What High Probability Trading Really Means - The Ultimate Guide
The True Nature Of Trading
Traders who are new, or who are not finding the success that they would like, are often working under the burden of a few common misconceptions and errors. The first and most broad-ranging of these misunderstanding is probably responsible for the most widespread and the biggest trading problems.
So, let’s tackle that. I’ll make this as brief, simple and easy as I can, but it’s a big subject.
The Correct Application of Probability Will Decide Your Success or Failure in Trading
There are other important elements in trading but none of them really matter if you don’t get this right. Trading results are driven by probability. Trading actions must therefore also be based on probability. This applies to every trading system in the world. But the practical application of this is widely misunderstood. And it’s a hidden killer.
There’s some math here. It’s easy but essential. But don’t misunderstand the significance or the importance of acting on it. If you want to trade successfully, you must have this core information.
Ok. Let’s get into it.
Trading Is A Business -- How To Use The Numbers To Your Advantage
Trading is a business. And, like any other business, we need to keep records. And we need to know a few simple things that we must do with the numbers, to confirm that everything is on track.
The first crucial element to take on board is this. We must have a set of trading rules, formulated into a trading plan. Otherwise the numbers and statistics we can derive are all meaningless. In order for the numbers from trading results to be useful, those numbers must come from a set of trades which were all taken on the exact same basis. Only then can we expect to repeat the performance of the trades we have taken.
So we must start with trading rules and a trading plan. This article would become impossibly long if I include that information here, so it is on a page of its own.
The Necessity For Trading Rules And A Trading Plan
With trading based on rules, we can evaluate our trading system in two scenarios. We can look at how it performed in the past, and we can evaluate it in real time, as it goes forward into the future.
Evaluating The Past - Backtesting
Evaluating the past performance is known as backtesting.
We can backtest our system with a computer. Many trading platforms have backtest capabilities. We would have to write the system rules into computer code.
Or we can backtest our system manually. In which case, we would look back through charts and note where events trigger our trading rules.
Evaluating The Future - Real Time Trading Records
In terms of real time trading, we have two options. We can paper trade, also known as simulated trading, or sim trading, or we can trade in real time with real money.
Note that we do not accept somebody else's claims about the performance numbers of a trading system. That is not one of our options. You must always confirm for yourself how a trading system performs.
Your Trading Records
At a minimum, you would want to record the following details for each trade taken:
- Opening Date
- Opening Price
- Number Of Shares
- Initial Stop
- Reason For Entry
- Exit Date
- Exit Price
- Reason For Exit
- Trading System (if you have more than one).
You can put these elements in as columns in a simple spreadsheet
The Trading Data And What We Can Do With It
Let’s assume that we have a decent number of trades recorded. Let’s look at what we can and should do with the data.
First we should count the winning trades and count the losing trades.
Next, calculate the average profit for winning trades. Then calculate the average loss for losing trades.
There is a lot we can do with just these four numbers.
First, we will just calculate two more statistics
Dividing the number of winning trades by the total number of trades gives us our win rate.
The last statistic we will calculate is the most important number for the trading system. This number is the expectancy of the trading system. This single number defines the profitability of our system. If it is above 1, the system has been profitable.
The expectancy of your system is a combination of two things. The first is the number of trades you win. The second is your win:loss ratio -- how big the average winning trade is in relation to the average losing trade. Which we calculated above
Plug in our numbers from above, and we get
As a sidenote, an expectancy number of 2.10 is high. High is good, but not easy to achieve.
Personally, I have a spreadsheet where I enter the details of every trade I take. It updates these numbers automatically. So I can always see at a glance if my recent trades continue to be on track.
Expectancy can be used to measure past performance. Past performance cannot tell us that our trading system can be expected to make a profit in the future. But it can give us proof of concept and a benchmark. Expectancy can then be monitored going forward into the future. Showing us that the trading system performance remains acceptable, or nor.
So,what other uses do we have for these simple statistics?
The numbers give us some other important clues about the system we are trading. We already calculated winrate above - to recap, it’s simply the percentage of your trades which are winners.
The number we need to look at in conjunction with win rate is your win:loss ratio, which we will calculate now. This is the size of the average winning trade in relation to the size of the average losing trade.
The formula is:
It’s very useful to look at these two numbers together. WinRate and WinLoss ratio. Like this:
Get A Feel For What Your Trading Experience Will Be Like
I think of this as the 'shape' of a trading system. It gives me a feel for what my trading experience will be like.
For example, I see that I am expecting to win around 70% of my trades. I see that my winning trades are a little smaller than my losing trades. This is what you would expect from a system that has a high winrate. These numbers also imply a few other things, which we'll look at next.
Winrate can also give me information about winning and losing streaks. We'll look at this shortly. It is pretty essential to know what I might be in for in a losing streak.
In conjunction with WinLoss Ratio, I also get a feel for how many trades it might take to recover from a losing streak. These numbers also give me a feel for what I might experience in a winning streak.
Winrate is the part of a trading plan or 'setup' that traders most frequently focus on. It's usually the first question people ask about a trading system. And often the only question they ask.
But it's usually for the wrong reasons. In terms of how well our system will do, winrate is actually meaningless on its own. It can't tell us whether a trading system will make money, or how much. We have already said that our expectancy is a function of both winrate and win:loss ratio.
Despite the misconceptions, winrate has some important inmplications. So, we'll look at this next.
A Serious Error Frequently Made In Trading
Which brings us to the first error that just about all traders make. And one of the most destructive. Let's say we have a trading system with a 70% winrate.
If we have a 70% winrate, then our next trade has a 70% chance of being a winner, right? That's pretty good odds. What if I could get an 80% winrate? My next trade is even more likely to be a winner, right?
Hold your horses. No, absolutely not. Your system can have a 70% winrate, an 80% winrate, or a 30% winrate and your next trade is always 50:50. How can that be? Surely that can't be true... How can we get our heads around that?
Well it is true because of two things
Each trade is independent of the one before and the one after. This simply means that the trades have no memory. The trades don't 'know' that you just had two losing trades and now you're 'due' a winner.
Each trade is 50:50 because it has only two possible outcomes. You either have a winning trade or a losing trade (ignore the very few times when you could break even)
The Outcome Of The Next Trade Can Only Be One Of Two Things
So, the outcome of the next trade can only be one of two things. And that outcome is is not affected by how many winners or losers you may have had. So, this next trade must be 50:50.
You only see the effect of your true winrate over a sufficiently large number of trades. This is crucial to understand.
If you take 100 trades, you will be getting close to seeing your true winrate. eg with a 70% winrate, you might see 65-75 winning trades.
If you take 1000 trades, you will be really close to seeing your true winrate -- maybe within 1 or 2. With a 70% winrate, you might see 68-72 winning trades.
To help see why the next trade is 50:50, Consider this:
- we know that the next trade can only be a winner or loser
- it is impossible to know where the winning trades are located, amongst the next 100 trades you are taking.
So, again we see that this next trade can only be 50:50
The error made my so many traders is in assuming that the next trade has the same probability as your winrate. So there is a tendency to put on too large a position. When the trade is a loser, the trader suffers a loss which is too big for his account.
If you truly take on board that the next trade is 50:50, this will never happen to you. You’ll never be tempted to trade too big. And, if a loss is equally likely to a win, you’ll find yourself focusing on the risk, and making sure it is not too large for your account.
A Second Common Error That Prevents Traders From Succeeding
There is another important property associated with winrate. And that is the size of winning and losing streaks. This is closely related to The 50:50 probability of the next trade that we just discussed above. It also gives yet another perspective on why that is the case.
If we had a trading system with a 50:50 winrate, it would be equivalent to flipping a coin. Coin flipping is a good substitute, because we inherently understand certain things about it.
We instinctively understand that a coin flip is a 50:50 proposition. From our work above, we now understand why. The coin toss is 50:50 because it only has two outcomes, heads or tails, and each coinflip is independent of the others. The coin has no memory to tell it that heads or tails is 'overdue'.
A String Of Wins - Or Losses
But we also know that we can get a number of heads in a row, or a string of tails in a row.
How big can these strings of heads or tails — or wins or losses — be? It's kind of crucial to understand this. When we relate this back to our trading system, it is our money that is on the line, and we need to know what to expect. Also, if we know what to expect, we can sit through the pain of a string of losses. We know that the true properties of the system will eventually re-assert themselves.
So, with a 50:50 coin flip, assume that Heads is a win and Tails is a loss. How many losing flips, or 'trades', might we see in a row?
I'll save you the work. I wrote some code to do the heavy lifting. What we are seeing below is the largest run of heads, and the largest run of tails from a number of coinflips
Here are the results from flipping a 50:50 coin 1000 times.
Surprisingly, even with a fair coin which has a 50:50 probability of coming up Heads/Tails, we have some unexpected results. We see runs of 8 Heads in a row and 7 Tails in a row.
Winning (And Losing) Streaks From 1000 Trades
And here are the results from performing that same 1000-flip test 10 times in a row
If we flip the coin 1,000 times, we see even larger runs. It's a fair coin with a 50:50 probability of coming up Heads/Tails. But we commonly see runs of 10, 11 or even 12 Heads or Tails.
What Does This Mean For Our Trading System?
It means that, with our 50:50 system, we know that it is quite common to see double-digit strings of both wins and losses. You might be thinking that nobody would trade a system with 50% wins and 50% losses, but you would be wrong. If such a system had average wins that were larger than average losses, it would be a money-making system.
And, in fact, many trend following trading systems have a win rate of less than 50%. They make their money by having winning trades that are much bigger than losing trades. But these systems also suffer very large drawdowns. Now you know why. With a win rate below 50%, The strings of losing trades can be very large. This results in a large drawdown, which is difficult to sit through for most traders.
And what do you think is one of the most common reasons why traders give up on a trading system? Encountering a drawdown which is beyond their comfort zone. And this is important to understand. If a trader quits in the middle of a large drawdown, he is ensuring a potentially large loss. If he fails to understand what is happening here, he is in danger. If he moves on to another trading and repeats his error, he is on a certain path to lose most or all of his money.
And this is exactly what happens with many many beginning traders. They hop from one system to the next. Always quitting because of a drawdown they could not withstand. Always cementing in their losses, and ultimately failing because of exactly that.
These traders quit because they think ‘the system failed’. They didn’t realize the string of losing trades that they might expect. Even if the system was working properly.
In other words, the system was probably working exactly as it should have been. Yet they have blamed their loss on system failure. But we can see that the loss was actually caused by their lack of knowledge and false expectations.
Let’s look at the results of the same coin-flip test, but with a coin biased such that it comes up Heads 70% of the time.
In other words, this is the equivalent of a 70% winrate trading system.
Viewing Probability Incorrectly In Your Trading
Remember, most people are expecting the next trade to have a 70% probabilty of being a winner. If that's what you expect, how do you feel when you see a winning streak of 7 trades, as occurred above?
The results that occur are governed by the characteristics of the trading system. You can't change what will happen unless you change the plan. But it is obviously much better if you truly know what to expect, and you can prepare in advance for it.
You might even decide not to trade a system at all, if you properly understand the results that are possible.
The Difference Between Trading Success And Failure
Understanding these simple numbers can allow you to make an informed decision. You can decide in advance whether a prospective trading system is right for you. And save yourself the time, pain and money lost by finding out the hard way.
Looking again at the table above, consider the difference in how you feel in these two of the possible scenarios:
Imagine how you would feel encountering a string of 22 winning trades. Then consider how you might feel with a string of 7 losing trades. If the string of 22 winning trades came first, you may possibly withstand the string of 7 losses. But what if the string of 7 losses comes first?
Whatever happened, however it turned out, you can now see that you are much better off to be prepared in advance. It has to be better to know in advance what worst case scenario you might expect.
What Do We Learn From 40,000 Trades?
Here's another 4 instances of 10 x 1000-flip runs of a 70% biased coin. Here we see that while the run of 7 losses we saw above is possible, it's not very likely. In these 40 x 1000-flip runs, we see that 6 is the highest number of consecutive losses.
We have just seen that a test of 40,000 trades resulted in a losing streak of no larger than six trades. But we also saw earlier that a streak of seven losing trades is possible. We see that different outcomes are possible from the same trading system. And we instinctively realize that, actually, anything is possible.
All those 40 runs of 1000 trades were from a system with the same 70 winrate. And yet all of the results were different. Why is this?
The answer is something that is very useful to understand about the nature of trading.
We Don't Know The Future But We Know What To Expect!
No analysis or projection tells you exactly what is going to happen in the future. The future trade-by-trade results from a trading system are always different. And they are always unknowable.
There is a perfectly rational explanation. We can know what winrate to expect from a trading system. But we never know where or when the winning trades occur within the total number of trades we take.
Eg we take 100 trades, we know to expect about 70 winners. But we don’t know where within those 100 trades the winners lie. We know that the winners and losers can bunch up into streaks of unknown length. The winners and losers are not evenly distributed.
And they are distributed differently in every set of trades we take. Which brings us right back to the next trade always being 50:50.
In this section, we have learned some important facts about the true nature of trading. I stress that they are facts, because much of this is counter-intuitive.
We can and will see strings of winning trades and losing trades
The size of these winning runs and losing runs is related to winrate
The outcome of the next trade is not is related to winrate
The outcome of the next trade is 50:50 in all trading systems
So we can boil all this information down to four essential nuggets.
But they are four very important points which will completely change the way you approach trading, if you fully take them on board and take action on them. It took all the detail to be able to get our heads around why these important facts about trading are true.
All successful traders know these facts about probability. Now you do, too. All successful traders incorporate these facts into their trading. But that part is up to you.
You can check out all our data and trading tools FREE by clicking this button