« Baseball ROI 2: scatter plots | Main | One for the cutting room floor »



The Pappas efficiency metric heavily rewards the first few wins above .300, i.e., every win just after 48 improves efficiency a lot. Using the Pappas metric, KC looks worse than the Yankees but even one additional win gives a big efficiency jump. The Pappas metric isn't very robust to small differences in the number of wins:

There are lots of alternatives. Here's one of them:

Of course, this alternative also suffers a flaw: in reality, the Royals really do suck.


Robert, your first graph would be great if we'd like to use scatter plots to divide the teams into segments using the Pappas metric.

How do you define your 2nd metric? I notice the lines are curved.


If you use the payroll/median payroll and win/loss ratio for the axes (instead of the team payroll and the projected wins), the lines are straight. However, the comparison between two different metrics is easier when the plot axes are the same and the contours are different than the other way 'round; try it the other way if you'd like to see why.

Using scatterplots to show X, Y, and f(X,Y) is a pretty handy technique that I think is underutilized. Here are a couple of other (sports-related) examples:

BTW, the baseball plots let you see payroll, wins, team name, league, and efficiency (for a particular metric).


Should've clarified for those who have a parochial view of sport that those are from the Tour de France. Here's yet another damn example of using a scatterplot to show X, Y, and f(X,Y):

The comments to this entry are closed.

Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR.

See my Youtube and Flickr.

Book Blog

Link to junkcharts

Graphics design by Amanda Lee

The Read

Keep in Touch

follow me on Twitter