The famous quote by economist John Maynard Keynes, “In the long run we are all dead,” has often been misinterpreted as flippant or dismissive. However, when read in full context, Keynes wasn’t making a nihilistic joke; he was issuing a pointed critique of economic thinking overly fixated on long-term outcomes. The full passage reads:
"The long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the storm is past the ocean is flat again."
Keynes' point wasn’t to discount long-term planning but to emphasize that focusing solely on eventual outcomes isn’t enough. To make a real impact, we must also grapple with immediate challenges. Saying that everything will even out eventually doesn't help if you're in the middle of a storm and need to figure out how to keep the ship afloat.
Here’s the key: making use of this insight does not mean throwing up one’s hands and disregarding strategies that may take a long time to work. Working with long time horizons and strategies that rely on methodical long-term execution can be powerful drivers of success in financial, business, and economic endeavors. The distinction is that to make these strategies workable in practice, patience and commitment are not enough. Proper reverence must be paid to the thoughtfulness and positioning required to allow you to withstand the storms that come while you wait.
This insight doesn’t just apply to economics—it resonates in any field where decisions must be made under uncertainty, especially when the number of opportunities to act is limited. From understanding statistical probabilities to crafting strategies in sports, the challenge of balancing long-term trends with short-term imperatives is a universal problem.
The Trap of Statistical Averages
Statistics often paint a comforting picture of predictability. For example, we know that flipping a fair coin a million times will result in approximately 50% heads and 50% tails. But what if you only flip the coin three times? The odds of deviating significantly from the "long-run average" are much higher, and in practice, you must make decisions based on the small sample size you actually have.
This distinction is critical. Just because a strategy or outcome is statistically likely over many repetitions doesn’t guarantee it will play out that way in the handful of opportunities you’re given. For instance, a casino can confidently rely on the long-term profitability of its games because it processes thousands of bets daily. But an individual gambler playing one hand of poker has no such assurances.
Similarly, decision-making under real-world constraints often involves sample sizes that are too small for the long-run averages to be helpful. In these scenarios, the outcome hinges more on understanding the immediate situation than on banking on eventual statistical convergence.
Put simply, you should understand when you and your situation present an actual sample size of “1”. If 990 out of 1,000 people are expected to experience a given outcome over a given time period, this will provide you with zero comfort if your “1” turns out to be in the excluded “10.” In a sample size of 1 with discrete outcomes, it’s often impossible to be 90% or 99% successful – you might be forced into a binary outcome even if you don’t want to think about it that way.
Lessons from Sports: Baseball vs. Football
Sports can offer a vivid illustration of the tension between long-term averages and short-term stakes.
Conjecture has occasionally been made that the NFL was slower to adapt to analytics than the MLB because of culture issues. But the argument could certainly be made that there’s more at play than that. After all, when it’s too easy to exploit an advantage that others refuse to exploit, you’ll often see someone do it, until the advantage goes away (arbitrage, market mechanics, etc.)
But what facts might have led this departure between the two sports to seem rational?
Baseball, with its 162-game season and hundreds of at-bats per player, provides an ample sample size for trends and probabilities to play out. Over time, a player's batting average is a reliable metric, and teams can use quantitative methods to optimize strategies across many games.
Football, by contrast, operates in an entirely different context. A typical NFL team plays only 17 regular-season games, with far fewer plays per season than a baseball team sees at-bats. The smaller sample size makes football outcomes much more prone to variability. A single play can determine the outcome of a season, as can a single injury or referee's decision.
It’s not surprising, then, that football was slower to embrace analytics compared to baseball. The smaller sample sizes make it harder to extract actionable insights from the data, and even when analytics suggest a certain strategy is "optimal" over the long run, that insight might not be helpful when the coach faces a critical fourth-down decision in a playoff game. At that moment, the probabilities are only part of the story; the specific context—opponent tendencies, player fatigue, weather conditions—matters just as much, if not more.
The key takeaway is that strategies optimized for the long run don’t necessarily translate to high-stakes, short-term scenarios. In baseball, you can play the averages because you have time to absorb losses and let the probabilities work in your favor. In football, when you only have one play to win the championship, the odds almost don’t matter—what matters is figuring out what will work right now.
Implications Beyond Sports
This principle extends to many areas of life and decision-making. Consider investing: while a portfolio diversified to attempt to track a market index like the S&P 500 might be predicted based on history to yield steady returns over decades, that insight doesn’t help a retiree who needs to liquidate assets during a market downturn. Long-term growth projections based on a working life of 30 more years essentially cease to matter if your actual working life is cut shorter than.
Similarly, in business, betting on a strategy that’s “likely to succeed eventually” can be disastrous if the company doesn’t have the runway to weather short-term losses.
The same goes for public policy. Telling people that inflation or unemployment will stabilize "eventually" is cold comfort if they can’t pay their bills this month. Effective leaders, like effective coaches, must combine long-term vision with short-term pragmatism.
Balancing Long-Term Insight with Short-Term Action
Keynes’ critique of long-term thinking is as relevant today as it was when he wrote it. While long-term trends and probabilities provide valuable guidance, they aren’t enough on their own. Real-world decision-making requires grappling with the messiness of the present—navigating the storm rather than waiting for the ocean to calm.
Understanding when to trust the averages and when to focus on the here-and-now is a delicate balancing act. Whether you’re an economist, a coach, or just someone trying to make the best decision with limited information, success depends on recognizing the limits of long-term thinking and embracing the complexity of the moment at hand.
Make sure you understand the full scope of your positioning for all the complexity that happens in the finer time slices that add up to “the long run.”
In the long run, the averages may smooth everything out—but only if you survive the short run first.
2024-7444792.1 Exp 12/26