Hi YXI friends,

The difference between a winning trader and a losing trader is often not their analytical capabilities, but their abilities to consistently execute trades with positive expected value (EV).

Today, we are going to discuss the art and science of making these trades.

Through a detailed breakdown of a live Bitcoin trade, we’ll explore how to calculate reward-to-risk ratios, set effective stop-losses, and size our positions for maximum efficiency.

Moreover, we’ll address the common psychological traps that lead traders to hold onto losing positions and exit winning trades too early - and how we can avoid them.

Let’s dive in!

DISCLAIMER: This newsletter is strictly educational. Any information or analysis in this note is not an offer to sell or the solicitation of an offer to buy any securities. Nothing in this note is intended to be investment advice and nor should it be relied upon to make investment decisions. Any opinions, analyses, or probabilities expressed in this note are those of the author as of the note's date of publication and are subject to change without notice.

1. Phasing Out Luck: The Key to Consistent Trading Success

At the end of Part 1, I highlighted that we are mainly playing two components of the winning formula: the odds of success and the right reward to risk (aka risk-reward).

Macro analysis, fundamental analysis, and technical analysis are all just tools we use to nudge both components in our favour.

The trade execution can often matter even more than the analysis itself. Good trade execution requires immense discipline over our greed and fear.

There are 3 main components of a trade setup: 1) the entry, 2) the take-profit, and 3) stop loss. One may prefer a trailing stop-loss to ride a winning trade as far as possible, but there should always be an exit strategy before we enter a trade.

We want the right entry, with the right risk-reward, on a consistent basis.

Over the long run, we want to phase out “luck”. If we are placed in 100 different markets, we want to succeed in not just the one we are currently playing, but the other 99 as well.

2. Understanding Expected Value (EV)

Feel free to skip ahead if you are well versed in Expected Values and Risk-reward.

Coin toss: an illustration

Heads pay $1, Tails pay $1

Let’s use a simple example of a fair coin toss to start. The chances of Heads versus Tails are 50:50. If we win $1 every time it’s Heads but lose $1 every time it’s Tails, the expected value of guessing Heads is 0. In the long run we just breakeven every time we guess Heads.

Heads pay $2, Tails pay $1

However, if we win $2 every time we guess Heads and are right, and lose $1 every time we are wrong, the expected value becomes $2 × ½ + ($-1) x ½ = $0.5. Each win more than covers the loss. This is suddenly a favourable game and we should want to play it infinite number of times, if the coin remains fair.

(If you think about it, the stock market is actually such a game in the long-run, but let’s not diverge.)

In fact, even if the odds of Heads drops to 40%, we still have a positive expected value of $0.2. Smaller than before, but above breakeven. It’s only when the coin shows Heads less than 33.3% of the time, we get a negative expected value if we keep guessing Heads.

Heads pay $3, Tails pay $1

Now we shift the payout of Heads being right to $3 (red line). As long as Heads show up more than 25% of the time, we are printing money.

Heads pay $0.8, Tails pay $1

However, if we just move the payout of Heads from $1 (or evens) to $0.8 (yellow line above), the EV significantly drops. We need Heads to happen more than 56% of the time for us to make a small profit.

(You might think - who’s going to play this game or not guess tails? I will later discuss how the scenario manifests in real life.)

After 1000 tosses, the contrast in the expected values is huge.

Where the Heads of a $0.8 win just breaks even, a $2 win is seeing $680 of profit and a $3-win is printing $1240, nearly doubling the $2-win.

3. Applying the EV Framework to a Live Bitcoin Trade

Through macro, fundamental, and price analysis, I usually look for setups that I judge to have better than even odds (50%) of success. But it’s not a requirement - I will explain why below.

Of course, the true underlying probability is far more uncertain and constantly evolving, but I try to get as much confidence as possible using all the available tools. I am aware of the possibility of being utterly wrong. That is why we manage risks.

Let’s look at a recent trade set up I’ve flagged in Bitcoin.

Bitcoin Daily Chart (expand here)

I have been tracking Bitcoin very closely after the Wave (3) high, and I believe the August 5 crash provided us with a likely Wave (4) low. However, I didn’t rush to buy the dip, and waited for a 1-2 set up on the hourly chart.

Bitcoin Hourly Chart (expand here)

Bitcoin formed 5 waves up from August 5 to form a Wave 1 at $62.7k. It spent the next two weeks retracing, and eventually got down to $56k. Because I tracked the move very closely with Elliott Wave and the Fibonacci levels, I grew confident that wave 2 was near completion at this point.

I flagged the trade entry at $57.7k on August 15, with a target of $92k, which completes the larger degree wave 5, and a stop loss of $49k (the larger degree wave 4 low mentioned earlier).

Bitcoin Daily Chart with trade setup (expand here)

If we win, we make 59%. If we stop out, we lose 15%. This is roughly a reward-to-risk ratio of 4.

Now, it may looks smart because the trade is already 11% in the money, but as seen in the previous bounces, we can never know for sure if this time will be a real break out. I am prepared to be wrong, which is why we have a stop loss.

While I think there’s a better than 50% chance of the trade working out (given my holistic analysis), I actually only need to win more than 20% of the time to be profitable in the long run. The higher the reward to risk, the more buffer we have for being wrong.

4. The Importance of Stop Losses and Position Sizing

Not all positions are equal. If I simultaneously like 4 trade ideas, I don’t just put 4 equal portions of capital into them. Intuitively, one may want to put highest amount of capital into ones with highest convictions.

But I have easier, more mathematical methods of setting position sizes, focused on limiting the downside.

The general rule of thumb is that a tighter stop loss affords a bigger trade position.

a) Have a fixed Stop Loss USD value

Step 1: When I place a trade, I don’t want to lose more than 1% of my capital. If my portfolio is $100,000, it means I don’t want to risk more than $1,000 on a single trade. I can afford being wrong a lot of times this way.

Step 2: Now I’ve established my loss tolerance, I can marry this $1,000 loss to my percentage stop-loss in a trade. In the Bitcoin example above, $1,000 represents 15% of $6,667 ($1,000 divided by 15%).

$6,667 is my position size. Now, my upside from the Bitcoin trade is $3,960, while my downside is limited to $1,000.

However, if I have a different setup with a 10% stop loss instead, I could increase my position size to $10,000. Being wrong would still risk me just $1,000 or 1% of my portfolio.

b) Consider using 2 standard deviations (Z-score) as a minimum Stop Loss.

Sometime we can be tempted to set a stop loss very close to entry to get a high theoretical reward vs. risk. For example, a tight stop at $56k in Bitcoin to ensure we quickly move on if the price pierced through support. This is a stop loss of just 3% from our entry.

The practical challenge is that Bitcoin is a volatile asset. In the past 1 year, Bitcoin’s daily moves had a mean of +0.2% with a standard deviation of 2.85%. (Surprisingly, this is actually not as volatile as many growth stocks.)

On 15.9% of the days, Bitcoin could drop by more than 2.65%. On 2.3% of the days, Bitcoin could drop by more than 5.5% (2 standard deviations from mean).

Piecing it together, a stop loss of 3% could get hit one in 6 days just through Bitcoin’s daily volatility. To me, that’s a bit too tight a stop.

Instead, we could use 5.5% as a stop - 2 standard deviations from the average daily moves to give us some buffer against random daily downside moves. This 5.5% could be regarded as a minimum stop loss when placing a trade.

If we then apply the $1,000 (or 1%) rule, our trade position should be $1,000 / 5.5% = $18,181 on a $100k portfolio.

If you trade on a shorter timeframe, where 5.5% is too wide, you can still apply the same principle to hourly or intraday setups.

c) Consider the worst Earnings reactions

For equities, it makes sense to examine the historical earnings performance. This ensures that if the trade has not yet taken off before earnings, we could still sit through earnings without being knocked out by a big gap down after the earnings report.

Using TSLA as an example, the worst Earnings next-day reaction was 12.3% since 2019 (this actually occurred in the latest earnings). We could employ something like 13% as the minimum stop-loss and apply the $1,000 loss rule to it.

The worst recent earnings reactions I’ve seen have been in FSLY, where the stock printed -30% in February and -32% in May 2024. The prices just fell without any short-term bounce. In this case, a smaller position size would have saved the losses in absolute USD amount. Conversely, when the stock does do 2x or 3x (e.g. in Jan-Aug 2023), we don’t actually need a big position to capture the large upside.

5. Should We Use a Stop Loss Order, Exit Manually, or Wait for a Bounce?

This is a difficult question in practice. It is ultimately up to one’s individual style, risk tolerance, and discipline.

Having a stop loss order ensures that we can stick to our risk-reward ratios. You may be curious why I used the 0.8 payout in the Coin Toss example - surely no one would play that game?

Actually, plenty of people do it in the market.

When stocks fall, they cannot emotionally bring themselves to the exit, often comforting themselves with some Warren Buffett quotes. I will talk more about this in the next section. The result is that a potentially high risk-reward trade often turns into a realised low risk-reward trade, as the trader lets the losses compound.

On the other hand, they either immediately take out the trade when the stock bounces back to breakeven (risk = -30%, reward = 0%) or cash out a trade too early to lock in the gains (risk = -30%, reward = 15%).

Therefore, having a realistic stop loss and being able to execute that stop are crucial.

Overtime, what best works for me is to exit a trade manually on the day of the stop loss trigger. It’s not quite automatic and allows me to watch the minute-charts to judge whether there is a likely bounce. However, generally, even missing the bounces, the earliest exits tend to work out much better.

I will cover more about bounces in the next section.

6. Managing Losing Trades: How to Emotionally Cut Losses

I think every trader goes through two important phases before they can graduate: 1) winning big early on, and 2) losing massively without cutting losses.

Survivorship bias means that those who continue to trade after 6 months or so probably made some handsome money in the first few trades. This usually made the trade feel at least a genius, and quite often, the master of universe.

I still remember the feeling when I bet big on a Bank of Canada rate hike and made 25% of my yearly budget in one day, just 2 months into my institutional trading career.

There might be two reasons for this “beginners luck”.

1) Survivorship bias - those who lost early on probably would have given up quite quickly, “this isn’t for me”, and

2) Not overthinking - beginners tend to just use a simple thesis or observation and get on with it, not aware of all the other noise. This means they can execute without too much ego or fear.

The problem with winning big early is that we become arrogant about our true ability and unjustifiably comfortable at taking large risks. Greed quickly kicks in, begging for a bigger and bigger position to take advantage of the next 10x opportunity.

Until the market goes massively against us and we find ourselves unable to dig out of the hole.

This brings me to the second part of a trader’s journey: dealing with being stuck.

I think for most people, when they get stuck on a trade (i.e. losing beyond their initial expectation, whether having a stop-loss or not), the instinct is to hold on and wait for the trade to get back to even.

That’s also when they start citing investment legends like Warren Buffett, like “be greedy when others are fearful” or “when it goes down we love it, because we'll buy more”. Alternatively, they insist on the trade being right and just a little early, or the market is being ruined by “algos”.

The truth is that humans are loss-averse.

We become more risk-seeking when we lose. Making an unrealised loss (with the potential to bounce back) into a realised loss can be extremely painful. When we are on a losing streak, our instinct is to dial up the bets to make back the money as quickly as possible. But this usually ends up creating a bigger hole.

On the other hand, we become risk-averse when we get into a winning position, wanting to lock in the gains as soon as possible.

The result is that people tend to to double down on losing trades, at the expense of cutting the winning trades short. Overtime, the portfolio piles up with a bunch of underperforming names.

Good News: There Is a Way Out

From my own experience, I have found a practical way out of this cycle, which can potentially reduce the psychological pain of loss-cutting without suffering as much FOMO on the next bounce.

The way out is this:

It is crucial that things happen in this order and not the other way round. Until one actually cuts their losses, they cannot separate their loss-averse emotions from evaluating the trade. Dangerously, people actively look for information that confirm their bias to hold on to losing trades.

But as soon as you cut your loss, you can evaluate whether you should start a new position in the same name. This includes all the macro, fundamental, and price analysis we normally do. It also requires us to establish a new entry, take profit, stop loss, and position size.

This is a fresh new trade, it just happens to share the same ticker as your previous trade. It doesn’t matter what you paid on the previous trade and where you got out - that old trade is closed now. We only look at the new trade in front of us.

Very often, once we got out of the trade, the loss aversion pain actually stops.

It’s important to remember we always have the opportunity to get back in. If there was a bounce and we didn’t get in, it’s because we didn’t think this new trade has the the right risk-reward profile. It doesn’t matter that it takes off as soon as you sold, because each new trade competes with the opportunity costs of not entering other available trade setups.

7. Trading as a Volume Game: The Long-Term Approach

Apart from stop-losses, sometimes pulling the trigger for the entry can also very challenging. This is especially if one deploys a “buy-the-pullback” type of strategy, when the pullbacks sometimes get deeper than expected or occur when the overall market is crashing down.

Or perhaps you find yourself taking profit too early, a few percentage points before the trade reaches the target, just in case it sours. It is all part of the human instinct to protect what we have. This is yet another factor that reduces the reward-to-risk profile of a trade.

The outcome of any given trade is binary - either we made a profit or suffered a loss. This can overshadow the probabilistic approach we take to trading. It is easy to become very result-oriented. With 20-20 hindsight, we often feel something is a “slam dunk” or “never going to happen”. The short term results can really skew our judgement.

However, trading is a long-term, consistency-based, volume game.

We want to become consistent about finding the right risk-reward profile for every setup, pulling the trigger regardless of our instinctive resistance, and executing the take-profits (or trailing stops) and stop-losses without fear or FOMO.

8. Striving for the Perfect Trade

All of this takes years of self-improvement.

You may find yourself being frustrated at not making the “perfect” trade.

That’s okay. It’s what makes us human. It’s what makes trading difficult.

It is important to remember that we are on a life-long journey of progress, and the journey itself matters as much as the end destination.

Tools I find helpful:

Growth Investor Pro (GIP) on Seeking Alpha: I have been writing on Growth Investor Pro by Cestrian Capital Research for over 4 years. GIP is a tech-focused investing group which offers pro-grade research that finds opportunities to make money in any market. Sign up here to receive a 20% discount on the Growth Investor Pro membership.

Seeking Alpha: I am the dedicated FOMC Commentator on Seeking Alpha News. Seeking Alpha provides in-depth investment news, research, and analysis to 250k+ members. Sign up here to get $25 off your Premium Membership (7-day Free Trial).

TradingView: I use TradingView for all my chart analysis. It is very fast and simple to play with, including the mobile and desktop apps. Sign up here to get $15 off your new TradingView plan.

Koyfin: I use Koyfin for stock screening, comparative analysis, and fundamental analysis. It provides very comprehensive data and features at a reasonable price. Sign up here to get 15% discount off all Koyfin plans.

Fastgraphs: I find Fastgraphs very useful in gauging whether a stock is underpriced or overpriced versus the fundamentals. Sign up here to get 25% off using the coupon “AFFILIATE25” at the check out.

(Disclaimer: I am an affiliate of the above service providers and receive a commission from new signups.)

A quick “Thank You”

I really appreciate your support for this newsletter.

Please feel free to share any suggestions or comments, either directly through email or DM me on X

— Yimin