Well it depends on the size. We get hit by tiny chunks of rock all the time, every year and even smaller ones by the thousands during meteor showers, but they all burn up in the atmosphere. But you probably mean large enough to get through the atmosphere probably and to hurt people.

So, first, you can see that the chance must be pretty small, because there has been no impact by a large asteroid or comet on any populated area for all of recorded history. Also the smaller ones are far more likely than the larger ones, as there are many more smaller ones than larger ones. The most likely impact is a 50 meters or less one. Not even as large as the Meteor crater in Arizona.

Meteor Crater - Wikipedia

That’s a one kilometer diameter crater. Also it hit a desert. And, yes - an asteroid hitting Earth is most likely to hit a desert or the sea because though our Earth may seem crowded if you live in a big city, actually four fifths of our planet is water and most of what is left is unpopulated ice sheets, tundra, deserts, mountains etc.

Something this small would have no effect at all in the sea, not large enough for a tsunami and in a desert it would make a crater and kick up a bit of dust, but have hardly any effect on a wider scale. If it happened to hit a city of course it would be devastating, but the chance of that is tiny.

Meteor crater formed 50,000 years ago. We don’t have any records of anything as dramatic as that in recorded history. So, they are clearly not very likely. We’ve had many tsunamis, volcanic eruptions and earthquakes, and a few really tiny craters forming, but nothing like Meteor crater, never mind anything larger.

As for even larger ones, the chance of being hit by a 10 km asteroid is pretty much eliminated, as we have plotted all the orbits of all the ones that could do flybys of Earth and they don’t hit us for the next several centuries. No surprise as it was only a 1 in a million chance per century anyway, but now we can rule that out. We can also rule out a 10 km comet because though we’d only spot it three or four years in advance, maybe a bit more than that, we get very few comets at present compared to asteroids. For every 146 asteroids that does a flyby of Earth you get only one comet. So - a comet might do a flyby of Earth but it is exceedingly unlikely for a comet say as big as 1 km or larger to hit Earth this century. We’d know about a 1 km comet also at least two years before it’s flyby.

But we do have some examples of deaths by smaller meteorites. Now that we’ve found all the 10 km asteroids that could hit us and shown they all miss, and are on track to find 99% of the ones of 140 meters upwards by the early 2030s, the attention is turning towards the smaller ones.

HISTORICAL DEATHS BY METEORITE

Sometimes you hear people say that nobody has ever been killed by a meteorite, but that’s not quite true. There have been a few fatalities. With the Chelyabinsk meteorite then there were some serious injuries, mainly from flying glass but nobody killed. You may also know of Anne Hodges who had a rather nasty bruise on her thigh from a meteorite.

The True Story of Ann Hodges: History's Only Meteorite Victim

That’s written before Chelyabinsk of course.

But, though very rare, there are quite a few historical accounts of injuries and of fatalities, including some very old ones that may be authentic. This is long before anyone knew or even guessed that asteroids existed or that meteorites came from the sky. It was a meteorite field fall in England, the Wold Cottage meteorite that eventually convinced the doubters. The theory that they came from space was only proposed the previous year in 1794, by Ernst Chladni

Wold Cottage meteorite in 1795. Confirmed the theory put forward only the previous year that meteorites came from space.

The historical records include e.g. many Chinese accounts of “iron rain” which made holes in houses and hillsides and killed people. They had no idea what they were of course. But looking back with the benefit of hindsight - what else could that be but meteorites?

There is one very early example where we know for sure it was a meteorite, the Ensisheim (meteorite). November 7, 1492, wheat field outside walled town of Ensisheim, Alsace, Further Austria (now France) This is a contemporary illustration of what happened.

And broadsheet about it:

Sebastian Brant, Donnerstein von Ensisheim, 1492

The reason we know for sure what it is is that they dug it out at the time though of course they had no idea what it was. The villagers started to break it up but a local magistrate stepped in and preserved the stone to give to king Maxillian. So we have it to this day.

Ensisheim meteorit donnerstein 1

It’s a carbonaceous chrondite. So this definitely was a meteorite. Details here:

And there are several fatalities from the twentieth century that most don’t know about.

You are far more likely to be killed by lightning than by a small meteorite but it can happen - here are all the examples I’ve found of people being killed by meteorites, or injured or hit by one and not injured at all:

• 616 jan 14, large shooting star destroyed wall-attacking tower, which collapsed and crushed 10 people to death
• 1020 some people killed by many stones in unknown place in Africa
• 1321–23 - “iron rain” in O’chia distrcit causes holes in hills and houses and kills people and animals
• 1341, similar story, “iron rain” in Chin-ning district, damages crops and kills people and animals
• 1342 - iron rain in Kun-ming district - damages crops and houses
• 1490 stones like rain in Ching-yang district kills tens of thousands of people - rather implausible for the small amount of meteorites. It could have been a Tunguska type event or, - it may have been hail with large hailstones.
• 1511, September 14, Monk and several animals killed at Cremona in Lombardy, Italy
• 1639 Large stone falls suddenly into market street in China, destroys several tens of houses and kill several tens of people
• 1664 -1665 Franciscan friar dies as a result of being struck on thigh by meteorite, Milan
• 1648 Eight pound meteorite may have killed two sailors on the ship Malacca traveling from Holland to Batavia
• 1874, huge stone falls in China during a thunderstorm, crushes half a cottage, kills a child, at Chin-kuei Shan in Ming-tung Li
• 1907, sept 5, stone falls and crushes to death entire Wang family - probably from collapse of their house at Hsin-p’ai Wei in Weng-li
• 1908, 30th June, two people died of their injuries in the Tunguska event in Siberia and several hundred reindeer were killed
• 1915 April 25, Meteorites fall into Ta-yang village, one of them falls on a woman’s shoulder and tears off her arm.
• 1929 December 8. Meteorite hits a bridal party and kills one person in Zvezvan Yugoslavia - reported in New York Times
• 1946 May 16, meteorite destroyed hamlet of Santa Ana in Nuevo Leon state of Mexico. 8 killed, 26 injured, reported Reuter and New York Times.
• 1951 Aug. 15, 62 houses destroyed in Tehran, Iran by meteorite shower. 12 killed, 19 injured, 300 livestock killed. Reported by Iranian newspapers and United Press.
• 1954, 30th November, 3.9 kg meteorite strikes woman on left hip and thigh while taking a nap on a couch (after bouncing off her radio),
• 1992, 14th August, boy hit on head by 3.6 gram meteorite which bounced off leaves of a banana tree in meteor shower
• 2013 Feb 15, Chlyabinsk, 1500 injured, 112 hospitalized, two in a serious condition.

Papers here: Human casualties in impact events

and here: Meteorite falls in China and some related human casualty events

There's also a long list by the International Comet Quarterly with cites here Interesting meteorite falls which includes damage to buildings and other structures as well as deaths and injuries to people. More links: Death by Meteorite! Also a few in Historical Meteor Shockwave Events That Destroyed Structures. Then there’s Lewis’s book Rain Of Iron And Ice which I’ve just ordered but not read yet often referred to as a source on deaths by meteorite.

For anyone interested in researching into this topic, there’s a very long list of meteorite falls here also, in chronological order, inclusive, would need to check individual ones to find out more. Chronology of Earth Impacts as it is very inclusive. For instance, one of its sources is this web page which only gives sources for some of its entries so you’d need to do a lot of searching for sources to verify them: Meteorites, Asteroids, and Comets: Damages, Disasters, Injuries, Deaths, and Very Close Calls

One of the deaths in the Tuguska event is also described here:

‘Probably the closest observers were some reindeer herders asleep in their tents in several camps about 30 km (20 mi) from the site. They were blown into the air and knocked unconscious; one man was blown into a tree and later died. "Everything around was shrouded in smoke and fog from the burning fallen trees."’ 1908 SIBERIA EXPLOSION: Reconstructing an Asteroid Impact from Eywitness Accounts

And in more detail here:

“[O]f the approximately twenty people who were within fifty kilometers of ground zero, it appears that all were slightly injured (Lewis 1996)." Thousands of reindeer were killed, as well as some dogs. One elderly man was thrown against a tree during the blast and later died, presumably from his injuries, and another old man died of shock. Another man was thrown to the ground and bit off his tongue. Several people were knocked unconscious, and one family’s hut was blown into the air, causing bruises to all inside “

Quote from Lewis’s book Rain Of Iron And Ice: The Very Real Threat Of Comet And Asteroid Bombardment which I have ordered but don’t have yet, will update with more extensive quote when I have the book. Found it here: Meteorite Impact Structures Student Research

The second paper attempts to calculate the fatality rate based on the figures for China, which had a better information gathering network than most places.

Based on those figures and increase in population and the current population of Earth and its land area the authors estimate that we should get one fatal accident worldwide from meteorites every 3.5 years. If so it’s rather surprising given that we have had no reports of deaths at all by meteorite anywhere in the world since that one in 1907. Also based on current fireball statistics we’d expect only a 28% probability of a single casualty over that entire time period.

• Maybe it is small number statistics, only based on five reports? (However they don’t seem to have reported single person casualties which would be expected to be far more common than these ones that affect many people)
• Maybe this is a particularly quiet time with fewer meteorites than usual?
• Maybe it’s just underreporting, and in many parts of the world maybe still people don’t recognize meteorites, or we just don’t hear because communications are poor.

It’s strange. But anyway - this is by far the most likely kind of meteorite deaths to happen. One person most likely of all. Or perhaps a handful of people.

It’s hard to say quite what the risk is, because you get conflicting figures and we don’t know the numbers of the very small ones. Those Chinese figures suggest there may be more of them than we think. But there have been no fatalities published from the US, and only one injury in the last century, in 1954 when a meteorite bruised a lady’s leg. And surely it wouldn’t be under-reported in the US? Surely most people would recognize a meteorite if they saw one? So I think the chance can’t be that high for the US. Far lower than lightning anyway. Lightning kills about 100 people a year and injures about 1000 a year in the US. NOAA Lightning safety facts

WARNINGS OF SMALL METEORITES

We would have warnings for anything large unless it comes from the direction of the sun. We have a blind spot in that direction.

Sadly, radar doesn’t help until it is very close. The problem is that radar’s effectiveness drops off as the inverse fourth power - the light has to go out to the object you shine the radar at, then reflect back and come back to Earth. So unlike ordinary observation where if it is ten times further away it is a hundred times fainter - if it is ten times further away for radar, it is ten thousand times fainter. So radar can give you hardly any warning at all, minutes perhaps.

The meteorites also may come from an awkward direction and not spotted by our ICBM detection radars

That’s why the Russians didn’t spot the Chelyabinsk meteorite in advance Did Russian early-warning system see the meteorite?

The new ATLAS system: Asteroid Terrestrial-impact Last Alert System can give early warnings from most directions, but it can’t do it if they come from the direction of the sun like the Chelyabinsk one. ATLAS can give warnings of from a week to several weeks for small objects unless they approach from the direction of the sun. ATLAS can detect a 120 meter diameter asteroid 3 weeks before impact unless it approaches form the direction of the sun when it couldn't detect it at all.

An asteroid from an awkward direction like that could still hit us with only a few hours of warning. A newly discovered asteroid did a flyby with very short notice earlier this year: Hours after discovery, asteroid swept by.

Note though that Chelyabinsk if it had been just a little larger, say 40 meters across, could have been spotted several months before the impact when it happened to be close enough to be seen - we just missed it because it was a little too faint.

FINDING MEDIUM SIZED ASTEROIDS 140 METERS UPWARDS

The Large Synoptic Survey telescope will find 80% of all NEOs larger than 140 meters within a decade, first light in 2021.

Rendering of the Large Synoptic Survey Telescope

So by around 2030 we will already 99% of the ones from 140 meters upwards.

MIND THE GAP - 40 METERS TO 140 METERS

Hopefully we'll also have space telescopes by then which would be needed to fill the gap from 40 meters to 140 meters as well as to help find the last few of the larger ones. There’s a system we can use to do that which costs only $50 million using eight cubesats.

Below 40 meters they usually burn up in the atmosphere but can get to the ground at a steep angle depending what they are made of.

NEW DEVELOPMENT SYNTHETIC TRACKING OF ASTEROIDS

However there are new developments in asteroid tracking which may help here. Especially, use of synthetic tracking. The idea is explained in techy detail in this paper. Finding Very Small Near-Earth Asteroids using Synthetic Tracking. For an easier to read summary of it, see “Synthetic Tracking” Set to Revolutionise Near-Earth Asteroid Discovery

The idea is that instead of doing a 30 second exposure, you do many shorter 2 second exposures. With conventional CCD's that adds to the read noise so you get more errors but there are new CCD's developed for medical imaging that permit fast accurate reading, called Scientific CMOS detectors. The Andor Zyla is an example here.

Andor Zyla 5.5 | sCMOS Camera medical imaging camera capable of fast read out with low read error

You can then use this to simulate tracking the asteroid with the camera, which makes the asteroid far brighter in the images.

This image shows a the result of stacking many photographs of asteroid 2009BL with camera set to follow the stars on the left - notice how the asteroid is shown as a streak, and rather faint. On the right, the same photos are stacked to follow the asteroid which then shows as a much brighter spot, and the stars are streaked and fainter.

Image from: DETECTION OF A FAINT FAST-MOVING NEAR-EARTH ASTEROID USING THE SYNTHETIC TRACKING TECHNIQUE

When the asteroid is small and traveling faster across the field of view, the trail can be so faint it can’t be distinguished from background noise when the camera follows the stars. If you know its velocity you can make it much brighter by following the asteroid. But what can you do if you haven’t detected it yet and don’t know which way it is moving? The idea of synthetic tracking is that you take lots of short exposure photos and just try stacking them in many different ways until you find the right velocity and an asteroid pops into vie win the photo. This is time consuming but modern graphics cards permit fast parallel processing which makes synthetic tracking feasible.

This approach can make it easier to spot fainter asteroids. It might mean for instance that you can spot an asteroid ten times further away than before. That means a thousand times the volume of space covered. So this technique can lead to a huge increase in the detection of asteroids.

We could retire most of the asteroid impact risk for $50 million with this new technique

The researchers found that fewer than eight cubesats, fitted with 15 centimeter synthetic tracking telescopes could find more than 70% of NEO's larger than 45 meters in diameter in less than six years (these are the asteroids that are most hazardous for us). The total cost would be $50 million. With larger 30 cm telescopes then eight satellites could find 95% of the NEOs larger than 45 meters in diameter in the same time period of less than six years. For details see their 2016 Annual Progress Report.

$50 million is not a lot. Any developed country could find that out of loose change from its defense budget. $50 million spread over the US population, say, is a one off cost of 15 cents per person. In the UK we could do it for a one off cost of £1.60 per person - that's to find them for the entire world. As an example, the UK recently voted to renew Trident at a cost variously estimated at £40 billion to £205 billion ($50 billion to $256 billion). The cost of finding 70% of Near Earth Asteroids down to 45 meters within six years is a tiny 0.1% of that. We can certainly afford to do that.

See also my Giant Asteroid Headed Your Way? - How We Can Detect And Deflect Them

And my Doomsday Debunked