Stats Explained

In 2006 and 2007 I spend long nights thinking about how to use last.fm data to produce even more meaningful information, many of this stats where inspired by the last.fm community, a few of them invented them myself.

AEP

The AEP is a number that is between -20 and 5 that shows how much preference a last.fm user has for his or her top artists. 5 is the max value for the AEP and it takes this value when the user has listened all the 50 top artists an equal number of times. It stands for Anti Exponiential Points, yeah I know very very silly but I can't change it now :)

The AEP Formula:

AEP = 5 - 25 * ( Slope / AverageTop50 )
Slope = (value of the first artist in the chart - value of the 50th artist in the chart) / 50
AverageTop50 = ( The sum of all values in the top 50 ) / 50

Example:

Tracks Listened from each artist in the Top50 chart:
96 83 82 76 72 66 66 64 63 60 59 58 52 51 51 50 50 49 48 47 47 47 46 45 45 44 44 43 41 41 40 40 39 39 38 38 38 37 37 37 37 37 35 35 35 35 35 34 33 33

Slope = (value of the first artist in the chart - value of the 50th artist in the chart) / 50
Slope = (96 - 33) / 50 = 1.26 , 96 from the first artist, 33 from the last artist.

AverageTop50 = ( The sum of all values in the top 50 ) / 50
AverageTop50 = (96 + 83 + 82 + 76 + 72 + 66 + 66 + 64 + 63 + 60 + 59 + 58 + 52 + 51 + 51 + 50 + 50 + 49 + 48 + 47 + 47 + 47 + 46 + 45 + 45 + 44 + 44 + 43 + 41 + 41 + 40 + 40 + 39 + 39 + 38 + 38 + 38 + 37 + 37 + 37 + 37 + 37 + 35 + 35 + 35 + 35 + 35 + 34 + 33 + 33) / 50 = 48.36

AEP = 5 - ( Slope / AverageTop50 ) * 25
AEP = 5 - ( 1.26 / 48.36 ) * 25
AEP = 4.35 Done!

Original AEP post

How ... Am I?

In the old days of last.fm a small meme/game was popular and it was about how EMO you were, this was of course based just in your music tastes.

I wrote a journal in last.fm critizicing the formula used for that calculation and then I came out with my own calculation, the real calculation done in last.fm extra stats has some extra steps to refine it more but the idea is about the same.

Find the emo score (Example) of each one of the your Top10 Artists and multiply it by the number of tracks played for each artist.

From First to Last = 80*123 =9840
Coheed and Cambria = 75*115 = 8625
Atreyu = 28*104 = 2912
Chiodos = 43*98 = 4214
Silverstein = 99*88 = 8712
Deftones = 8*87 = 696
Hopesfall = 53*85 = 4505
Funeral for a Friend = 100*82 = 8200
Explosions in the Sky = 1*81 = 81
Avenged Sevenfold = 22*80 = 1760

Sum all the emo scores = 49 545
divide it by the total of tracks listeneds in your top 10 = 123+115+104+98+88+87+85+82+81+80=943
And my emo rating is 52.54%!

Original How ... Am I? post

Mainstreamness

It tells you how popular are the music that you listen to. It uses the most listened artists in last.fm to compare with yours.

Tag Cloud

I will just copy-paste from the most popular tag cloud online generator at the time:

A musical user tag cloud is a textual depiction of one's listening habits, here based on musical social networking service Last.fm. The weight of a tag in the cloud is determined by the play counts of top artists in the user's profile, and the top tags for those artists, labeled as such by the Last.fm community

Diversity

It's like an AEP but for the tags/genres. It first creates your personal tag cloud and then uses for calculating how diverse is your music taste using the same AEP formula detailed earlier.

Original Diversity post

Compatibility

Mesaures the music compatibility between two last.fm users

The procedure is like this, starting with 2 last.fm profiles:

C26000 Profile 162 The Appleseed Cast
159 Atreyu
.
.
.
65 On The Might of Princes
64 Incubus

Neocronos Profile
550 Still Remains
491 Jimmy Eat World
.
.
.
50 My Chemical Romance
48 Ours

The first step is normalize each profile, divide each for number of tracks value by a special number N, which can be calculated using the following formula

N = (1st#Tracks^2 + 2nd#Tracks^2 + 3rd#Tracks^2 + ... and so on until the 50th#tracks^2) ^ (1/2)

for C26000 N = ( 162^2 + 159^2 + ... + 65^2 + 64^2 )^(1/2)
for C26000 N = 716.43
and for Neocronos N = 1460.31

Normalizing the C26000's tracks played
0,226121184(=162 / 716,43 ) The Appleseed Cast
0,221933755(=159 / 716,43 ) Atreyu
.
.
.
0,090727636(=65 / 716,43 ) On the Might of Princes
0,089331826(=64 / 716,43 ) Incubus

and the Neocronos's tracks played
0,376632359 Still Remains
0,336229979 Jimmy Eat World
.
.
.
0,034239305 My Chemical Romance
0,032869733 Ours

The second step is to find the artists present in both profiles and multiply the normalized value for the corresponding value in the other profile, finding this way a number for each artist that represents the compatibility level for each one.

Atreyu 0,2219 * 0,2157 = 0,04786383
My Chemical Romance 0,2164 * 0,0342 = 0,00740088
.
.
.
Ours 0,0935 * 0,0329 = 0,00307615
Emery 0,0935 * 0,2479 = 0,02317865

With these values is easy to see that the most compatible artist are
Atreyu, Unjust, 36 Crazyfists, The Hurt Process And Emery.

The final step is to sum all the values and multiply it by 100.
Compatibility Rating = (0,04786383 + 0,00740088 + ... + 0,00307615 + 0,02317865) * 100
Compatibility Rating = 23.07

Original Compatibility post

Genre Compatibility

It's the same as Compatibility but it uses the users top tags instead of artists.

Chart

It's just a chart in the same way last.fm shows them but it has the advantage that you can choose any time interval.

Line Graph

It's a simple line graph that shows the ups and downs of your music listening history. It only works fine for just a couple of artists because it can get to crowded and hard to read very fast.

Wave Graph

And the best for last! The wave graphs are undoubtedly the main feature of Last.fm Extra Stats, they represent your music listening history in a very intuitive and aesthetic pleasing way. They were inspired by the work of Lee Byron . You can learn much more about the wave graphs in the FAQ section.

Original Wave Graph post