This site will look much better in a browser that supports web standards, but it is accessible to any browser or Internet device.

No Kill Buffalo-Niagara homepage

No Kill Buffalo-Niagara
"Sometimes it falls upon a generation to be great. [We] can be that great generation." — Nelson Mandela


This website was designed and is maintained by volunteers of No Kill Buffalo-Niagara.

Stats graphic

Statistics . . .

Current No Kill status of area shelters: No Kill = save rate of 90% or better

Quicklinks to downloadable statistics:

Notes on Computing Statistics:

Asilomar statistics for area shelters are provided below. These documents are provided by the shelters themselves and follow the standard Asilomar format. Please note that the Asilomar save rate, sometimes printed at the top of such reports, is NOT the raw live save rate. The Asilomar save rate excludes animals who are lost or died in care as well as "owner-requested euthanasia" cases. Therefore, the Asilomar save rate is inflated compared to what No Kill standards dictate. For a discussion of the problem see No Kill News (Note: site temporarily down at this time).

To determine the actual, uninflated raw live save rate, there are many methods that are generally accepted by No Kill standards. In all cases, a proportion will be computed, in which the total number of live outcomes (adoptions, transfers out, returns to owner/habitat) will be divided by the total outcomes. The problem is how best to compute the relevant number of total outcomes.

(1) Method 1: Total outcomes is computed as (Starting Shelter Count plus Total Intakes minus Ending Shelter Count).

(2) Method 2: Total outcomes is computed as all live outcomes plus all non-live outcomes. In other words: Live Outcomes (Adoptions plus Transfers Out plus Returns to Owner or Habitat) plus Non-Live Outcomes (Total Euthanasia plus Lost or Died in Care).

(3) Method 3: Total outcomes is computed as (Starting Shelter Count plus Total Intakes).

Ideally, Method 1 is the most accurate. But particularly in large shelters, there will be discrepancies and some animals will not be accounted for. Sometimes where will be outcomes over what they should be, perhaps animals taken in but never logged, some adoptions not logged.

Method 2 looks only at recorded outcomes, and ignores "slippage" in record-keeping. Usually, "missing data" are few, so this method may be reasonable.

Method 3 actually penalizes for a carryover of animals -- which can be a reasonable penalty in the calculation for shelters that do excessive "warehousing" of animals and particularly where the starting/ending shelter count is increasing over time. However, any shelter will have a certain capacity and one would expect for No Kill shelters to remain at capacity.

Our preferred method is to use Method 2, but then also track and look at separate trends to see if large-scale shelters are engaging in some potential machinations to pump up their save rates. Relevant trends to track include:

  • Intake -- Are shelters contracting their intakes? One way to log "improvements" in save rates is to not take in as many animals or to selectively intake animals. In some cases, this may be a good thing. For example, a shelter that at one time took in feral cats and killed nearly 100% would at least be improving by not taking such cats at all. The expectation is that intakes would remain relatively stable, unless the shelter made major improvements in outreach to the community to keep animals in the home, significant spay/neuter outreach (70-80% sterilization in an area is generally required to reduce population expansion), and implemented other programs that have had a measurable impact on shelter intakes.
  • Classification of animals -- For those shelters who make much of definitions such as "adoptable" or "treatable" it can be valuable to see if the proportion of animals so classified is not changing, with more and more being designated "unadoptable" or "unhealthy/untreatable." This is another reason raw live save rate is superior to looking at adoptables or treatables.
  • Adoptions -- Good shelters will constantly be working to find ways to increase adoptions, even of animals who may be seen as difficult to adopt. That might include animals with special needs or who may have behavioral issues that necessitate careful placement.
  • Starting/ending shelter counts -- Shelters should NOT be going No Kill simply by warehousing animals. The point is to move them out into homes. By watching trends here, or counts in relation to "comfortable shelter capacity," we can get a picture of whether the shelter is not doing all it can to place animals or get them the rehabilitative care they need to be placed into homes.
  • Transfers out to kill shelters -- Look to make sure that the shelter isn't just shipping out animals who will then likely be killed by the receiving shelter or rescue. As an example, a municipal shelter could kill 10% and send its other "problem" animals to a shelter that will also "cull" a certain percentage, sometimes far in excess as might have occured at the hands of the sending shelter.

City of Buffalo Animal Shelter statistical reports (downloads) — yearly:

Erie County SPCA statistical reports (downloads) — yearly:

Niagara SPCA statistical reports (downloads) — monthly:

Where to find your shelter's statistics? If they voluntarily report Asilomar statistics, they might just be online at: Participating Organizations - Asilomar Accords. Organizations who get certain grants may also be required to report as a condition of their grant. No Kill rescue groups are unlikely to report such statistics because they do not operate shelters per se, and are frequently dealing with far fewer cases than animal control or county shelter facilities. For information on No Kill rescues, you will probably have to look at annual reports or contact them to give you information about their intakes and non-live outcomes.