Do Snow and Ice Cause the Deadliest Crashes in Massachusetts?

No, and the data runs hard the other way. MassDOT crash records for 2022 through 2024 hold 405,240 Massachusetts crashes. Dry pavement carried 317,337 of them and 896 of the deaths, a rate of 2.82 fatal crashes per 1,000. Snow-covered roads carried 11,465 crashes and 6 deaths, a rate of 0.52 per 1,000. Ice carried 5,751 crashes and 5 deaths, 0.87 per 1,000. Taken together, snow, ice and slush account for 18,896 crashes, 4.66 per cent of the three-year total, at 0.79 fatal per 1,000, which is less than a third of the dry-pavement rate and well under the statewide figure of 2.68. The injury rate moves the same way: 23.33 per cent of dry-pavement crashes involved an injury against 15.40 per cent on snow, ice or slush. Winter produces more crashes and less severe ones. Where winter does show up is geographic, and sharply so: 8.76 per cent of Fitchburg’s crashes happened on snow, ice or slush, against 1.79 per cent of Boston’s, a difference of nearly five to one. To talk through a Massachusetts crash, call Jim Glaser Law at (617) JIM-WINS. The line is answered 24 hours a day. The first telephone consultation is offered without charge.

Where the Numbers Come From

MassDOT publishes Massachusetts crash records as open data, one dataset per calendar year, built from the reports police departments and drivers file with the Registry of Motor Vehicles. Every record carries a field for the road surface condition at the time of the crash and a separate field for the weather. Those two fields are not the same thing, and keeping them apart matters: it can be raining on dry pavement in the first minute of a storm, and a road can be icy under a clear sky.

The figures in this article come from a query run on 2026-09-22 against the 2022, 2023 and 2024 datasets, grouped on the road surface and weather fields and cut by severity. The same query run by community produced the town table further down. Those town totals match the figures already published on this site for all but one of the 349 communities, which is the check that makes the rest of it trustworthy: if the denominators agree town by town with the Massachusetts crash report, the splits built on them are working from the same records.

The one exception is Monson, where the live dataset now returns 80 crashes and the site’s published file holds 79. A single record has been added since the site’s data was last refreshed on 2026-09-01. It changes nothing in this article beyond the last digit of the statewide total.

How to Read These Numbers, In Order

1. Separate the road surface reading from the weather reading. They are not the same field, and they can diverge: it can be raining on dry pavement in the first minute of a storm, and a road can be icy under a clear sky.
2. Compare each surface’s rate against the statewide baselines, 2.68 fatal crashes per 1,000 and a 22.89 per cent injury rate, rather than against raw crash counts.
3. Check the crash count behind any rate before leaning on it. The sand, mud, dirt, oil and gravel fatal rate rests on 3 crashes, and the snow and ice fatal rates rest on 6 and 5.
4. Set aside the “Not Reported” weather category as a reporting artifact rather than a finding, since a fatal crash is investigated differently and the conditions box is more often left for a reconstruction than filled in at the roadside.
5. Read the injury rate alongside the fatal rate rather than in place of it. On road surface the two move together; on other cuts of this same file, such as darkness, they have moved in opposite directions.
6. Check a town’s own total against the Massachusetts crash report before relying on its winter share, since one added record can move the count between refreshes, as it did for Monson.

The Severity Table, by Road Surface

Two rates are reported for every surface, because reporting only one of them is how a reader gets misled. The first is how many crashes happened. The second is how bad they were.

Road surface	Crashes	Share of all	Fatal	Fatal per 1,000	Injury crashes	Injury rate
Dry	317,337	78.31%	896	2.82	74,029	23.33%
Wet	61,974	15.29%	151	2.44	14,643	23.63%
Snow	11,465	2.83%	6	0.52	1,577	13.75%
Ice	5,751	1.42%	5	0.87	1,051	18.28%
Slush	1,680	0.41%	4	2.38	282	16.79%
Sand, mud, dirt, oil, gravel	681	0.17%	3	4.41	208	30.54%

The statewide baselines against which each of those should be read are 2.68 fatal crashes per 1,000 and an injury rate of 22.89 per cent.

Dry pavement is above the statewide fatal rate. Snow is at roughly a fifth of it. A Massachusetts crash on dry pavement is about 5.4 times as likely to kill someone as a crash on snow, and about 3.2 times as likely as a crash on ice.

Answer Capsule

Snow and ice do not produce the deadliest Massachusetts crashes; dry pavement does. Across 405,240 crashes recorded by MassDOT for 2022 through 2024, dry pavement accounted for 317,337 crashes, 78.31 per cent of the total, and 896 of the deaths, a rate of 2.82 fatal crashes per 1,000. Snow accounted for 11,465 crashes and 6 deaths, 0.52 per 1,000, and ice for 5,751 crashes and 5 deaths, 0.87 per 1,000. Snow, ice and slush combined account for 18,896 crashes, 4.66 per cent of the three-year total, at 0.79 fatal per 1,000 against a statewide 2.68. Injury severity moves the same direction rather than the opposite one: 23.33 per cent of dry-pavement crashes involved an injury against 15.40 per cent on snow, ice or slush, where the statewide rate is 22.89 per cent. Wet pavement sits close to dry on both measures, at 2.44 fatal per 1,000 and 23.63 per cent injury, but carries a disproportionate share of pedestrian crashes: 18.01 per cent of Massachusetts pedestrian crashes happened on wet pavement, against wet pavement’s 15.29 per cent share of crashes generally. The two surfaces that break the pattern are small and dangerous: sand, mud, dirt, oil or gravel, 681 crashes at 4.41 fatal per 1,000 and a 30.54 per cent injury rate, the worst of any surface on both measures. By community, the winter share of a town’s own crashes ranges from 8.76 per cent in Fitchburg and 6.30 per cent in Pittsfield down to 2.07 per cent in New Bedford and 1.79 per cent in Boston. In law, none of this makes weather a defense: M.G.L. c. 90, sec. 17 provides that notwithstanding any posted speed limit, “every person operating a motor vehicle shall decrease the speed of the same when a special hazard exists with respect to pedestrians or other traffic, or by reason of weather or highway conditions.” Call Jim Glaser Law at (617) JIM-WINS; the first telephone consultation is free.

Both Rates Move Together Here, and That Is New

This site has now cut the same MassDOT file three ways on severity, and this is the first cut where the two measures agree.

The night crash analysis found that darkness moves the fatal rate more than fourfold, from 1.81 per 1,000 in daylight to 8.49 on an unlit road, while leaving the injury rate essentially flat at 22.47, 25.00 and 23.78 per cent. An earlier cut by community size found the same split: the fatal rate moved more than fourfold from the smallest towns to the largest cities while injury share stayed within a fraction of a point of the statewide 22.89 per cent.

Road surface does not behave that way. On snow the fatal rate falls to 0.52 per 1,000 and the injury rate falls to 13.75 per cent. Both drop, and they drop together.

That difference is informative rather than inconvenient. When a condition moves the fatal rate and leaves the injury rate alone, something is changing what happens after impact without changing how often people are hurt, which is the signature of speed and of delayed discovery. When a condition moves both rates down together, the plainest reading is that the crashes themselves are different events: slower, and more often metal against metal rather than metal against a person.

Why Winter Crashes Are Less Severe

The explanation below is INFERRED from the numbers above rather than measured directly, and it is worth labelling as such.

Snow and ice change driver behaviour before they change anything else. Speeds fall, following distances open up, and a meaningful number of trips do not happen at all. What is left is a large number of low-speed events: sliding through a stop sign, failing to hold a curve, sliding into the car ahead at a light. Those produce property damage and a police report, and the record counts each one as a crash. They rarely produce a fatality.

The bicycle and pedestrian counts support that reading. Of 4,356 Massachusetts bicycle crashes in the three years, 3,904, or 89.62 per cent, happened on dry pavement, against dry pavement’s 78.31 per cent share of crashes generally. Ten bicycle crashes in three years happened on snow. Cyclists are simply not out there, and the people most likely to die in a crash are the ones outside a vehicle.

The corollary is the part worth carrying away. The conditions under which Massachusetts road deaths actually happen are ordinary ones. Dry pavement, in clear weather, accounts for the large majority of them. The statewide fatal crash analysis covers where those deaths concentrate geographically.

The Two Surfaces That Break the Pattern

Sand, mud, dirt, oil or gravel. This is the smallest category in the table and the worst on both measures: 681 crashes, 3 of them fatal, a rate of 4.41 per 1,000 against the statewide 2.68, and an injury rate of 30.54 per cent against 22.89. It is the only surface in the dataset that is above the statewide line on both severity measures at once. Three fatal crashes is a very small number and the rate that follows from it should not be leaned on hard. The injury figure rests on 208 injury crashes and is a good deal more solid, and it points the same way.

Fog. Weather is recorded separately from surface, and the standout there is not snow either. Fog, smog or smoke accounts for 777 crashes, 0.19 per cent of the total, and 7 deaths, a rate of 9.01 fatal per 1,000. That is roughly 3.4 times the statewide rate and higher than any road surface in the table. Again, 7 fatal crashes is a small base. But fog is the one condition in this file that behaves the way people expect bad weather to behave, and it is also the one that most reduces the distance at which a driver can see a hazard.

By comparison, rain accounts for 26,170 crashes at 2.25 fatal per 1,000, below the statewide rate, and snow as a weather condition accounts for 7,704 crashes at 0.65 per 1,000.

One category that is an artifact and not a finding. The weather field reads “Not Reported” on 3,770 crashes, and those carry 35 deaths, a rate of 9.28 per 1,000. That number should not be published as though it said something about the weather. The likelier explanation is about how reports get filled in: a fatal crash is investigated differently from a fender bender, and the conditions box is more often left for a reconstruction than completed at the roadside. It is named here so that nobody later mistakes it for a result.

Where Winter Actually Shows Up: the Town Table

The statewide winter share is 4.66 per cent. It is not distributed anything like evenly, and the pattern follows elevation and distance from the coast almost exactly.

Among the communities with at least 2,000 crashes over the three years:

Community	Crashes	On snow, ice or slush	Share
Fitchburg	3,424	300	8.76%
Pittsfield	2,809	177	6.30%
Haverhill	4,815	296	6.15%
Leominster	3,543	211	5.96%
Shrewsbury	2,229	132	5.92%
Andover	3,302	189	5.72%
Marlborough	3,390	190	5.60%
Weston	2,281	126	5.52%
Chelmsford	2,108	112	5.31%
Worcester	15,598	762	4.89%
Cambridge	4,774	108	2.26%
Falmouth	2,146	45	2.10%
New Bedford	11,070	229	2.07%
Boston	17,407	311	1.79%

Fitchburg’s winter share is 4.9 times Boston’s. The towns at the top of that list are inland and north-central or in the Berkshires: Fitchburg, Leominster and Pittsfield sit in the part of the state that gets the most snow and holds it longest. The towns at the bottom are coastal, where the same storm arrives warmer and often as rain, or dense urban cores where traffic never moves fast enough for the surface to matter much.

By raw count rather than share, the biggest winter crash loads are Worcester at 762, Springfield at 421, Lowell at 332, Boston at 311 and Fitchburg at 300. Worcester alone carries more snow and ice crashes than Boston, on 1,809 fewer crashes overall.

Every figure above can be checked against this site’s own town pages. The full three-year record for each of the 349 communities is in the Massachusetts crash report, including Fitchburg, Pittsfield, Leominster and Shrewsbury. For the larger communities there are also local practice-area pages: Worcester car accidents, Boston car accidents, Haverhill car accidents and New Bedford car accidents.

What Massachusetts Law Says About Bad Weather

The reason any of this matters in a claim is that bad weather is the most common thing an insurer reaches for when it wants to move a crash out of the category of somebody’s fault and into the category of nobody’s fault. The Massachusetts speed statute closes that door in terms.

M.G.L. c. 90, sec. 17 opens with a standard that does not mention a number at all: “No person operating a motor vehicle on any way shall run it at a rate of speed greater than is reasonable and proper, having regard to traffic and the use of the way and the safety of the public.”

Later in the same section, after the posted-limit machinery, comes the sentence that decides most weather arguments:

but, notwithstanding such establishment of a speed limit, every person operating a motor vehicle shall decrease the speed of the same when a special hazard exists with respect to pedestrians or other traffic, or by reason of weather or highway conditions.

Read that carefully, because the direction of the duty is the opposite of the intuition. Bad weather does not lower what is expected of a driver. It raises it. A driver travelling at the posted limit on an icy road has not satisfied sec. 17 by pointing at the sign, because the statute says in terms that the limit does not relieve them of the obligation to slow down for the conditions. Snow is not an excuse under this section. It is a circumstance that the section requires a driver to respond to.

Being partly at fault does not end a claim either. Under M.G.L. c. 231, sec. 85, Massachusetts bars recovery only where the injured person’s own negligence is greater than the total negligence of those they are claiming against, and otherwise reduces damages in proportion to their share. The mechanics are covered in the comparative negligence guide.

What This Means for a Claim

Three practical points follow from the numbers and the statute together.

“It was snowing” is an argument, not a conclusion. It describes a circumstance that sec. 17 required the driver to account for. The questions that actually decide the case are what speed the vehicle was travelling, what the driver could see, and what a reasonable response to those conditions would have been.

The conditions defense is raised far more often than the conditions warrant. On this data, 95.34 per cent of Massachusetts crashes happened on a surface that was not snow, ice or slush, and 78.31 per cent happened on dry pavement. The same argument appears in motorcycle files, where it fits even less: where motorcycle accidents happen in Massachusetts found dry pavement in 92.4 per cent of them.

The surface is evidence, and it is perishable. Whether the road was treated, when it was last plowed or sanded, and what the surface actually looked like are all facts that exist on the day and are gone a week later. Photographs taken at the scene, and the timing of any municipal treatment, are worth preserving early. How a Massachusetts crash gets reported in the first place is covered in reporting a car accident, and a fall on ice at a property rather than a crash on the road runs under an entirely different set of rules, set out in slip and fall on ice and snow.

Key Terms

Road surface condition. The MassDOT crash record field for the condition of the pavement at the time of the crash, kept separate from the weather field.

Weather condition. The MassDOT crash record field for the weather at the time of the crash, distinct from the road surface field because the two can point different ways at once.

Fatal crashes per 1,000. The rate used throughout this article, fatal crashes divided by total crashes for a category and multiplied by 1,000, which allows comparison across surfaces of very different sizes.

Comparative negligence. The M.G.L. c. 231, sec. 85 rule that bars recovery only where the injured person’s negligence is greater than the total negligence of the parties they are claiming against, and otherwise reduces damages in proportion.

Duty to decrease speed. The M.G.L. c. 90, sec. 17 requirement that a driver decrease speed for weather or highway conditions notwithstanding any posted speed limit.

Frequently Asked Questions

Q: Do most fatal Massachusetts crashes happen in bad weather?

A: No. MassDOT records for 2022 through 2024 put 896 of the 1,087 fatal crashes on dry pavement, and dry pavement’s fatal rate of 2.82 per 1,000 is above the statewide 2.68. Snow-covered roads carried 6 fatal crashes and ice 5, rates of 0.52 and 0.87 per 1,000 respectively. Snow, ice and slush together account for 4.66 per cent of all Massachusetts crashes and a smaller share of the deaths.

Q: How many Massachusetts crashes happen on snow or ice?

A: 18,896 over the three years, counting snow, ice and slush together, out of 405,240 crashes. That is 4.66 per cent. Snow alone accounts for 11,465, ice for 5,751 and slush for 1,680.

Q: Which Massachusetts city has the highest share of winter crashes?

A: Among communities with at least 2,000 crashes in the three years, Fitchburg, where 300 of 3,424 crashes, or 8.76 per cent, happened on snow, ice or slush. Pittsfield follows at 6.30 per cent, Haverhill at 6.15 per cent and Leominster at 5.96 per cent. Boston is the lowest of the large communities at 1.79 per cent. By raw count the largest winter loads are Worcester at 762 and Springfield at 421.

Q: Is wet pavement dangerous in Massachusetts?

A: Wet pavement behaves much like dry on severity, at 2.44 fatal crashes per 1,000 against 2.82 on dry, and a 23.63 per cent injury rate against 23.33. It carries 61,974 crashes, 15.29 per cent of the total. The place it stands out is pedestrians: 1,077 of the 5,980 Massachusetts pedestrian crashes, 18.01 per cent, happened on wet pavement, against wet pavement’s 15.29 per cent share of crashes generally.

Q: What is the most dangerous road surface in Massachusetts?

A: On these three years of data, sand, mud, dirt, oil or gravel. It is the smallest category at 681 crashes, and it is the only surface above the statewide line on both severity measures: 4.41 fatal crashes per 1,000 against 2.68, and a 30.54 per cent injury rate against 22.89. The fatal figure rests on 3 crashes and should be read as a direction rather than a precise rate; the injury figure rests on 208 and is firmer.

Q: Can a driver blame the weather for a Massachusetts crash?

A: Weather is a circumstance the law requires a driver to respond to rather than a defense to having failed to. M.G.L. c. 90, sec. 17 requires that speed be no “greater than is reasonable and proper, having regard to traffic and the use of the way and the safety of the public,” and provides that notwithstanding any posted limit, “every person operating a motor vehicle shall decrease the speed of the same when a special hazard exists with respect to pedestrians or other traffic, or by reason of weather or highway conditions.” Whether a particular driver met that standard is a question of fact about that crash.

Q: Does being partly at fault in a snowstorm end my claim?

A: Not by itself. M.G.L. c. 231, sec. 85 bars recovery only where the injured person’s negligence is greater than the total negligence of the parties they are claiming against, and otherwise reduces damages in proportion to their share. The statute also presumes that the injured person was in the exercise of due care, and places the burden of proving otherwise on the party asserting it.

Q: Where can I check the crash figures for my own town?

A: The Massachusetts crash report on this site carries the full three-year MassDOT record for each of the 349 communities, including total crashes, fatal and injury counts, bicycle and pedestrian counts, and the highest-crash roads by name. Every figure in this article was produced from the same MassDOT datasets and the town totals match that report for 348 of the 349 communities, the exception being a single record added to Monson since the site’s data was last refreshed.

Q: Is Jimmy Knows AI giving me legal advice?

A: No. This is general information about Massachusetts law and public crash data, not legal advice, and reading it does not create an attorney-client relationship. The figures here describe patterns across hundreds of thousands of crashes and say nothing about any particular one. For advice about a specific situation, speak with a Massachusetts attorney.

Talking to a Lawyer

The numbers on this page are background rather than evidence. What decides a case is the particular crash: the speed, the sight lines, the condition of that stretch of road at that hour, and what the vehicles and the people in them did. Several of those facts have a short life, and the road surface is the shortest-lived of all of them.

Surface is one of several conditions that move Massachusetts crash severity, and they do not all move it the same way. The posted speed limit at the crash location separates the two severity measures instead of moving them together, which is set out in which Massachusetts speed limits carry the deadliest crashes.

Jim Glaser Law can be reached at (617) JIM-WINS. The line is answered 24 hours a day. The first telephone consultation is offered without charge.