15 Dec 2015

Statistics result 10-km run Carrera de las Empresas Madrid 2014

#48 Statistics result 10-km run Carrera de las Empresas Madrid 2014

Note 21-12-2016:
For stats (made with R and Power BI) of last edition of this race, see this post:

Yesterday, so Sun. 13 Dec. 2015, the 10K run "Carrera de las Empresas" was held here in Madrid.
In this 'business run', companies (teams of collegues) compete in 18 categories (dimensions: number of runners in team (2,3,4), gender team-members (men, women, mix) and distance (6, 10 km)). I also wanted to participate with my company, Ilunion (coparate group ONCE), like I did last year, but unfortunately this time I could´t because I was injured. For the 2014 edition of this run I made an Excel with the results but I never published it because of time problems, but I thought to finish it last sunday, so although I did´t participate, my mind was with the race..

In the 2014-edition there were in total about 9000 runners (about 300 teams) from about 800 companies, For more info about the run, see:



and for some photos:


and for some statistics:


In fig.1 you can see the result, and, as always, in the end of this post you can find the
 download-URL´s of the Excel(s).

fig.1 statistics result (finish times) 10K run

So my Excel only has statistics of the 10K run, which (netto/chip-) finish-times I got from this PDF:


As you can see in this PDF, for each runner his team is normally specified as:
Name company +´'-' + number or letter, e.g. my team was named Ílunion-26.
I wanted to create also statistics about the companies (number of runners per company), so first I transformed the data of this PDF to get the company of a runner, so for my row in this Excel the transformation was: team  Ílunion-26 -> company  Ílunion. Then I calculated in the Power Pivot datamodel, with a DAX-formula 'distinct count' the number of companies (687) derived from the field team (total: 1848), see fig.2.

fig.2: calculating data 'Company' (EquipoDef) from field 'Team' (Equipo).

And then I created a statistic 'total runners per company', see fig.3 for top 25 companies
 (with most runners). NB: I don´t know if 'TR' is a company-name or maybe a dummy-value.

fig.3: total runners per company

I wondered if there was a correlation between the number of runners of a company and the best 'total-times' of a team (of 2,3 or 4 members).  I used the category '10K - 2 (male) runners per team' to test this, see fig.4 for the result, which shows the correlation is weak (-0.15, so far for the max (negative) correlation of -1). And in the graph (scatter-plot) you can see that companies with about 30 runners or more always have a total-time (sum of time of runner 1 (best runner of company) and runner-2 (2nd best runner of company) lower then 5000 sec, which is not the case for smaller companies (in the 'bin' of companies with 2 to 10 runners, the slowest total-time is about 8000 sec.). Although in this category, the winner was a team of a company with only 2 runners, from New Balance. I guess they run on NB-shoes, so this run can be a nice way to get some free promotion...

fig.4: correlation between the number of runners of a company and the best 'total-times' of a team


#Mirror 1: Google Drive (zip files with 2 Excel and PDF files):

#Mirror 2: Microsoft Onedrive (1 Excel file):
NB: this site has 'Excel-Online', so you can view my Excel-doc if you don´t have MS Excel on your PC


#Mirror 3: Scribd.com (1 PDF file):


No comments: