Tennis Analysis Ariana Zuberovic BDA 761 Historical data from the - - PDF document

tennis analysis
SMART_READER_LITE
LIVE PREVIEW

Tennis Analysis Ariana Zuberovic BDA 761 Historical data from the - - PDF document

12/17/2015 Tennis Analysis Ariana Zuberovic BDA 761 Historical data from the ATP Association of Tennis Professionals Network of mens tennis players formed by tennis players Years 1968-2015 Attributes include:


slide-1
SLIDE 1

12/17/2015 1

Tennis Analysis

Ariana Zuberovic BDA 761

Data

  • Historical data from the ATP
  • “Association of Tennis Professionals”
  • Network of men’s tennis players formed by tennis players
  • Years 1968-2015
  • Attributes include:
  • Tournament Name
  • Ex: US Open
  • Surface
  • Ex: Grass
  • Score
  • Ex: 6-3 6-2 6-4
  • Player name, country and ID number
slide-2
SLIDE 2

12/17/2015 2

Data Source

  • Historical Data, Open Data
  • Available on GitHub
  • JeffSackmann/tennis_atp
  • Yearly .csv file
  • Data Prior to 1983:
  • Missing Ranks (intermittent)
  • Missing Columns:
  • Number of Aces
  • Minutes Played

Why Tennis Data?

  • Why has the popularity of tennis in America been facing

a downward slope?

  • Suspected reasons:
  • 1968 - "Open Era” Prior to this tennis players could

not make a living through tennis.

  • “Rich-man's country club sport”
  • 1968- is that the US Open was broadcast on television

for the first time (by CBS)

slide-3
SLIDE 3

12/17/2015 3

What about it?

  • Analytics of the data by decade
  • Winning Countries?
  • Nationalism has been linked to popularity among

tennis, where has America fallen?

  • Attributes of the top countries?
  • Ranks
  • Grand Slams
  • Surface
  • Which surfaces do the top countries dominate?
  • Age
  • When is a players peak performance?
  • Making of a top country for the next decade.

Inaccuracies with the Data Sets

  • Missing Data
  • Dates of the tournaments
  • Inaccuracy: Each match date is recorded on the same

day, of course this is not possible as matches are played regularly for a series of weeks

  • Columns are not completely identical
  • Can be an issue when combining data sets
slide-4
SLIDE 4

12/17/2015 4

Organization

  • Decade Data
  • 1970s
  • 1980s
  • 1990s
  • 2000s
  • Tableau
  • Extract and append data from separate files
  • Ensure: Data Columns are Identical
  • Calculated fields for
  • Weighted number of Winning Countries
  • Restricted Ranking Numbers

1970s-Matches Won

slide-5
SLIDE 5

12/17/2015 5

Number of Country Wins by Decade

1 9 7 s 1 9 9 s 1 9 8 s 2 s

1970s-Number of Players

slide-6
SLIDE 6

12/17/2015 6

Number of Players by Decade

1970s 1980s 1990s 2000s

1970s-Number of Players

slide-7
SLIDE 7

12/17/2015 7

Number of Players by Decade (Yearly)

1970s 1980s 1990s 2000s

1970s-Weighted Number of Matches Won

slide-8
SLIDE 8

12/17/2015 8

Weighted Number of Matches Won by Decade

1 9 7 s 1 9 9 s 1 9 8 s 2 s

GRAND SLAM WINS

1970-2000

slide-9
SLIDE 9

12/17/2015 9

1970s-Weighted Number of Grand Slam Wins 1980s- Weighted Number of Grand Slam Wins

slide-10
SLIDE 10

12/17/2015 10

1990s- Weighted Number of Grand Slam Wins 2000s- Weighted Number of Grand Slam Wins

slide-11
SLIDE 11

12/17/2015 11

ATP RANK:WHICH COUNTRY COMES OUT ON TOP?

1990-2000 Data only

1990s-Weighted Highest Presence in the Top 4

slide-12
SLIDE 12

12/17/2015 12

2000s- Weighted Highest Presence in the Top 4

2000s-American Presence at the Top 10 Rank

slide-13
SLIDE 13

12/17/2015 13

Two Decade Rank Comparison

2 s 1 9 9 s

WHO DOMINATES ON CLAY? HARDTOP? GRASS?

1970-2000

slide-14
SLIDE 14

12/17/2015 14

1970s 1980s

slide-15
SLIDE 15

12/17/2015 15

1990s 2000s

slide-16
SLIDE 16

12/17/2015 16

DOES AGE MATTER?

1970-2000

Age of peak performance (Yearly)

1970s 1980s 1990s 2000s

slide-17
SLIDE 17

12/17/2015 17

Conclusion and Winning Country Profile

  • Earlier decades: More variation in winners of each attribute
  • USA:
  • Downward slope for number of players
  • Are not making the top 4 rankings
  • Not winning as many matches
  • Future analysis
  • Predictive Analysis- What is in store for 2010s?