The Giants QB Eli Manning is in the news for the wrong reason this season. His hometown paper, the New York Times, looked the other way, focusing on one metric that he still excels at, which is longevity. This is like the Cal Ripken of baseball. The graphic (link) though is fun to look at while managing to put Eli's streak in context. It is a great illustration of recognition of foreground/background issues. (I had to snip the bottom of the chart.)
After playing around with this graphic, please go read Kevin QuigleyQuealy's behind-the-scenes description of the various looks that were discarded (link). He showed 19 sketches of the data. Sketching cannot be stressed enough. If you don't have discarded sketches, you don't have a great chart.
Pay attention to tradeoffs that are being made along the way. For example, one of the sketches showed the proportion of possible games started:
I like this chart quite a bit. The final selection arranges the data by team rather than by player so necessarily, the information about proportion of possible games started fell by the wayside.
(Disclosure: I'm on Team Philip. Good to see that he is right there with Eli even on this metric.)