Correlation of Stats Part 1
All stats used for this fanpost are from 2008.
Each year, many teams have a huge amounted total of home runs. From the White Sox hitting 235 homers to the Giants hitting 94 homers. They are many things to factor in, such as ballparks, wind, and things of that sort. Now we will see if there is any correlation between home runs and record.
Arizona Diamondbacks - 82-80, 159HR
Atlanta Braves - 72-90, 130HR
Baltimore Orioles - 68-93, 172HR
Boston Red Sox - 95-67, 173HR
Chicago Cubs - 97-64, 184HR
Chicago White Sox - 89-74, 235HR
Cincinnati Reds - 74-88, 187HR
Cleveland Indians - 81-81, 171HR
Colorado Rockies - 74-88, 160HR
Detroit TIgers - 74-88, 200HR
Florida Marlins - 84-77, 208HR
Houston Astros - 86-75, 167HR
Kansas City Royals - 75-87, 120HR
Los Angeles Angels - 100-62, 159HR
Los Angeles Dodgers - 84-78, 137HR
Milwaukee Brewers - 90-72, 198HR
Minnesota Twins - 88-75, 111HR
New York Mets - 89-73, 172HR
New York Yankees - 89-73, 180HR
Oakland Athetics - 75-86, 125HR
Philadelphia Phillies - 92-70, 214HR
Pittsburgh Pirates - 67-95, 153HR
San Diego Padres - 63-99, 154HR
San Francisco Giants - 72-90, 94HR
Seattle Mariners - 61-101, 124HR
St. Louis Cardinals - 86-76, 174HR
Tampa Bay Rays - 97-65, 180HR
Texas Rangers - 79-83, 194HR
Toronto Blue Jays - 86-76, 126HR
Washington Nationals - 59-102, 117HR
From all the stats here, you can see that the number of home runs doesn't have anything to do with the teams record. The Sox hit 235HR while winning 89. Then the Angels hit 159HR while winning 100. The numbers here show that home runs doesn't have anything to do with winning. They just give a big show. What do you think on the correlation between home runs and record?
0 recs |
14 comments
|
Comments
For most people, they would like to see a Graph of Winning % vs HR to visualize better
and us more stat heavy nerds will definitely want to know the r-squared (stat that actually measures correlation between numbers). If you need help figuring out either of these, let me know.
by Jeff Zimmerman (TucsonRoyal) on May 28, 2009 11:47 PM EDT reply actions
Agreed
I’d like a graph and R-squared. At the very least, this list should have been in record order or HR order to better illustrate it
The artist formerly known as "mlbnotebook".
MLB Notebook.com
Inside the Majors
Roto Rat.com
by Zach Sanders on May 29, 2009 12:06 AM EDT up reply actions
Something like this? 2008 data

Beyond the Boxscore // Calling BJ Upton lazy is lazy.
by Sky Kalkman on May 29, 2009 10:25 AM EDT up reply actions
Beautiful
The artist formerly known as "mlbnotebook".
MLB Notebook.com
Inside the Majors
by Zach Sanders on May 29, 2009 3:55 PM EDT up reply actions
Which program did you use to make that graph
St. Louis Cardinals... defying win expectancy since 2008
by vivaelpujols on May 29, 2009 4:06 PM EDT up reply actions
Excel. Whatever the last version for Mac was. Scatterplot.
Beyond the Boxscore // Calling BJ Upton lazy is lazy.
Correlation will go up if you use net HR.
Or HR – HR_Against. Which, if the question is “how do home runs correlate with team wins,” I think answers it better. Of course I’m weird like taht.
When an x-axis and a y-axis love each other very much...
…they get together and they make a graph.
(Yes, I’m being snarky and uphelpful. Do you have a copy of Microsoft Excel available? That’s a great place to start.)
Also, I don't think the numbers show that homers don't mean anything
Since they do make it a whole lot easier to score runs. But they’re not the only way to score runs.
And scoring runs is only half the battle.
So homers are important, but maybe not as important as people think. Are there a lot of people out there who think home runs are that important on the team level?
I just look at team wOBA
much simpler
St. Louis Cardinals... defying win expectancy since 2008
by vivaelpujols on May 29, 2009 6:21 PM EDT up reply actions
You're going to have random variation with everything
That said, there are other measures that are going to have a much higher correlation to record than home runs. But that’s really the point of sabrmetrics.
Vogt early, Vogt often.

by 

























