Analyzing the Tampa Bay Lightning: Corsi Ratings

How good is your favorite top line, really?

Conventional stats such as goals, assists, and plus/minus ratings don't factor into the computations of the impact a hot (or cold) goalie can have on the bottom line. Who's really performing? How do we measure a players worth when they are shut out some nights? This is where a Corsi rating really shows it's value.

To the uninitiated, a Corsi rating adds shots on goal, blocked shots, and missed shots taken by both teams while any (or every) given player is on the ice. The amount of shots taken is indicative of time spent in the opposing zone, showing puck possession. If on some night Mattias Ohlund is on the ice for 22 minutes, the shots tallied for each team during his TOI is used to calculate his Corsi value. Tampa Bay gets 8 shots on goal, had 3 shots blocked against, and missed 6 more. Toronto takes 6 shots on goal, has 3 shots blocked and missed 4, just during Ohlund's ice time. Add each variable, (8+3+6=17) - (6+3+4=13) = 4. Ohlund's Corsi value for the night is plus-four.

For computing Corsi over a season, it's more plausible to use a decimal instead of an integer. So in our test case, 17+13=30, so 17/30 = .567. Pretty high, right? It was probably a Lightning win. The thing is, when we use a larger sample size, say 69 games (minus injuries), we can get a much better picture on who is really getting it done for the Lightning more nights than not. With that in mind, here is your current Tampa Bay Lightning roster:

Brandon Segal .600
Marc-Andre Bergeron .523
Bruno Gervais .514
Teddy Purcell .505
Ryan Shannon .497
Steven Stamkos .494
Tim Wallace .488
Ryan Malone .483
Victor Hedman .483
Mike Commodore .483
Vincent LeCavalier .482
Brendan Mikkelson .478
Brett Connolly .477
Martin St. Louis. 471
Eric Brewer .455
Brett Clark .453
Tom Pyatt .452
Dana Tyrell .449
Nate Thompson .444
Brian Lee .436
Adam Hall .434
J.T. Wyman .428
Keith Aulie .398

It's my opinion that Corsi reflects favorably on better players, especially over a larger sample duration, thus smaller sample sizes should be taken with a grain of salt. In other words, this is not a good metric by which to judge players such as Keith Aulie (six games) or Brandon Sagal (two games) Tampa Bay's team Corsi rating is .477, which should be a benchmark. Thanks for reading!