardly ever during NY lunch). The only other time you see that many contracts traded in such a short amount of time is when price freight trains 50 pips straight in one direction. But this time, price didn’t even move more than 10 pips until the battle was long over.

algo-war

May as well show you today’s trades as well (including the poor entries mid-range), leading up to the algo battle on the right-hand side:
4-12-11-why-midrange

Posted in trades | Leave a comment

New Tools

I’ve been using a couple new tools lately to help confirm entries and for post-trade analysis. The first is the Better Volume indicator, which is the same as a normal volume indicator except that some bars are colored according to different classifications. See the creator’s site (http://emini-watch.com/free-stuff/volume-indicator/ for more info. At first I was skeptical…I mean how can a “Volume Climax Up” bar mean the start of an uptrend OR a top OR downtrend continuation?? But that’s how volume works…and the more I continued to see the places that these colored bars occurred, I realized that interpretation of a high-volume bar has a lot to do with where it occurs within the macro price context. As long as you realize that the colored bars are not explicit trade triggers (after all, they are just being classified based on a simple algorithm), it’s a great tool. If you understand WHY the algorithm may have colored the bar a certain way (for example ignoring bars that got colored due to news, or a breakout from a tight zone, etc), it can guide you towards the correct interpretation of a volume spike a little quicker (and you’ll be amazed pretty often at how well it can call swing highs/lows).

The second tool I’ve been using is Tradervue, for journaling trades and reviewing statistics about trade types/categories/timestamps/etc. (it’s free at www.tradervue.com) The main advantage it has over other products I’ve used, including tradingspreadsheets, is that it can import trades from a bunch of different brokers/platforms, so you avoid the hassle of importing each trade’s info one by one. Once trades are imported, it has a pretty user-friendly and fast interface for categorizing trades and adding notes, then automatically generates reports and statistics. It makes sharing trades with others easy too, using an automatic blog widget.

Here are a few trades from a couple days ago detailing my current favorite setup, based on price extension with heavy volume, with an entry just after price hits a previous S/R zone and stalls: (these are 55-tick charts of 6E)

3-14-11-658am

3-14-11-107pm

3-15-12-656am

What you don’t see on here is the volume surge, because I track volume on 1-min increments on another chart. You will also notice that I leave money on the table, but I prefer leaving money on the table rather than getting nothing out of a trade that offered some. It’s easier for me to execute consistently when I’m making base hits rather than hoping for a bigger move and getting stopped out at breakeven, then hoping for an even bigger move next time to make up for getting nothing out of the previous trade, etc…that slippery slope we’ve all dealt with. Plus, I always remind myself that as someone wise once said, “if you aren’t leaving money on the table, you aren’t making money.” So if you’ve just closed a trade at your profit target, who cares what price does after that? Just look for the next “grade A” setup.

Posted in strategy | Leave a comment

Volume Signals

Are you looking at volume when you trade? If you’re not, you should be. Trading just by looking at price only shows half the picture…you can see where price is going, but you can’t tell who’s participating. I absolutely love dissecting price action, so let’s look at today’s chart of the euro, and how you can scalp for an easy 4-5 pips or more based on what volume tells you:

(click to open in a new window)

3-8-11-volume

A) Well, before we get to A, look at the giant volume spike at about 6:35. This was just after unemployment claims had come out and also the ECB was having a press conference that started at the same time, where one of the first things that was said was “Available survey indicators confirm signs of a stabilisation in the euro area economy” and they talked about interest rates. So, since I avoid trading during news volatility, I skipped this first volume spike. So starting with A, what’s going on? It clearly shows traders who thought price would be taking another run to the top, but price is quickly rejected and they’re trapped. The high volume indicates the frenzy of not wanting to miss price rocketing off to the moon followed by the panic of being trapped in a reversal–a setup with very predictable order flow about to happen. Two minutes later there’s even greater volume from longs getting stopped out and reversal-callers getting on board.

B) Not as significant a volume spike, but there was just enough of a pause in price a minute before B to get any late-coming shorts a chance to hop on board…but when price snaps back up, they too are trapped. There’s also volume from prior short profit-taking as well as price hitting a previous area of demand. When B starts bouncing back up, 4-5 pips were easy to snag.

C) If you sit and wait for volume to tell you something, C is an obvious spike in volume. The tape shows us that not only has price broken out from consolidation around 1.3250, but it is also trying to break out from the swing low at 6:30. Well, price rarely moves in a straight line, and to try to deplete all the demand below the swing low after price had already extended itself a bit isn’t likely…another easy fade. Remember, if volume is that big, it means everyone is already on board, so who’s there to keep selling to drive price lower?

D) D is the next spike, with volume most likely created by breakout traders in addition to bottom-callers at C who didn’t take profits and are now getting stopped out. When price hesitates, though, breakout shorts start getting anxious and nervous, and that is exactly the emotion that provides for a good entry! It wasn’t an immediately profitable bounce back, but the hesitation spooked enough people so that at their next opportunity to close out at breakeven or + a few pips (just below D) they took it, and price snapped up offering a profit potential on the long. When I took this trade, I didn’t like that I took some heat (however a manageable 5-6 pips max), but the fact that the continuation volume died off told me there wasn’t anyone serious about shorting at this point, and I held until it popped up and hit my take-profit.

E) Ok at this point, price is definitely going down, anyone can see that! Except that is exactly what the pros want you to think. As soon as this breakout fails, it’s easy pippin’ to the north when shorts cover for a loss. There’s also additional order flow from any previous shorts who got scared on the upspike between D and E and are happy for the chance to exit at breakeven.

F) The volume surge at F shows that now everyone is convinced that we’re back to the uptrend and nobody wants to miss the train. Except that price has just hit a previous area of supply/demand imbalance and has also just taken out the stops that anyone placed above the correction between C and D. If there were any real weight behind this move, price wouldn’t have stalled at the top of F. A stall after a surge in volume means fade the move.

G) After patiently waiting for the next spike in volume, it comes at G when it seems obvious that the downtrend is going to continue. This gets shorts on board, who then begin cursing when they realize they just shorted into the obvious support level formed after F. When price moves back up and they get in the red, they start closing out their shorts in a hurry, and 4-5 pips are easy to snag. Plus, isn’t that the “pin bar” that everyone looks for? (Then notice when the pin bar fails, there’s a rush of order flow the other way when longs are getting stopped, until it hits the support level again…oh forgot about that, now we’re going back up, thanks for the cheap contracts!)

H) The next volume spike is at H, and at this point, it should be kind of obvious that the trend has decided which way it wants to go, so this is an ill-advised short. However, you were given 2-3 minutes to realize that price wasn’t following through to the downside and that you were about to get freight-trained (so scratch the trade)…but, this also shows that no method or strategy wins all the time…you have to deal with the fact that sometimes your take will just be wrong.

I) The volume spike at I was an opportunity for a quick fade, but at this point, we’re in a pretty strong uptrend, and there really isn’t anyone “trapped”, so it’s more risky of a trade. If you can learn to distinguish the times when you are CLEARLY trading against a strong trend and avoid them, your account will thank you.

J) Look at how volume dropped off to its lowest average level of the session just prior to the spike at J. This should be an indication that trend strength is petering out. There is one last moment of panic when any longer-term shorts get their stops blown that they put above A or the spike before A and absolute fools think price is breaking out from said high and going to the moon. When price is quickly rejected and starts coming back into the prior zone, it is finally a safe time to fade the strong move. The difference in J as opposed to H or I is that J represents trend exhaustion, capitulation, and the volume tells us that a LOT of people are in a frenzy (it is 2-3x the volume of H or I).

K) K is a sucker move for anyone who has been sitting on their hands and is now confident that a reversal is in order. But you can’t be that late to the party! Who would be shorting here, directly into support formed before the capitulation high? When you see dumb moves like this, the volume represents people not wanting to miss the short train combined with big money putting in big orders saying “thanks for the cheap shares, but in case you didn’t notice, we’re in an up trend”.

L) This is basically the same thing as A. But there are so many reasons why anyone going long at the L bar is a complete fool–we’ve just had serious rejection at J, anyone who got long in the J region is happy to cut their losses (adding to selling order flow), the smart money who got long at K is taking profits, etc.

There really isn’t anything special about today’s action…pull up another day and see if you can spot similarities. These moves happen day in and day out, because panic and greed never change. The best part is, there’s no prediction taking place. You aren’t calling tops or bottoms, you’re waiting for other people to make calls and then you’re just scalping the predictable order flow when they’re proven wrong and start to panic. There isn’t even any guesswork–a volume spike is EASY to spot, and once you see one, determine who will be trapped if that move fails.

I know I just wrote an entire novel about just 3 hours of price action, so thanks for reading, and I can’t help it, I love this stuff!!

Posted in psychology, strategy | Leave a comment

Active Trade Management

Should you actively manage open positions, or is a “set it and forget it” method better? Many times in the past, I’ve found that any time I made an intra-trade decision, like moving a stop or target order, or closing a trade before the stop/target got hit, it was somehow the exact wrong decision to make at the time. I’ve decided to do some number-crunching, because my scalping style of late has less to do with predicting market swings, and more with waiting for temporary moments of the market being out of balance and quickly nabbing a few pips as it returns to balance. For example, failing reversals within the context of a trend.

I noticed that in many cases, it’s obvious when the trade isn’t working, and I’m almost certain my stop is about to get hit. But, I don’t close out the position, because lots of times, price does come back in my favor. I mean, that’s what the stop loss is there for, right? Well, I decided to do some analysis on the actual odds of a trade being a winner or a loser the longer it takes for either to occur. I only looked at setups that occurred this week, but even with a limited sample size, the conclusion is notable.

When I take a set-it-and-forget-it approach, I found that my setup yields about a 65% win rate. However, about 19% of the time, it takes >10 bars (an arbitrary number representing price hesitation) to hit my profit. But the most interesting discovery was that price immediately runs to my stop loss only 6% of the time…the other 29% of the 35% of losses take longer than 10 bars to hit my stop. So, if price hangs around my entry point for >10 bars, there’s about a 61% chance it will eventually hit my stop loss rather than my profit target.

Actively managing trades that aren’t working after 10 bars mean I will be closing about 48% of trades near breakeven. I plugged my numbers into my Expectancy Calculator to see how active management affects my trade expectancy.

First, the results of not managing open trades–a 65% win rate of 4 pips and 35% win rate of 6 pips:

nonactive-mgmt

Next, the results of a 46% win rate of an immediate 4 pips, 48% scratch trades (averaging out to 0 pips), and 6% of full 6-pip stop losses:

active-mgmt

The conclusion is the same regardless of my target/stop size…the point is, if only a small percentage of my losses occur immediately, I have ample opportunity to close losing trades before my stop is hit. My R:R is “worse” with active trade management (if you count scratches as wins), but it doesn’t matter. As you can see, closing trades that don’t go my way quickly increases my expectancy substantially. Getting more conservative, with a win rate of 36%, scratch of 50%, and loss of 14%, expectancy is still better with active management. This conclusion is nothing new, as I’ve been a fan of Mike Reed’s style of scalping for quite some time. He says:

No matter what the market is doing, you must demand that it moves in your favor soon after you enter, otherwise you get out as close to break even as possible. This means you’ll be closing a lot of trades near break-even within the first minute. This is the foundation of learning to trade for consistent gains.

But crunching the numbers for myself really drives home the point. Now, your style may be different, if you like to let your trade “breathe”, or if you are trying to make larger directional swing predictions. But for my style of profiting from very temporary moments of market imbalance, active trade management is essential. What about for your style?

Posted in strategy | Leave a comment

Zombie Trades

One of my biggest weaknesses lately has been letting a trade that initially goes in my favor turn into a full stop-out. The emotional barrier to action is very powerful, because it forces me to admit that I had a chance to profit but missed it. I think to myself, “Price was JUST in the profitable zone, so it should get back there soon, and THEN I’ll close the trade.” This inaction during the time when I should admit the trade worked, but isn’t working anymore, and scratch the trade causes unnecessary loss. I guess it depends on your definition of when a trade becomes a “winner”, but as a scalper, I think of anything that goes 4 pips or more in my favor as a potential “winner”, and I too often commit the cardinal sin of “don’t let a winner turn into a loser”. I find it frustrating to close a trade for nothing when I had a chance to close it for a nice profit. Getting over these frustrations and realizing I have to play the odds correctly is key to winning.

Similarly, there are times when a trade goes against me quickly and I almost get stopped out, but price doesn’t quite hit my stop and then comes back into profitable territory. I like to call these trades “zombie trades” because they have miraculously come back from the dead. What just happened, however, is that the market told me that my entry was probably incorrect and I have a lucky chance at avoiding a loss. Even though the trade is now in profit territory, holding until my original target is met is probably incorrect, since my entry wasn’t quite right. So unless price starts shooting off in my favor, I always remember to “kill the zombies” ASAP, and in that way, I can achieve the opposite of the cardinal sin, and I can turn a losing trade into a winner, even if it’s just a small one. Make sure to kill your zombies when they’ve come back from the dead, before they kill you!

Posted in psychology, strategy | Leave a comment

Expectancy Spreadsheet

Recently I developed an Excel sheet for calculating expectancy. There are several calculators online already, but I couldn’t find one that incorporated scratch trades into the formula–just ones using wins and losses only. Further, I couldn’t find any that included commissions in the expectancy formula. Instead of manually averaging my scratches with wins (or losses) and subtracting commissions, I decided to just make my own calculator.

UPDATE 11/26: I just added an “outlier” trade category, sweet!

A few notes:
-A “Scratch trade” in this particular sense is equivalent to a breakeven trade, i.e. when you have moved your stop to breakeven, or +1 or +2, etc, and price comes back and stops you out there.
-I am no longer trading forex–I have switched to futures (6E), so “win size” is in ticks and dollar amounts are multiplied by tick value ($12.50). To use this with forex, you can change the $12.50 multiplier to your particular pip value.
-Commissions are per round trip–and if you are trading forex, your “commission” is the average spread of your instrument.

I hope you will find this tool as useful as I do. What it makes obvious is how commissions really add up and impact your bottom line, and they are a serious consideration when calculating expectancy. If you are a frequent trader, you need to be calculating expectancy including commissions!

wb-Expectancy.xlsx

Please offer any feedback or calculation corrections, as I developed this fairly quickly and without a lot of testing.

Eventually I would like to code this to javascript to embed it on this site! (Any volunteers?)

Posted in strategy | 1 Comment