All Trading Is Predicting

A lot of people online, and I don’t mean just the “gurus,” get their jimmies rustled when you imply that trading is predicting.

They will say stuff like “I don’t predict, I react.”

Or, “I don’t predict, I trade based on statistical analysis of the markets.”

Or “I am playing the odds.”

Let’s cut the nonsense, it’s all the same thing.

What is a prediction?  A prediction is when you think that one thing is more likely to happen than another thing.

Why do you take a long position?  You do so because you think price is going to go up.  After all, if you thought price was going to go down, why would you take a long position?  You wouldn’t.

It doesn’t matter if you take a long position because you are “reacting” to something that makes you think price is more likely to rise, or because your “statistical analysis of the markets” suggests that in a certain scenario price is more likely to rise, you are still predicting that price is going to rise, and as a result you take a long position.

People don’t like to use the word “predict” because it ties into their ego.  If they get a trade wrong, their ego takes a blow.  Their prediction didn’t come true.  That hurts.

But when they’re “just playing the odds” or whatever, it’s ok to have some losing trades.

All trading is predicting.  Regardless of your reason for entering with a long position, whether it’s because you see a particular pattern, or you did statistical analysis, or your indicator gave a signal, or because you are “reacting” to a certain situation which has lead you to believe that price is more likely to go up than to go down, what do all of those situations have in common?

They all involve you thinking price is more likely to go up than to go down.

They all involve you predicting that price is more likely to go up than to go down.

Here is a flow chart that explains the process for every single non-random long entry (random entries are not predictions; see the paragraph about random entries below):

long flow chart

(click to enlarge)
This is how every non-random long entry is made by every directional trader.

As you know, I frequently state that I cannot predict price direction, and that is why I trade the way I do.  But even when I trade I am still predicting!  I am predicting that the S&P will eventually go back up.  I don’t know when, and I don’t know how far up it will go, but I am predicting that it will eventually go back up, and that’s why I average into my positions.

If I did not believe that the S&P was eventually going to go back up, I would not take a long position.

All trading is predicting.  The reasons don’t matter.  When you take a long position you are predicting that price is going to go up (after all, if you didn’t think price was going to go up, why would you go long?).  When you take a short position you are predicting that price is going to go down (after all, if you didn’t think price was going to go down, why would you go short?).

Even if you use statistics you are still predicting.  If you are going to flip a fair coin 100 times I will predict that you will get heads 50 times.  If you are going to roll a six-sided die 60 times, I predict that you will roll a five 10 times.

If my statistical analysis tells me price is more likely to go up than to go down, I predict it will go up.  If my statistical analysis tells me price is more likely to go down than to go up, I predict it will go down.

As for “reacting,” that’s the same thing, too.  Trading “gurus” love to say they “react” rather than “predict,” as if you can’t be wrong with a “reaction.”  But even that is all the same.

We’ve already established what a prediction is: a prediction is when you feel that one thing is more likely to happen than another thing.

When a trader “reacts,” what is happening?  He sees a situation and feels that because of that situation, price is more likely to do one thing than to do something else.  In other words, he is predicting that price is going to do something, and taking the appropriate position.

A “reaction” is a prediction of what will happen now.  All predictions are reactions to the current situation.

The only way trading is not predicting is if your entries and your direction are random.  If you flip a coin and heads means go long and tails means go short, that is the only situation in which trading is not predicting because you are entering randomly rather than because you think one outcome is more likely than another.

All (non-random) trading is predicting.

Hopefully your jimmies are unrustled now.Rustled Jimmies

Weekly Silver Fibonacci

The people who think Fibonacci levels have magic powers might be interested in seeing this weekly silver chart that shows that price is just about to hit the 0.764 level.  That’s not even one of the main Fibonacci levels.  Also notice how price pretty much ignored all the other levels on the way down.

It’s funny how Fibonacci numbers only seem to work sometimes and no one can ever predict ahead of time when that will be.

silver fibonacci

Higher High? Higher Low? Lower Low? Lower High?

Some people will tell you that trends “move in waves.”  They will say, for example, that up trends are made of a series of higher highs and higher lows, like this:


And a downtrend is just the opposite (a series of lower lows and lower highs).


Now that’s cool and it makes sense, but there are two other things price can do: make double tops and double bottoms (when a high is the same as a previous high and when a low is the same as a previous low):

double tops and bottoms

As far as this kind of price action is concerned, those are all the possibilities you can have.

Conventional thinking will tell us that higher lows, for instance, are bullish because they mean price was unable to break the previous low before going back up.

Double tops are likely to be bearish because it means price was unable to break above the previous high (and you could also look at it as price bouncing off of previous resistance, which is bearish).

Double bottoms are likely to be bullish because it means price was unable to go beneath the previous low (and you could also look at it as price bouncing off of previous support, which is bullish).

Now, is this enough off of which to build a trading system?

There are a few problems with this.  Let’s say you were going to trade an uptrend that was formed by HHs and HLs.  Price starts going up and forms a high, and then it starts retracing.

uptrend forming 1

You are going to wait for the HL and then go long, but there is a problem: how do you know when price is done retracing and is going to go back up?  How do you know that retrace isn’t going to keep going down and form a lower low?

uptrend forming 2

Did you enter yet?

uptrend forming 3

Did you enter yet?

uptrend forming 4

Did you enter yet?

uptrend forming 5

Oh no, price formed a lower low!  Looks like we’re not in an uptrend anymore.

Now we should be looking for a lower high to confirm the downtrend, right?

uptrend forming 6

A LL followed by a HH?!  How does that happen?!

People will come up with all sorts of rules, like:

– don’t enter long until a high, a HL, and another HH (this is supposed to “confirm” the uptrend)

– don’t enter long after more than 3 waves (as if after making 3 HHs, price won’t make another HH)

All of these rules are just things people make up because they don’t understand the random price of the market.

Sometimes price trends and makes beautiful HHs, HLs, and HHs.  If you can figure out when to enter (because the HL can form at any point), you can make money as long as you get out before price goes against you.  Be careful adding to winning trades because as soon as that retracement goes a bit too far, your winning trade can turn into a losing trade.

Sometimes price makes LL, HH, LL.

Sometimes price makes HL, LH, HL.

Price can do whatever it wants and doesn’t care what it has done in the past.

You will see people chime in on these kinds of discussions and say “well obviously you just don’t understand it” subtly implying that they do understand it, yet you will never see them make real time calls.  I wonder why?

Price can go wherever it wants, and unless you know ahead of time that it’s going to make a LH or whatever, there’s no way to know that a HL will follow a HH.  Here is a list of possible formations that can follow a HH:

– HL

– DB

– LL

(obviously a LH can’t follow a HH because a LH is formed when price is rising and if you’re already at an HH you can’t have a LH without some sort of low coming first).

Here’s a chart showing what can follow what.  Start with the vertical column and go across to see if a certain formation can follow it.

high low table

This is a very active topic and I will likely revisit it in the future.

I encourage your comments and discussion about how to use HH/HL/LL/LH/DB/DT in your trading.

How I Trade As If Price Is Random – Part 3

(continued from part 1 and part 2)

My original method, the one used in my journal thread, was based on Fibonacci retracements. Keep reading before you flame me: I don’t think Fibonaccis are special, and I joke about people who think they have magic powers.I do not believe that a 38.2% retracement is really any fundamentally different than 38.1% or 38.3%, or 43.6%, or 19% or 39.4%, or any other number. Literally any number will work just as well because, say it with me, price is random. Do you think price goes “hey, this is a 38.2% retracement! Time to go the other way now!” No. And do you think the market makers do that?

I do not believe Fibonacci numbers are special. I know that some people are pretty serious about using Fibonacci number tick or volume charts as if the Fibonacci number is somehow magic and will work better than any other number. That’s all nonsense.

When I first shared my method elsewhere a few years ago, the discussion quickly turned into an argument between people who think that because Fibonacci numbers may sometimes be found in nature, they are therefore applicable to trading, and the people who think that’s a bunch of crazy talk.

I decided to share this method because of a few conversations I had been having with people who were paying a monthly fee to some “guru” to learn how to trade with Fibonaccis. None of them were making any money (other than the “guru” who was collecting monthly subscriptions), and the thing I noticed is that these “gurus” were always extremely vague.

They would post after the fact charts where price happened to bounce off of a Fibonacci level and say “see?! Fibonacci retracements work!”

The gullible students would say “wow u r so smart! Here pleez take my money to teach me!”

But the people who had more accurate BS detectors see that that is all nonsense. Without knowing ahead of time if price was going to bounce off a Fibonacci level, and if so, which one, such a method is useless.

I originally used two methods: one for daytrading the ES, and one for long term trading SPY (or SSO). The daytrading system is much crazier and I will talk about it later. Since I don’t daytrade regularly anymore, let’s just talk about the longer term system.

This is a long term system so we’re using daily charts.

The general idea is that since price is random, we know that it goes down and up, but we don’t know when it’s going to either, nor for how long. We do know that it can’t go to negative, and we have a pretty good chance that the S&P will not go to 0.

Do not use margin. You are going to be averaging down quite often. Using margin is a good way to blow your account when SPY goes one penny further than you can afford. If you do not use margin you cannot blow your account.

The initial profit target will be the previous high. Depending on how far your position goes against you, you may wish to change the target in the future (for example, if you start buying SPY at 140, and it eventually drops to 90, you might not still want to have to wait for it to get back up to 140 before you close your position).

Step 1: Wait for SPY to make a pullback. You can use Fibonacci retracements if you want. I did for the sake of proving the point that Fibonaccis can be traded profitably but it really doesn’t matter. You can use 25% increments, you can use 30%, or 42%, or 51% or anything you want. There are no magic numbers. When price has pulled back to your level, buy a small amount.

Step 2: Buy more SPY when it goes against you another predetermined amount (perhaps the second Fib retracement, perhaps a random percentage, etc).

Step 3-x: Continue to buy more SPY when it continues to go against you. Again, you can use Fibonaccis, or you can divide its current price into increments (such as if SPY is at $140 and you buy every time it drops $10), or into percentages (such as if you buy every time it drops a certain percentage), etc. Remember that each add is going to be bigger than the previous. This is averaging down. Remember that you are not using margin. Begin your first add with a small enough size that you can continue to buy more as it goes against you. If SPY is $145 and you keep buying all the way down to $100 and then run out of money and then SPY goes down to $80, you did it wrong.

Remember, price is random so you have no idea how far it’s going to go down.

Some of the trades will be awfully slow and boring. I currently have a trade in QLD that has been open since November of 2012 and has barely gone anywhere. I didn’t know QLD was going to chop for a few months when I bought it. Patience, remember. Not every trade will end up in a huge winner. Some of them will just be small winners.

Ok I’ve been spending a few hours writing this now and need to take a break. There is much more to discuss, including:

– hedging a SPY position with SH to make money when your SPY position is drawing down
– uptrends and random entries
– using covered calls in slow boring uptrending markets

And some more rules to be discussed, such as why I never add to a winning position.

And now to preempt some of the inevitable questions and criticisms:

1) “Dude you are so dumb, this isn’t trading.”

Well, I’m making money buy buying and selling stocks, so yeah, it is trading. And I’ve been consistently doing so for years with every trading posted in real time, so… yeah.

2) “This system sucks and doesn’t make money in big up trends. That’s when all the big money is made. I know because some “guru” told me so. You would know that if you were a real trader. Trend following for lyfe!!1!1!!!!”

Yeah, I don’t make much in big up trends unless I happen to have a big position from a previous downtrend. That doesn’t always happen. But I also don’t have big losses following downtrends the way most people do. And when price goes up and down I do pretty well.

3) “You idiot, you are basically just throwing a lot of money at the market and when it eventually rebounds, you make money.”

Yup. Since I cannot predict price, this is how I have to do it. I view the markets as a mathmatical random sequence rather than whatever you view them as. If you were going to make money from a random sequence, how would you do it? Averaging down, that’s how.

Look, this is not everyone’s holy grail. Sometimes it works very well. Sometimes it doesn’t produce big winners.

If you can predict price, keep doing what you are doing. If I could predict price I would just enter in the correct direction with my entire account on each trade. There would be no reason to start small and add more as price goes against me.

I fully admit that this method is inferior to price prediction methods. But I’ve never seen anyone who could successfully predict price, and until I am able to do so myself, I will continue to trade this way.

To give you an example of a complete trade that lasted a whlie, on October 27, 2011, I closed out a trade that had been open for a while. This was the previously mentioned trade that at one point was drawn down over $60,000. On October 31, 2011, the dividend was paid and the final numbers looked like this:

SPY: $31,115.00 (closed)
hedge: -$3,281 (closed)

$27,834.00 (SPY gain + open hedge loss)
$1,223.60 (realized hedge gain)
$2,931.69 (realized hedge gain)
$1,129.72 (dividend)
$3,562.22 (dividend)
$5,221.55 (realized hedge gain)
$41,902.78 (total for trade)

I had a SPY position which I had been continually adding to, several SH positions that had been closed and reopened, two dividend payments, and a currently open SH position that was closed at a loss.

That’s how I trade as if price is random.

How I Trade As If Price Is Random – Part 2

(continued from part 1)

Let’s talk account size for a bit. This is not to brag, but just to paint a realistic picture of where I’m at. This style of trading is not going to be feasible if you have a $5,000 account.

Ever since college I have been very frugal and tried to live below my means. This isn’t meant to be a financial lesson, but just because your salary increases doesn’t mean your expenses have to increase. I’m sure you’ve all read “The Millionaire Next Door” and that type of books (if not, read it. Cliffs notes: rich people are frugal, poor people are flashy).

This is especially true in trading. If you have a normal salaried job, you know that every week you will get paid $500 or $1,000 or however much you make. Although your salary may be limited, it still allows you to plan for future expenses. If you own your own business (trading is included in this) then this is all uncertain. You may make $10,000 one week but that doesn’t mean you won’t make $0 the next week. Or -$10,000.

Being frugal helps ensure your survival.

Lesson over, back to trading.

Let me back up even further and say that the reason my account is where it is today is because during the “recession” of 2008, I averaged down heavily into weighted index funds (QLD, SSO). This is already breaking two “rules” that traders love to quote: 1. don’t average down, 2. don’t use weighted ETFs long term.

Everyone else was freaking out and selling. “Oh noes, the economy is collapsing!”

No it’s not.

It hit me that this could be a great buying opportunity. My first thought was to buy as much SPY as I could afford, but then I learned about the weighted ETFs which were not only cheaper per share, but also double weighted, so for example, if the S&P 500 goes up 1%, SSO goes up 2%.

I picked the indexes because I had no idea what would happen to individual stocks but I was pretty confident that the indexes would not go to zero, and if they did, I had bigger problems than blowing my account.

While I watched everyone on the forums talk about the collapse and how they were selling, I just continued buying more every time it dropped a certain %.

“Be greedy when others are fearful.”

I can’t predict price direction, but I was pretty sure those funds would go back up.

So, remember when I said I regularly break commonly-accepted trading rules?

What is it, 95% of traders lose? And they all probably follow those rules.

– Never add to a losing position
– Never let a winner turn into a loser
– Never risk more than 2% of your account
– The trend is your friend

Let’s talk about those:

– I regularly add to losing positions. I do it within certain boundaries, never with margin, and never with risk of blowing my account.

– My winners often turn into losers. I have no idea if price is going up or down. Sometimes they go against me.

– My drawdown sometimes goes well past 2% of my account.

– Every trade I make is counter-trend.

“The trend” is a bunch of nonsense and is a great topic for another post. Let me summarize it like this:

– price can reverse at any point and the “trend” only exists in hindsight. Just because whatever trend identification method you use (MA slope, higher highs/higher lows, whatever) happens to say “hey, we are now in a trend whereas one tick prior to this we were not in a trend” doesn’t mean price is going to keep going in that same direction.

(continue reading in part 3)

How I Trade As If Price Is Random – Part 1

I also posted these on the excellent forum

Let me start by saying this style of trading is not for everyone. I expect a lot of criticism and people to tell me I’m doing it wrong. I pretty much break every trading “rule” in every trade. More on that later.

Let me also say I’m not selling anything nor am I a vendor, nor do I accept “donations.” This is all free for the sake of discussion and creative thinking. I have a website and a Twitter account but there’s nothing for sale on any of those, either, and all I really do on them is call trades and write about how to avoid trading scams. No one is going to steal my edge and prevent me from making money unless they somehow manage to make the market go up forever without ever retracing in which case my retirement accounts will thank them.

Let me start by talking about my “edge.” I hate that word because it gets thrown around all the time on the forums by people who don’t always know what it means (the same people who calculate risk/reward after the fact).

My edge consists of:

Account size – I know there are stories of people starting with a $10,000 account or whatever and building it up to $1M. I’m sure that’s possible if you can predict direction, but I can’t, and this is not one of those stories.

Discipline – I don’t have trouble following rules. I’m also patient. Some of you may have been following my journal thread for the last few years on another big trading forum where I’ve posted live calls for every trade and remember a point when I was sitting on around -$60,000 of drawdown and everyone was telling me I was doing it wrong and I should close the trade before it gets any worse and etc. I didn’t care. I sat through big drawdown for a while, collecting dividends and profits from hedging the other side (more on that later) in the process, and ended up closing the trade for over $40,000 in profit. This was all posted in real time with daily updates, and that forum doesn’t let you edit old posts so I couldn’t change anything if I wanted.

Good thing I didn’t listen to those people who told me I was doing it wrong.

Patience – I know that there are some times when I won’t have any trades. Sometimes the market just slowly trends upward over time and I don’t make very much money. I don’t care. Focus on the big picture.

Money management – I never use margin. I plan entries and exits ahead of time. I know how to hedge the other side. I never use margin. Also, I never use margin.

Knowing what I don’t know – I cannot predict price. I’ve studied every indicator, I’ve reverse engineered some of the popular ones, I’ve created my own. All useless. I’ve studied price action. I’ve studied volume analysis. I am still studying these things in my spare time but currently am still unable to predict direction, but maybe one day I will.

But until that time, I’ll continue to trade the way I do, having no idea where price is going tomorrow, only knowing that it’s going to go up and down.

(continued in the part 2 and part 3)

Is Price Movement Actually Random?

Take a look at this chart.  I’ve labeled some of the technical patterns that you might identify:

chart with patterns labeled

You can see a double bottom as price tests support and then rallies up to a previous level of resistance.

You can see price bounce off that previous resistance and then continue upward, and then come down and retest that same level, bounce off as it flips to support, and then continue upward again.

You can see a period of consolidation after a new high where it just kind of chops around for a bit, then breaks through to the bottom, bounces off that new support level, goes back up and is currently forming a triple top at a new resistance level.

A trader armed with this knowledge who can identify these patterns should be able to make some money, right?

Well, I hate to tell you this, but this is just a random chart that was generated by 1,000 coin flips.  Heads is +1, tails is -1.  The only change I made was to move it over to the left a little bit so it didn’t start from zero, since no stock ever starts from zero.

1000 coin flips

You can try it yourself here.  Obviously some of the results won’t look like stock charts, especially the ones that go negative, but run it a few times and you should get some good ones.

Are you being fooled by randomness?

Why I Don’t Add To Winning Trades

As you know, I trade by adding to my position when price goes against me.

Most people say you’re not supposed to do that.  Here is where I point out that when I add to my position, I do it with specific risk tolerances rather than haphazardly adding as long as price keeps going down, which is a good way to lose all your money.

Adding to a position brings the average cost closer to the most recent add.  For example, if I go long ES at 1,000 with 1 contract and then price decreases to 995 and I add another contract, my average price is now 997.50, which is closer to the most recent add.  The more you add to an open position, the more price moves closer to the most recent add.  For example, if I have one ES contract long at 1,000 and price decreases to 995 and I add 2, my average cost is now 996.66, which is closer to the most recent add than it was when I only added 1.

When you add to a losing long position, your average cost decreases which means it only takes a smaller movement for your trade to turn into a winner.

But when you add to a winning position, that means it takes a smaller move against you for your position to turn into a loser.

Let’s assume price is trending upward and is making a few pullbacks along the way, making HHs (higher highs) and HLs (higher lows).  If you were long at the start of the trend and then added to your position at each pullback, it would look like this:

adding to winners 1

That is an ideal situation, but let’s assume for now that that’s what happens.

The green lines are buys and the green circle is selling your entire position.

If you could trade like this, you’d have a nice winning trade.

But let’s look at what is happening.  When you first open the position, your average cost is right at the first entry point.  No problem.

When you add the second contract, your average cost has now risen.

The red dash shows where your average cost is after each add (this assumes you start with one contract and add one contract at each buy level):

adding to winners 2

Now that’s fine and all if price keeps going up, but what if it doesn’t?

How do you know that each of those pullbacks is just going to be a LH and that price isn’t going to keep going down?  When you’re looking at the hard right edge of the chart, can you tell?

adding to winners 3

If price doesn’t keep going up and instead keeps going down, your previous winning trade has just turned into a losing trade:

adding to winners 4

In that last situation, if you hadn’t added to your winning trade, it would still be a profitable trade as price is still above your initial entry point (and average cost).

Adding to winning trades is only profitable if price keeps going in your direction, and a small movement against you can turn your winner into a loser.

I’m not saying it’s wrong.  If you can make it work for you then you should keep doing it.  I’m just explaining why I don’t add to winners.  Since I cannot predict direction, I have no idea if that pullback is going to be a pullback as part of a bigger trend or if it’s the beginning of a bigger movement in the opposite direction which would turn my winning trade into a losing trade.  I have also never seen anyone who can tell in real time if a pullback is a LH or the beginning of a downtrend.  I’ve seen a lot of gurus post after the fact charts where everything looks all perfect, but I’ve never seen a single one make a real time call.  In fact, this is a good way to tell if someone who claims to be a trading instructor is scamming you or not.  If they only show you after the fact charts and cannot demonstrate in real time that they can trade profitably, then they are probably scamming you.  If they show a chart and say “add here on the pullbacks” without explaining how they knew they were just pullbacks and not the beginning of trends in the opposite direction, they are scamming you.

Refer back to this image:

adding to winners 3

When you’re in that position, can your “guru” trade instructor teach you how to tell if that’s going to be a pullback (in which case you should buy) or if it’s the beginning of a down trend (in which case you should not buy)?

Sure, he can post after the fact charts like the first one I posted in this entry, but after the fact charts don’t help you.

July 29, 2011 – Why I Closed Half The Hedge

Mostly because I felt like it. I cannot predict where price is going to go so I never know when the best time is to close it. Back on 6/15 the hedge was up around $5,379. I should’ve closed it then. But I didn’t know if it was going to keep going up or down or what. Plus, I moved and we had those big up days in a row and my SPY position was way up in a matter of days. There’s no way of knowing how big of swings price is going to make.

I’d rather close it for a bit of profit than for a loss.

I mean, it really doesn’t matter, though. I will end this trade net profitable no matter where price goes and no matter when I close the hedge, so any positive hedge profit is extra bottom line profit in the end.

I’m thinking actually maybe I should make my next buy point at the 100 line of where I originally drew the first fib lines. I’d be buying some soon, and it would lower my average close sooner than waiting to see if price hits $124.10 where my next buy line is. And then I could potentially lower the profit target cuz who knows when price is going to get back up to $137 where my current sell area is? Or maybe I should just ask one of those people who claims to be able to predict price.


If I were to buy 1,400 @ $128.50, that would put my average cost at around $130.86. So then I could make the target profit 133 or so for a net of $6,848 from the SPY, plus the $1,223 from the hedge, plus whatever is left of the hedge, plus the $1,130 dividend.

Of course, if I do that, then SPY will probably go to 137 the following day. Cuz that’s how it works

Why Hedge Instead of Just Reducing Position Size?

So I can cash out in both directions. I can’t predict price.

Right now I’m just in a less than optimal place in that I’m still in the short position while the long position is in the money. I should’ve closed the short position when I was up around $5,000, but I ended up moving and being away from my computer for a few days while we had a rally. lol.

Mentally, I’m not treating it as a directional bet. I’m treating it as two separate things.

Also, I don’t usually enter both positions at the same time, which would make adjusting position size more logical (eg. instead of going long 2 and short 1, just go long 1).

I understand your point, however.

And since I can’t predict direction I have no way of knowing where price is going to go, or for how long. The hedge is basically a “bonus,” whether it’s an extra $100 or $5,000. even if I get it wrong and never close out the hedge for a profit (because price is random), I’m still net profitable for the trade.

I get paid more SPY dividend this way, too, because I’m holding larger SPY positions.