Entropy Isn’t About Disorder—It’s About Probability: A Guide to Science’s Most Misunderstood Concept
Entropy ranks among the most misrepresented and widely misunderstood ideas in all of science. You have almost certainly heard it defined as the measure of “disorder” in a system, paired with the second law of thermodynamics that holds entropy of a closed system always increases over time. This leads to a common, lazy punchline: why bother tidying your office if it is only guaranteed to get messier again? That logic might sound intuitive, but you cannot blame entropy for your messy workspace. The untidy bedroom metaphor is a classic go-to for introducing the topic—nearly everyone can relate to a teenager’s perpetually cluttered room—but it is deeply misleading.
“Disorder” in thermodynamics does not mean clutter or chaos. Instead, it describes how many different ways you can rearrange a system’s individual parts without changing its overall, large-scale properties. For example: imagine your system is a sealed box full of air. At the microscopic level, every gas molecule inside zips and bounces around like a bumper car at a carnival. If you could freeze time and map the exact position and velocity of every particle, that specific arrangement is called a microstate. There are infinitely many possible microstates, and they shift trillions of times every second. What you actually observe, though, are the macroscopic properties you can measure: things like air pressure. If you sealed the box at sea level, that pressure will always read 14.7 pounds per square inch, no matter how the individual molecules move—unless you add energy to the system, like heating it up. Every one of those countless microstates corresponds to the same 14.7 psi macrostate. Put simply, entropy describes the link between the invisible atomic realm and the visible, measurable world we inhabit every day. It is a conceptual and mathematical bridge between two very different scales of reality—that is pretty cool, if you think about it.
When we talk about which outcomes we actually see in a system, everything boils down to random chance and probability. Probability is fundamental to entropy, and that is the big piece the messy room analogy leaves out. So let’s ditch the bedroom metaphor and use a far clearer one: rolling dice. Albert Einstein once famously quipped that “God doesn’t play dice with the universe.” Let’s test that idea, shall we?
Rolling the Dice
If you roll one standard six-sided die, you can get any number from 1 to 6—that is six total possible outcomes, or states. Roll a 20-sided die for your Dungeons & Dragons campaign, and you have 20 possible outcomes. If you want to impress your fellow players mid-session, you can casually note that the 20-sided die actually has higher entropy than the six-sided one: it has more possible outcomes, after all.
Now let’s say you are rolling three six-sided dice to calculate your character’s ability score. The sum of the three dice can be anywhere from 3 (all 1s) to 18 (all 6s), but not every sum is equally likely. If you want maximum dexterity, you need a total of 18. There is only one way to get that: every single die has to land on 6. But if you are okay with moderate dexterity, a total of 10 works just fine. That is far easier to roll, because there are many more combinations of three dice that add up to 10. To break it down: if you count unique sets of values, there are six distinct combinations that sum to 10. If you count the order you roll them in (which matters, since time only moves one direction), that number jumps even higher. For example, the set 6-3-1 can be rearranged into five other valid sequences, for a total of 27 different ordered outcomes that add up to 10.
Across all possible three-dice rolls, there are 216 total distinct microstates. The only thing that matters for your character sheet is the total sum—that is our macrostate. So the odds of rolling an 18 are just 0.4% (1 out of 216), while the odds of rolling a 10 are 12.5% (27 out of 216). We can say the macrostate of 10 has higher entropy, because there are more ways to get it. And because there are more ways to get it, it is far more likely to happen. There is no mysterious invisible force pushing a system toward higher entropy. It is just simple probability: higher entropy states are more probable. That is really all there is to it.
Opposite World: When the Unlikely Could Happen (But Never Does)
Let’s apply this logic to energy flow. Suppose you have a glass of cold water at 50 degrees Fahrenheit, and you drop a hot 120-degree copper ball into it. What happens? From everyday experience, you know the water will warm up and the ball will cool down, until both reach the same temperature somewhere in between. That is thermal equilibrium.
But what does that actually mean at the particle level? When something warms up, its atoms and molecules gain kinetic energy—they jiggle around faster. If the water gains 50 joules of thermal energy, the copper ball has to lose 50 joules, because energy is always conserved. But here is a mind-bender: what if one day, when you drop the ball in, the ball actually gets hotter, gaining 10 joules of energy, while the water loses 10 joules and gets even colder? Did you just break the laws of physics? Nope. Energy is still conserved. It is not impossible—it is just extremely unlikely. Why? Entropy. That outcome is one possible arrangement of energy, but it is so improbable that you would never see it in a trillion lifetimes.
Again, there is no secret physical force pushing energy to spread out, like a teen tossing clothes all over their bedroom. It is just that spreading energy across more particles creates a vastly larger number of possible microstates, making energy dispersion a statistical inevitability. That is why your hot coffee always cools down to room temperature.
An Object Lesson in Energy Arrangement
Let’s make this concrete with a tiny simplified model, just like our three-dice example. Imagine you have a solid so small it only has three atoms. Quantum mechanics tells us atoms can only hold specific amounts of energy, just like a die can land on 2 or 3 but never 2.5. If the total energy of our three-atom solid is 10 units, just like our three dice, we already know there are 27 different ways to split that energy between the three atoms.
Now let’s expand that: say we have two tiny solids, A and B, with different amounts of thermal energy. Object A has two atoms (like two dice) and 3 units of total energy. Object B has three atoms (three dice) and 7 units of total energy. The total energy of the combined system is 10 units, which follows the rule of energy conservation. Now we put A and B in contact, so energy can transfer between them. As long as the total energy stays 10, there are lots of different ways energy can be arranged:
If Object A only has 2 units of total energy, there is only one way that can happen: each of its two atoms has 1 unit. That means B has to hold 8 units, which has 21 possible arrangements.
If A has 4 units and B has 6 units, that combination gives us 30 total possible arrangements.
That second state has more possible microstates, which means it has higher entropy—and it is far more likely to occur.
The Law of Large Numbers Makes Probability a Law
This leads us to the classic definition of entropy: it is a measure of how many ways you can arrange the energy in a system, captured by the famous Boltzmann equation:
$$S = k_B \ln \Omega$$
Where S is entropy, k_B is Boltzmann’s constant, and Ω is the number of possible microstates for the system.
Of course, our tiny two- and three-atom model is just a toy for explanation. Most objects we interact with are made of staggeringly more particles than that: a single drop of water contains roughly 1.7 sextillion ($1.7 \times 10^{21}$) water molecules. What happens when we scale up this logic by orders of magnitude? The core idea stays the same, but the probability distribution becomes so heavily concentrated around the highest entropy state that any other outcome is effectively impossible.
While it is technically true that a hot copper ball could get hotter in cold water, it is so unlikely that it has never happened in the history of the universe. The state of thermal equilibrium—where both objects end up at the same temperature—is so overwhelmingly probable that we treat it as a guaranteed outcome in everyday life. That is what the second law of thermodynamics actually says: heat will always flow from a hotter object to a cooler object. But it is not an ironclad law that can never be broken—it is just a statement of overwhelming odds.
At the end of the day, it turns out Einstein was wrong: God really does play dice with the universe.
