Let's say you are a new member of the SARS-CoV-2 family, with a few genetic tweaks that distinguish you from the rest of the clan. Maybe you have changes in your spike protein that allow you to more easily to invade cells, or a random mutation that helps you elude the virus-killing effects of a COVID-19 treatment.

You could be a contender. But first, you have to get out of the body you're in and infect some more people.

What would it take for you to prosper? Alternatively, what would cause you to die, a short-lived experiment in evolution's harsh proving chamber?

A modeling exercise by researchers at the Fred Hutchinson Cancer Research Center in Seattle suggests that for a genetic variant to become a menacing new presence, it's not enough to be scary and highly transmissible. It also needs lucky breaks.

Most notably, it must gain entree to a superspreader event if it is to have a fighting chance of planting its flag in a population. In the brief period during which its carrier is at or close to his peak viral load, the new variant needs to hitch a ride to someplace where people are mingling at close quarters and many are not wearing masks.

Once there, it needs to infect at least a handful of people. Five would be enough to live on to compete for more victims. Infecting 20 or more will give it a real chance of becoming predominant in its new community.

Time is of the essence, research suggests: Even for a new variant that's armed with transmission superpowers, that first superspreader event needs to come within a month of its arrival for the variant to stand a chance of becoming established.

That's a lot of narrow windows and a lot of high hurdles, and the likelihood that a new variant will clear all of those obstacles is actually pretty slim, the researchers concluded.

But then there's reality: At least five new "variants of concern" have apparently overcome these forbidding odds in the span of about six months.

That suggests there are probably many more such variants out there, each looking for its lucky break. It takes only one or two with the right constellation of mutations to prolong or escalate the pandemic — or to undermine the vaccines that could end it.

One thing researchers are increasingly sure of is that in patients with compromised immune systems, the coronavirus is more likely to take on not just one but a passel of mutations. Those genetic changes could make it even harder to fight the virus with medicines, masks and vaccines.

A model like this doesn't yield calibrated measurements of how effectively an intervention like masking can stop spread. It doesn't generate predictions about the pandemic's next stage. But it does provide insights into how a virus behaves under a range of circumstances, along with probabilistic estimates that can sharpen the intuitions of public health officials.

"We will in all likelihood create new variants on top of those that have emerged," said Dr. Joshua T. Schiffer. "And the ones that will win are the ones that dodge the vaccine or transmit more easily."