Michigan's Deer-Crash Hotspots: Eight Counties Never Left the Top Quarter

IThe 2024 answer

Huron takes first place among leaders from several regions

For the greatest likelihood of hitting a deer relative to driving distance, look to Huron County in 2024: its rate reached 3,393.7 deer collisions for every billion vehicle miles. Next was Presque Isle, with 3,252.1. Alcona recorded 3,081.8, Hillsdale had 2,735.0; Sanilac posted 2,608.0. Rounded to one decimal place, Charlevoix and Dickinson both show 2,197.3. Their placement as 9th and 10th in the following table is arbitrary.

To calculate the measure, divide crashes involving deer by total driving distance within a county and express the result for a billion miles. This addresses a familiar flaw in counting crashes alone: deer and traffic both affect the total. During 2024, Wayne County recorded 624 collisions involving deer, exceeding Alcona's 409. Drivers covered about 14.7 billion miles in Wayne, compared with 0.13 billion in Alcona. The paper on why deer-crash maps mislead we wrote measured that issue in Iowa: harvest and collisions involving animals and vehicles had a correlation of only r = 0.306 when compared across 99 counties. The paper identified a rate based on miles driven as the useful ranking measure.

The Thumb accounts for three counties among the leading seven: Huron, Sanilac and Tuscola. Among the lowest rates were Wayne's 42.3, Macomb's 110.1 and Menominee's 153.0. Huron therefore had about 80 times the rate recorded in Wayne.

IIEight for eight

Counties that kept making the cut

A yearly snapshot can include both lasting hotspots and counties with an unusually high rate that year; these data cannot explain the reasons. We distinguished them by sorting all 83 counties according to their rates annually during 1997 to 2024. We selected the highest 21, rounding a quarter upward. Eight counties qualified through all 28 years.

Those eight counties cover several areas. The Thumb has two, Huron and Sanilac; the Upper Peninsula has two, Dickinson and Iron. The remaining four, Alcona, Mason, Missaukee and Mecosta, lie elsewhere within the Lower Peninsula. Rates declined in every county over that span. Huron had the smallest drop, 13.5%: its rate began at 3,921.2 and finished at 3,393.7. The largest decline, 59.2%, belonged to Mecosta, whose rate started at 4,044.8 and ended at 1,650.4. Even with these decreases, all kept a place among the highest quarter annually. For 2024, Mecosta recorded 1,650.4, exceeding the statewide county median, 1,152.4.

Compare that group with the 2024 top 15: it includes seven counties from the eight consistently high-ranking counties, with Mecosta absent. Another eight made that year's ranking without qualifying for the consistent group: Presque Isle, Tuscola, Hillsdale, Charlevoix, Oceana, Gladwin, Manistee and Branch. Every one of them placed outside the highest quarter during at least one year. Even Presque Isle, second for 2024, failed to qualify. That gives a good reason to avoid planning from just a lone year's ranking.

IIIThe typical county

A third lower at the median

County rates opened with a median of 1,719.3 for 1997. The high point was 1,831.2 for 1999; the low was 842.4 for 2016. From 2017 onward, the median has remained within a range of 1,041.7 to 1,277.8. Its closing value for 2024 was 1,152.4, a decline of 33.0% from the starting value.

For the rate to drop, collisions must decline in relation to driving distance. A smaller crash total or greater mileage alone does not ensure that result when the other measure changes too. This record does not separate their effects. Nor does a declining rate establish a decline in deer numbers. There is also an unexplained jump in the series: the median was 842.4 for 2016, then rose to 1,079.6 for 2017. We did not examine that jump. Such a large shift reminds us: rates can change for reasons barely related to deer.

IVMethod and limits

What the rate tells you, and its limits

We use each county's crashes involving deer as the numerator, taking them from Michigan's traffic-crash statistics. Those figures originate in crash reports filed by police; collisions never reported to police are excluded. For the denominator, we use yearly vehicle mileage across every road within each county, supplied by the HPMS tables for counties from Michigan's Department of Transportation. According to MDOT, mileage is estimated using traffic counts, substituting annualized estimates when counts are absent. We divide the crash total by that mileage to produce the rate. Using MDOT's mileage tables alongside the crash counts we had stored, we recalculated four rates for particular counties and years: Alcona and Huron for 1997, Iron and Macomb for 2024. All four results agreed.

Deer numbers help shape a rate, but so do roads, traffic patterns and time spent driving. It therefore does not count the herd. With little driving mileage, a county can record a high rate from relatively few collisions. Where traffic is heavy, a county can record a low rate even with a large crash total.

Choosing the cutoff affects membership in the highest quarter. Exclude one more county at that cutoff, and two members of that eight-county group, Mecosta and Sanilac, no longer qualify. Near the other end, Menominee sits in the Upper Peninsula, as do Dickinson and Iron, yet its 2024 rate is third-lowest (153.0). We did not test the reason.

VOn the road

Different guidance for driving and hunting

Drivers get a fair comparison from the rate because mileage is already accounted for. Measured that way, the eight consistently high-ranking counties have presented a high chance per mile of hitting a deer over 28 years, including many autumns. The chance for a mile driven in Wayne County or Macomb County is much lower. These rankings apply to counties; they do not distinguish individual roads or hours. They cannot identify the stretch of a particular route where you should reduce speed.

Hunters need a different approach. Crash rates mark encounters between vehicles and deer. Our Illinois collision-density paper examined a county index that accounts for mileage when adjusting crash totals. In a single year's comparison, its relationship with harvest was only moderate (r of about 0.31 for 2024). The paper regarded the trend across multiple years as evidence that relative deer abundance had changed. Apply that approach to Michigan's rates. Repeated placement among the highest quarter over many years puts a county on your shortlist, but warrants nothing beyond that. Next, review county harvest results: Michigan tallies mandatory harvest registrations at the county level (83 counties, as described in our discussion of the state's two harvest numbers). Also find out how much land is actually available for you to hunt. Remove a county from consideration if its crash rate is the only evidence in its favor.