The models agree: once chronic wasting disease saturates a herd, deer numbers should fall. Southern Wisconsin's core provides a strong test of that prediction. In the 2025 sampling year, roughly one out of three sampled deer tested positive in Richland (35.7%) and Sauk (33.0%) counties. Prevalence across the eleven core counties ranged from 16.5% to 35.7%. We used two independent field measures, deer-vehicle collisions and total harvest, to check whether the core herd was declining faster than herds in counties with prevalence below 5%. Each measure can give a misleading picture of herd decline for different reasons. Through 2024, the core showed a more favorable trend in crashes, but a somewhat larger harvest decline (7% versus 3%). The harvest difference was modest and showed no collapse. Population effects could still appear later.
KEY FINDINGS
- Across Wisconsin's eleven counties with the highest prevalence, the rate of deer-vehicle collisions, measured per billion vehicle-miles, dropped by a median of only 6% between 2010 to 12 and 2022 to 24. Counties with lower prevalence recorded a 30% decline.
- Total harvest declined by a median of 7% in the core, compared with 3% in counties below 5% prevalence. The gap is real but small, and the decline does not amount to a collapse.
- In plain terms, neither measure showed a core collapse. Harvest alone points toward the models' prediction. Collision data showed less decline in the core than in counties below 5% prevalence.
- The survival findings remain sound: infected deer die at young ages. Still, herd-size effects trail prevalence by years. That delay ends earliest in the core, making this a herd that deserves attention rather than dismissal.
IThe prediction
Where prevalence is expected to lead to fewer deer
Wisconsin research conducted over many years modeled a prevalence threshold for CWD in adult female deer. At that level, population growth would fall below replacement, and deer numbers would be expected to drop. If this threshold model is correct, effects on herd size should emerge over years to decades following increased prevalence. The core should then lose deer at a measurably faster pace than counties below 5% prevalence. This paper tests that expectation using collision and harvest data collected independently.
IITwo honest signals
Crashes and harvest mislead in different directions, so check both
Harvest may look like a useful measure of herd size, but it can be misleading. Permit totals, hunter effort, and hunting rules affect it as much as deer numbers do. Deer-vehicle collisions have the opposite limitation: they rise and fall with traffic. Adjusting collision counts for vehicle-miles gives a collision rate, but road and traffic changes still affect that rate. Different factors confound the two measures, so agreement between them does not confirm a change in deer numbers or its cause. Here is what the measures show for the core.
IIIWhat the core actually shows
County results vary without a consistent downward pattern
The core shows mixed signals across counties. Richland, with the highest prevalence in the state, had a 46% drop in its collision rate, the sort of result the model predicts. Yet Iowa County's rate climbed 50%, Walworth's 51%, and Grant's 27%, with all carrying 20-plus-percent prevalence. Harvest tells a mixed story too: Iowa fell 44%, while Rock, Walworth, and Juneau increased. The signals do not form a consistent picture of a herd being wiped out. If CWD were reducing deer numbers in the core as the threshold model warns, eleven of the counties with the highest prevalence would show a different pattern.
Swipe table to see every column
| County | 2025 prev % | Crash-rate Δ | Harvest Δ |
|---|---|---|---|
| Richland | 35.7 | −46% | −12% |
| Sauk | 33.0 | +1% | −20% |
| Lafayette | 31.0 | −21% | −34% |
| Columbia | 27.4 | −41% | −6% |
| Green | 26.6 | −6% | −7% |
| Iowa | 23.5 | +50% | −44% |
| Rock | 23.1 | +20% | +1% |
| Grant | 20.7 | +27% | −6% |
| Walworth | 20.0 | +51% | +4% |
| Dane | 19.3 | −16% | −10% |
| Juneau | 16.5 | −25% | +2% |
IVA clock, not an all-clear
The delay is central to the finding
These findings do not show that CWD causes no harm. Evidence demonstrates that infected deer die at young ages, while prevalence continues to rise. Models place herd-size effects years to decades behind rising prevalence. A decline absent through 2024 therefore cannot establish that CWD is harmless. At very high prevalence, a Wyoming study using collars found an annual decline of 10.4%. Continued attention to the core is warranted. If the models are correct, the highest-prevalence areas should show effects earliest, giving us a reason to keep tracking the core.
The models forecast falling deer numbers. Through 2024, neither measure showed a collapse in Wisconsin's core. Harvest alone showed a slightly larger decline than in counties below 5% prevalence. These findings do not determine the cause. Continue tracking the core.
NOTES AND SOURCES
- The core consists of 11 Wisconsin counties at ≥15% CWD prevalence for sampling year 2025, with ≥50 deer analyzed. Counties below 5% prevalence form the comparison group.
- The crash rate counts collisions involving deer for each billion vehicle-miles (WI DNR Deer Metrics / WisTransPortal), removing the effect of traffic-volume growth. Harvest counts all deer harvested within a county (WI DNR deermetrics). Each measure uses 3-year averages for the starting and ending periods (2010 to 12 compared with 2022 to 24), reducing noise from individual years.
- Different factors independently confound the measures: hunting rules and effort affect harvest, while road and traffic changes affect crashes. A faster core decline qualifies only if both indicators show it. Here, the evidence for that decline is weak and limited to harvest.
- This statement concerns herd populations rather than the risk to an individual deer. Evidence establishes that CWD kills infected animals; the companion paper discusses Wisconsin's epicenter and the threshold of 29%.
- The lack of a decline signal through 2024 provides no forecast for the decades ahead; the delay between prevalence and its effect on deer numbers is the question being examined, rather than a qualification attached to it.
Fully Drawn Research is an independent data analysis desk, not affiliated with any state wildlife agency, transportation agency, mapping provider or outfitter. Figures trace to Wisconsin DNR and WisTransPortal public records, retrieved June–July 2026.
Keep reading
FDR Scout puts county harvest history like this on the map of the ground you hunt. Open Scout →