Evidence (track)
by Sampha
Evidence appears on the following album(s) by Sampha:
- Lahai (track #11) (this album) (2023)
Listen to Evidence on YouTube
Evidence ratings
Average Rating = (n ÷ (n + m)) × av + (m ÷ (n + m)) × AVwhere:
av = trimmed mean average rating an item has currently received.
n = number of ratings an item has currently received.
m = minimum number of ratings required for an item to appear in a 'top-rated' chart (currently 10).
AV = the site mean average rating.
Showing latest 5 ratings for this track. | Show all 22 ratings for this track.
| Rating | Date updated | Member | Track ratings | Avg. track rating |
|---|---|---|---|---|
| 03/12/2026 12:06 | Exist-en-ciel | 100,917 | 71/100 | |
| 06/17/2024 02:35 | leniad | 14,843 | 89/100 | |
| 04/25/2024 12:54 | Rayzer6 | 19,140 | 89/100 | |
| 03/24/2024 16:54 | migl | 9,972 | 86/100 | |
| 03/06/2024 15:03 |  covecove | 68,107 | 84/100 |
Rating metrics:
Outliers can be removed when calculating a mean average to dampen the effects of ratings outside the normal distribution. This figure is provided as the trimmed mean. A high standard deviation can be legitimate, but can sometimes indicate 'gaming' is occurring. Consider a simplified example* of an item receiving ratings of 100, 50, & 0. The mean average rating would be 50. However, ratings of 55, 50 & 45 could also result in the same average. The second average might be more trusted because there is more consensus around a particular rating (a lower deviation).
(*In practice, some tracks can have several thousand ratings)
This track has a Bayesian average rating of 74.9/100, a mean average of 74.3/100, and a trimmed mean (excluding outliers) of 74.3/100. The standard deviation for this track is 13.4.
Evidence comments
Be the first to add a comment for this track - add your comment!
Your feedback for Evidence
A lot of hard work happens in the background to keep BEA running, and it's especially difficult to do this when we can't pay our hosting fees :(
We work very hard to ensure our site is as fast (and FREE!) as possible, and we respect your privacy.
