The Art of the Home Run Trade

Waiting for Your Pitch

When you get a pitch to hit, put a good swing on it. There's no better way to explain a concept than through an analogy, and in this case, there's no better analogy than baseball. The top players in today's game, Ohtani, Judge, Guerrero Jr., and Soto, hit the ball about 30% of the time on average. There has never been a player in the history of the game to hit 50% of the time. Trading works the same way. We can make millions with a sub-50% win rate, and in some extreme cases, even sub-20%. The key is in sizing and execution.

MLB batting averages during the 2023-2024 regular season

Most of a discretionary trader's day, week, month, or even career is spent waiting. Truly great trades are hard to find. But every now and again (or maybe more frequently if you've got serious alpha), the stars align. This is where careers are made. People obsess over trading data and building systems, but one of the biggest differences between huge winners and everyone else is their ability to recognize great opportunities and then size them appropriately.

Let's go back to baseball. Great hitters know which pitches to swing at, but on a deeper level, they understand which pitches will be balls, which will be singles, and which will be home run balls. Not every pitch can be hit out of the park. The talent of truly gifted hitters is putting the right swing on the right ball. When the pitcher misses his spot, the greatest hitters put it in the stands.

Markets work the same way. Some trades aren't home runs; they're just there to sustain the PnL curve and give us more experience. But every now and again, a trade appears that ticks every box. An opportunity so good you feel it in your balls. These are the career-defining moments. It doesn't take a genius to spot these opportunities, but it takes a truly skilled and fearless market participant to size them correctly and hold them long enough to see the entire thesis play out. The best example of this would be Stanley Druckenmiller and George Soros's bet against the pound.

A Lesson In Conviction: How Soros and Druckenmiller Broke the Bank of England

In 1992, while working at George Soros's Quantum Fund, Stanley Druckenmiller and his analyst identified that the British economy was struggling and the pound was severely overvalued. The thesis was so strong that Druckenmiller wanted to put 100% of the fund into the trade. Here's the story, as told by Druckenmiller himself:

“I go in at 4:00 and I said, ‘George, I’m going to sell $5.5 billion worth of British pounds tonight and buy deutsche marks. Here’s why I’m doing it; that means we’ll have 100 percent of the fund in this one trade.’ And as I’m talking, he starts wincing like what is wrong with this kid, and I think he’s about to blow away my thesis, and he says, ‘That is the most ridiculous use of money management I ever heard. What you described is an incredible one-way bet. We should have 200 percent of our net worth in this trade, not 100 percent. Do you know how often something like this comes around? Like once in 20 years. What is wrong with you?’”

This trade netted over $1B in PnL and is perhaps the greatest example of all time when it comes to swinging for the fences. Now, we're not saying you should leverage yourself to the max on every single trade. But you need to know which opportunities require home run swings and which ones shouldn't even be swung at. That's the skill that separates the good traders from the great ones.

Quantum Fund NAV vs DEM/GBP pair trade

Putting Theory Into Practice

So how do we actually translate this philosophy into executable sizing decisions? The answer largely depends on your strategy. Quant funds will use different metrics than L/S equity shops, and L/S shops will use different metrics than macro funds. At Ceruleus Capital, we're a discretionary macro fund, which means we use a variety of metrics including volatility targeting, correlation-adjusted positioning, and exposure caps. But the foundation of our sizing methodology, like many other funds, is the Kelly Criterion, which is a formula that's as useful for retail traders as it is for massive investment funds.

Kelly Criterion

Kelly Criterion is great, but it has its cons, one of which is that it's insanely sensitive to inputs. If you misjudge your win probability or expected payoff, Kelly will tell you to size massive, and you'll consequently get blown up. This is why Kelly requires a large sample size of trades to estimate accurate probabilities. Even then, markets evolve, and edges can deteriorate.

The solution to input sensitivity is to implement Fractional Kelly. Instead of betting the full Kelly recommendation, you bet a fraction, typically 1/4. Sizing at 25% of the original Kelly formula dramatically reduces the risk of ruin while still capturing most of the upside. The real art of sizing, however, comes during those rare, career-defining setups, which is when you absolutely do not want to stick to 1/4. You throttle up to 1/2, 3/4, or even full Kelly. Knowing when to make that adjustment is the difference between good traders and icons like Druckenmiller.

Sizing Constraints

Kelly is just the starting point, not the finished product. At the fund level, there are constraints that matter much more:

1. Max drawdown tolerance: No matter how large your conviction or what Kelly says, risk must be capped at a predetermined percentage.

2. Correlation risk: If you're already long equities and short bonds, adding another risk-on trade, even if Kelly says to size it big, requires you to either trim existing positions or reduce the new one. Understanding correlation risk is paramount in reducing a blowup in tail events.

3. Liquidity risk: For funds that are extremely large or operating in smaller markets, liquidity risk becomes a major factor. Positional losses can easily double or triple if the market experiences a liquidity crunch, such as the Covid-19 decline, or more recently, the tariff tantrum. Position size must account for how much you can actually exit without moving the market against yourself.

The Framework

Here is how we think about sizing in practice:

• Baseline: Start with fixed risk per trade (0.25%, 1%, etc.)

• Conviction scaling: Increase that baseline based on setup quality, confluence of factors, and personal edge

• Kelly check: Run fractional Kelly (1/4 to 1/2) to get the mathematical recommendation

• Blend: Weight your conviction-based sizing against Kelly's suggestion to find the optimal size

• Override: On true home runs, such as the Soros-Druckenmiller trade, you may need to exceed your risk models. This requires deep experience and intimate knowledge of your own system

Waiting for Your Moment

Not every trade is going to be a winner. In fact, over half your trades might be losers. But that doesn't matter if you're sizing them correctly. The point of risk management is to keep you in the game long enough to find great opportunities with outsized risk-reward profiles. A great crypto trader once said, "Five times a year there's free money on the ground, pick it up, and do nothing else." While I don't fully agree that you should do nothing else (experience matters), the core message is true. In discretionary trading, truly great theses don't come around often. Waiting for the stars to align can take a hell of a long time. But like Soros said, when they do appear, you need to be ready to swing for the fences.