You know the story, or at least some of it. Sixty-five million years ago, a rock about the size of Mount Everest careened out of space and slammed into modern-day Mexico. It opened a Hawaii-sized hole in the crust, launching soot and sulfur high into the atmosphere, blocking out the sun.

Within days, air and water temperatures plummeted around the world. Within weeks, the food chain on both the land and the ocean had collapsed. Within years, the dinosaurs—the rulers of Earth for more than 150 million years—had perished (except, of course, for the birds).

This moment, the Cretaceous-Paleogene mass extinction, is the most recent widespread die-out in our planet’s history. The dinosaurs’ departure opened the gates for a new class of creature—that is, mammals, like us—to inherit the top of food chains around the Earth. And the moment was also, according to a paper published Thursday in Nature Scientific Reports, quite improbable.

When the asteroid slammed into Mexico 65 million years ago, there was only a 13 percent chance that it would trigger a mass extinction, argue Kunio Kaiho, a geoscientist at Tohoku University, and Naga Oshima, a senior researcher at the Japan Meterological Agency.

“The probability of significant global cooling, mass extinction, and the subsequent appearance of mammals was quite low after an asteroid impact on the Earth’s surface,” they write in their paper.

Odds were 87 percent that the asteroid wouldn’t have prompted the evolution of mammals. In other words, the odds were longer that mammals would triumph than that Donald Trump would, on the eve of the 2016 president election (at least according to FiveThirtyEight).

What set these odds? The location of the asteroid’s impact with Earth—and the mechanism that actually killed the dinosaurs in the days after the impact.

Researchers once believed that the Chicxulub asteroid (so named because it struck the modern-day location of Chicxulub, Mexico) was so devastating because it prompted forest fires around the world. Those fires released soot and ash into the high atmosphere, blocking out the sun’s rays and cooling global temperatures. Indeed, all around the world, geologists find the remnants of soot and sulfur at the rock layer that signifies this moment in geological history.

But geologists and climate scientists have questioned whether forest fires alone could have set off a mass global cooling. Two years ago, a team led by Kaiho looked at the molecular structure of the soot. They argued that all that soot came from one origin point, and that its ratio of elements suggested a higher-energy burn than would normally occur in a wildfire.

The asteroid, they said, must have struck a site rich in hydrocarbons, like crude oil, natural gas, or kerogen. Only about 13 percent of the Earth’s surface contains enough subterranean fuel to set off a mass global cooling.

“Hence, if you assume that such an impact occurred at a random place on Earth, there would have been a 87 percent change that there would have been no mass extinction at all,” said Johan Vellekoop, a paleoclimatologist at the University of Leuven in Belgium who was not connected to this paper, in an email.

“It turns out the dinosaurs just happened to be very, very unlucky. This aspect of chance is often neglected in our field, and I think it is really cool that Kaiho and Oshima have tried to put a number on it,” he said.

Kaiho and Oshima are not the only researchers who have looked at how contingent the timing of the asteroid’s impact was on its consequences. Earlier this year, a group of British and American scientists showed that if the asteroid had entered Earth’s atmosphere mere seconds earlier or later, it would have collided with the sea instead of North America. An aquatic impact would have sent much less dust and soot into the sky, dampening its effects.

Vellekoop took issue with how Kaiho and Oshima arrived at their 13 percent number. They used climate models to estimate global temperature decline with the amount of soot in the atmosphere. He said this linked climate change too directly with the mass extinction:

For their calculations, they basically assume that all the extinctions that occurred at the Cretaceous-Paleogene boundary can be ascribed to climate change, and then go on to put a number on it: “Approximately 8–10 degrees Celsius temperature changes causes a mass extinction.” In doing so, they basically dismiss all other detrimental consequences of the Chicxulub impact, such as the blocking of sunlight and surface-water acidification.

The severity of climate change–driven extinctions greatly depend on the rate of change. Does it occur over days, months, years? Also, different species of plants and animals react differently to changes. Many species will already go extinct with a 5-degree change, some others might require much more. Putting actual reliable numbers on this would require a whole new study, probably involving a complex biological modeling component.

But he said their experiment was valid as a “thought experiment,” and as a first-pass estimation of how unlikely the rise of mammals may have been.

Kaiho’s work is not the only work being done on the mass extinction of the dinosaurs. Last year, the evolutionary biologist Manabu Sakamoto argued that dinosaurs began declining 40 million years before the infamous asteroid collision. For tens of millions of years, more species of dinosaurs disappeared than were replaced by new species.

The time scale, by the way, points to the longevity of dino dominance on Earth. Dinosaurs first appeared on this planet about 230 million years ago. They declined for 40 million years before they went extinct, at least in Sakamoto’s telling. But only 65 million years separates us from that asteroid collision. During that interval, a tiny shrewlike creature branched into horses and antelopes, sloths and polar bears, orangutans and blue whales—and the entirety of human history took place.

“Their long decline had made dinosaurs vulnerable to sudden environmental changes, such as those caused by the asteroid impact,” said Sakamoto in an email. “A sudden catastrophe followed by a substantial global cooling may have been too much of a stressor for dinosaurs to recover from.” He said that Kaiho’s paper did not conflict with his team’s hypothesis.

Lucky for us, the odds that an asteroid collision prompts a mass extinction haven’t changed much in the past 50 million years. In an email, Kaiho said there’s still about a 13 percent chance that any asteroid impact would toss up enough soot and ash to cool global temperatures.