Click here to see the SAS code.
Click here to see the example.

---------------------------------------------------------------

This is a SAS/Graph imitation (and enhancement) of the following BonavistaSystems Win/Lose Chart:

   soccer_orig.jpg
   http://www.bonavistasystems.com/OnlineDemoReports.html

I copy-n-pasted the soccer data into an Excel spreadsheet, and read the 
data directly from there into SAS.

   http://www.soccerstats.com/allmatches.asp?league=england


I look at the score, and determine whether each team won, or lost, each game
(and I ignore 'ties', since this is a win/lose chart).  If the team won a 
given game, I use annotate to draw a line in the "upward" direction, if they
lost, I draw in the downward direction.

The basic chart is pretty simple, but I get a little tricky to get the 
y-axis and the text stats (rank and number of wins) in there, and to 
order them based on the number of wins.

-----

Enhancements:  (better than the original)

You can hover your mouse over the win/lose ticks, and see the team names
and scores for that game.

I use an actual date axis, and space the ticks proportionally by the date.
This allows you to see visual trends that you might not have seen with 
the equally-spaced ticks.

I used a better color-scheme, which was more intuitive, and less harsh
on the eyes.

I added a url/link to the footnote, so you can click on it to go to the 
soccerstats.com website.

Also, instead of using the alternating gray/white bars behind each 'row'
of data in the chart, I use a simple line down the middle of the 
win/lose ticks, connecting them and pointing to the rank/team/wins
text on the left.  I think this uses "less ink" and is less "busy",
and makes the actual data of the chart stick out more.

In addition to the wins & losses, I also show the 'ties' (using a 
subtle gray circle).  Without showing the ties in the win/loss chart,
it looks like there are big gaps where certain teams had no games,
when in fact they did have games (but just tied, rather than win or lose).
Now the gaps truly represent gaps in time where there were no matches.

Back to Samples Index