Statistical median: median

The median template function calculates the statistical median of a collection of values. Give it a list of numbers and it returns the middle value when they are sorted. If the list has an even number of elements, it returns the average of the two middle values.

This is useful when you want a representative value from multiple sensorsSensors return information about a thing, for instance the level of water in a tank. [Learn more] that is less affected by outliers than average. For example, if you have five temperature sensors and one is reading incorrectly high, the median will ignore that outlier and give you a more accurate picture of the actual temperature. It works well for any situation where a single extreme value should not skew the result.

Usage

Here’s how to use this template function. Copy any example and adjust it to your setup.

{{ median(21.5, 22.0, 19.8) }}

21.5

{{ [21.5, 22.0, 19.8] | median }}

21.5

Function signature

The signature is a technical summary of this template function. It shows the name of the function, the values (called parameters) it accepts, and what type of data each parameter expects (for example, a piece of text or a number).

Function parameters that have a = with a value after them are optional. If you leave them out, the default value shown is used automatically. Function parameters without a default are required.

median(
    *args: list | float,
    default: Any = None,
) -> float

Function parameters

The following parameters can be provided to this function.

Using a default value

If the list might be empty or contain invalid values, provide a default to avoid errors. This prevents your templateA template is an automation definition that can include variables for the action or data from the trigger values. This allows automations to generate dynamic actions. [Learn more] from breaking when a sensor is temporarily unavailable.

{{
  [states("sensor.maybe_broken") | float(none)]
  | reject("none")
  | list
  | median(default=0)
}}

0

Good to know

• An empty list raises an error unless you supply a default.
• With an even number of values, the result is the average of the two middle values.
• Less sensitive to outliers than average, which makes it better for noisy sensor data.

Try it yourself

Ready to test this? Open Developer tools > Template, paste the example into the Template editor, and watch the result update on the right. Edit the values to see how the function adapts to your own entitiesAn entity represents a sensor, actor, or function in Home Assistant. Entities are used to monitor physical properties or to control other entities. An entity is usually part of a device or a service. [Learn more].

More examples

Real scenarios where this function comes up in automations and templates. Copy any example and adapt it to your setup.

Median of sensor values

Calculate the median temperature across multiple rooms.

{{
  median(
    states("sensor.living_room_temperature") | float,
    states("sensor.bedroom_temperature") | float,
    states("sensor.kitchen_temperature") | float
  )
}}

21.5

Median across a group of entities

If you have a groupGroups are a way to organize your entities into a single unit. [Learn more] of temperature sensors, you can expand the group and find the median of all their values in one expression.

{{
  expand("group.indoor_temperatures")
  | map(attribute="state")
  | map("float")
  | median
}}

21.5

Still stuck?

The Home Assistant community is quick to help: join Discord for real-time chat, post on the community forum with your template and expected result, or share on our subreddit /r/homeassistant.

Related template functions

These functions work well alongside this one:

• Average (arithmetic mean): average - Calculates the arithmetic mean of a list of values.

• Statistical mode: statistical_mode - Returns the most commonly occurring value in a list of values.

• Minimum value: min - Returns the smallest value from a list of values.

• Maximum value: max - Returns the largest value from a list of values.

• Expand groups into entities: expand - Expands groups and zones into a sorted list of individual entity state objects.