BaseEngine
BaseEngine is the platform's modeling layer. It fills gaps where local data is sparse, inconsistently reported, or unavailable.
It is on by default and is the recommended setting for all analysis.
Watch it work
Two minutes, using Paris as the example — from a citywide annual police release with no locations and no dates, to street-level scores.
Open the explainer in its own tab if you would rather watch it full-width.
What it does
Crime reporting is not uniform. Some police departments report promptly and completely; others report infrequently, irregularly, or not at all. Raw reported data in those places is misleadingly sparse — it looks like less crime when it is actually less reporting.
BaseEngine models across those gaps so that a location in a low-reporting environment is scored against what is actually happening rather than against what happened to get published.
With BaseEngine on, you get the platform's best normalized view of risk. With it off, you see only raw reported data.
Leave it on
Turn BaseEngine off only when you have a specific reason to isolate raw reported data — a research question about reporting behavior itself, for example. For every standard assessment, comparison, and client-facing output, leave it on.
Where the toggle is
In the filter panel. See Filters.
In the API
BaseEngine inclusion is controlled by the isGTM parameter, which defaults to true and matches the UI default.
isGTM is not a timezone flag
The name suggests GMT. It does not control timezone. It controls whether BaseEngine modeled data is included in results. Setting it to false because it looked like a timezone setting produces sparse, misleading results, particularly for international locations.
See API quickstart.