Groundball% for Major Leaguers
+ Now, there's proof that Roger McDowell was an extreme ground ball pitcher. Every Major Leaguer now has their GB% listed next to their stat records from 1957 to 2006. The calculation is fairly simple:
Every ball put into play was assigned as either a groundball or a non-groundball. Lineouts, popouts, foulouts and flyouts are considered to be non-grounders.
Otis Nixon hit 57% of the balls he put in play on the ground. Imagine how effective he might have been had he aimed for a 65% gb%.
Kirby Puckett started his career as a very good ground-ball hitter but evolved into a flyball hitter once he became a power hitter.
Brady Anderson was a poor man's power hitter until 1996 when he clubbed 50 homers, twice as much as his next best total (24). 37% of his hits were in the air that year. His career average was 45%.
You would think that Bert Blyleven would have an extremely low gb% the season he gave up 50 homers (1986). Strangely enough, his gb% was 53 which suggests that when the ball was hit in the air, it went pretty far a lot of the time.
Just another dimension to the Cube's data.
+ I noticed that Baseball-Reference got into the Minor League Statistics game. I can't say that I am ecstatic about the competition but its a smart move on his part. As always, we'll both coexist and offer the baseball world a slightly different view of the same data. And we'll continue to make baseball the most advanced sport on the Internet.
8 comments:
nice work. the gb% is very interesting.
i agree it's nice to have minor league stats on baseball-reference but i find it annoying to do a search for a player and then have to choose between his major league stats and minor league stats (one more step...it adds up). i much prefer how you have it on the same page.
actually, i would prefer major league stats on the player's main page with a link that gives you both, like you have.
This is pretty awesome. Did you get this from retrosheet? I didn't know they had batted ball type for the older seasons.
I checked an old season, 1982, against retrosheet. My GB% are a little different from what's on the site. Retrosheet has batted ball type for plays that result in outs, but its usually blank if the play results in a hit.
Did you find another source to get batted ball type for hits?
The event files have information on all batted balls and whether they were hit on the ground or in the air.
its not just outs.
Gary, that's not the case for all years. In Retrosheet, 2000-2002, they are missing a substantial portion of batted ball types.
Do a year-by-year, and you'll see the gaps. This usually points to the different main source that Retrosheet went with that year.
While I'm hear, your position info of Ichiro in Japan is wrong:
http://www.thebaseballcube.com/players/S/Ichiro-Suzuki.shtml
As you can see by the research here:
http://marinerds.blogspot.com/2006/10/truth-about-ichiro.html
For 1982 American League, there were 14,474 singles. Batted ball type shows 3 popups, 101 lines, 483 groundballs, and 6 flyballs. The rest have blank for batted ball type.
There are some you can make assumptions on, use 'fielded by' to see which stay in the infield and almost always going to be groundballs, but for the vast majority of these hits there's no way to tell if its a grounder or not.
I suggest starting a blog or message board for people to report data errors.
A few are Doug Jennings, 1985: 81 hits, 19 at bats.
1987 Wichita has two batting lines for Roberto Alomar - one of them has to belong to Sandy. I'll go out on a limb and guess the one with 43 steals in 44 attempts is Roberto's.
Post a Comment