Statistics integration works not as it should be

Configuration
1

Hi,

I have an outside temperature sensor. It reads values every 30 seconds (which might be important or not). I’ve made a sensor using statistics integration to show avarage (mean) temperature from last 12h.
It completelly doesn’t work as expected. Let’s say that minimum recorded temperature from last 12h was 7 degress, maximum was 12,5 and current is 8,5. Avarage from last 12h is currenly 10,8 and even tho current temp. (8,5) is LOWER than avarage (10,8) from last 12h this avarage temperature keeps RISING! Maybe I’m not the best in math but this doesn’t make any sense. If current temp. is lower than avarage then avarage should be FALLING. Here’s my config:

sensor outside_avarage_temperature_over_last_12h:
  - platform: statistics
    name: "Outside avarage temperature over last 12h"
    entity_id: sensor.outdoor_temperature
    state_characteristic: mean
    precision: 1
    sampling_size: 1440
    max_age:
      hours: 12

What am I doing wrong or what do I don’t understand how this integration works?

2

In addition - here’s graph of temperature from last 12h. On bare eye you can see that avarage is not 10,8 but much less.

image
image1033×417 6.85 KB

3

I know I’m replying to myself like idiot but I made manual count hour after hour based on graph and it really looks that this avarage 10,8 is correct. Nethertheless I still don’t get it how it’s possible that if current temp. is lower than avarage then how come this avarage got rising.