Baseclogging
Last year, when there was NOT a man on on second, runners on first scored 42.6% of the time on a double. When there was a guy on second, runners from first scored 45.3%. You read that right, there is no such thing as clogging the bases.
about 2 years ago
The Cheat
57 comments
1 recs |
Email
PrintComments
hey i don't adhere to this base clogging philosophy.
I would have brought back Thome AND Dye. FPK though.
Kenwo4life=ratings
i deny this evidence and repute it's existence. i will continue to hold onto my own beliefs.
/dusty baker and the catholic church
Cashing checks and having sex.
I don't think you understand what "baseclogging" is.
What you’re describing is clogging the basePATHS. Baseclogging refers to the clogging of the base itself.
by The Actual El Guapo on Mar 26, 2010 12:53 PM CDT reply actions
baseclogging
runs/(times on 1b mark mcgwire)
minus
runs/(times on 1b carl crawford)
-————————————————————
=degree of baseclogginess
i’d guess its about 10-20%
crazy shit
mcgwire scored roughly 27% of the time he got on first
crawford scores 55%
rough math
39.4% of mcgwire’s ABs came with 2 outs as opposed to 28.4% for crawford.
and that’s just one point i’d make. it’s much.
alright. give him 10%
still a huge difference.
you seriously want to argue that slow guys score as often as fast guys?
of course not.
perhaps you were simply making a facetious point. if so, apologies. otherwise, your methodology is really bad.
i assume dave cameron
includes his baserunning in WAR? how does he figure that out? stopwatch?
amazingly, he's written articles about such things.
you’re not exactly breaking new ground here. this stuff has been studied to death. see, e.g.:
http://www.insidethebook.com/ee/index.php/site/comments/baseclogging/
yeah, but is it in WAR?
which has become de rigueur? i’m asking cuz i don’t know. more rigorous studies show the difference b/t bad and good basrunners is 10-15 runs/year.
historical guys, we don’t have detailed stats about how many times a guy advanced on ground balls, etc. but you compare guys with 5000-10000 plate appearances, differences in where they hit and who hit behind them should diminish in importance.
here’s another guy: dave winfield. hit in the middle of the order like mcgwire. played on pretty good teams. a somewhat similar hitter. only he was pretty fast. scored 38% of the time.
why would sample size
erase batting order differences? This only makes sense if the players concerned have their batting order locations randomly distributed so you would need a certain about of samples to even out the randomness.
however, this is not true. Crawford will spend the vast majority of his career leading off and having sluggers hitting behind him, while McGwire hit in the 3-4 holes and didn’t have the benefit of high powered guys behind him.
yeah
Thus the winfield comp. It was just an offhand comment to show that speed matters.
We don’t have historcal baserunning data. Have to use what is available.
by Hatchetm on Mar 29, 2010 4:42 PM CDT via mobile up reply actions
wasn't there a point where
caught stealing weren’t recorded? I feel like there is a question about Cobb’s base stealing efficiency.
if we're going to have an honest discussion
about who was the more valuable racist, we’ll need to properly account for Cobb’s baserunning advantage to counteract Anson’s slugging advantage.
I wonder if BR has a ‘racist level’ sortable column.
by coffeepac on Mar 29, 2010 5:19 PM CDT up reply actions 1 recs
cobb's got 60 WAR on anson.
even if we assume that anson strung up six negroes, that would only be worth 39 more WAR (era-adjustment of 6.5 WAR/negro).
by larry on Mar 29, 2010 5:28 PM CDT up reply actions 1 recs
Gotcha.
I thought that was a bit of a weird point. However, you still shouldn’t say that the increase in at-bat samples decreases the importance of lineup location.
And no one says speed doesn’t matter at all. What people are really saying is that speed isn’t as important as getting on base/hitting the ball far in the first place. The best to worst hitter is a range of a hundred runs a year (10 wins). Whereas best to worst base runner is 10 to 15 runs (1-1.5 wins).
So with that as the backdrop, the person who is a better hitter relative to league average is more valuable to the person who is a better baserunner relative to league average.
That’s why, for the most part, speed doesn’t matter.
this is not making much sense to me.
yes, it’s in WAR.
yes, the difference is about 12 runs a year.
no, as i pointed out in your original example, more plate appearances does not diminish in importance.
alright, i revamped my analysis...
this time i only gave 40% credit for scoring after hitting a double, 20% credit for scoring after hitting a triple (obviously no credit for scoring on a homer). 100% credit for single, walk, hbp. the numbers come out a bit different. (you could quibble with these discount rates, but 2B and 3B are rare compared to the overall on base, they don’t make a big difference).
looked at five guys who spent considerable time at leadoff and #3 (boggs, raines, molitor, madlock, downing).
when batting #1, they scored 26%
when batting #3, they scored 23%
(not that much of a difference, huh?)
looking at 2009, the average AL leadoff man scored 26%
the average AL #3 man scored 19%
#3-8 are roughly the same 17-19%, the #9 and #2 score 24% (why? i do not know).
it appears that batting order and baserunning ability combine to account for the difference in scoring percentage (duh).
looking at mcgwire, you have 17% (mostly batting #4)
crawford 26% (not as good as raines 28%)
supposing a guy gets on via walk, single, hbp about 200 at bats, the difference between mcgwire and crawford is roughly 9%. let’s adjust 4% for batting order differences. so crawford is about 5% better at scoring runs than mcgwire. given roughly 200 times on base in a year you get in the neighborhood of about 10 runs.
pretty simple analysis that is roughly what we would expect given other analysis using PBP data to estimate runs.
I would explain all of those differences with batting order position
Your best hitters are going to be in the 2-3-4 slots. The people who score the most runs are in the 9-1-2 slots. Which implies your 2-3-4 hitters are driving in your 9-1-2 hitters.
And while you’re using runs, that isn’t the same type of runs that WAR uses. You’re using context dependent runs, while WAR uses context neutral runs.
In your accounting, you’re hoping luck is neutral over the length of your samples, which is probably foolish. WAR, or whatever the baserunning component is, will assign, say, .6 runs for a batter going from 1st to 3rd on a single no matter if he scores. Your system is hoping each batter will score 60% of the time he goes 1st to 3rd on a single.
If your samples are large enough, then yes, you will have luck neutral numbers, but your sample of 5 people to determine batting order affect on runs scored is probably insufficient. Especially since you have several hall of famers in your sample. Note: that makes it not representative.
it's not perfect, but like 25,000 ABs is a pretty big sample size
i could do 10 guys and 50,000 ABs, but i’m satisfied that baserunning matters….as you say 10-15 runs per year between a good one and a bad one.
now, as far as i can tell WAR as reported in Fangraphs does NOT include baserunning. (see http://www.fangraphs.com/blogs/index.php/runnin-the-bases-part-1).
this means we need to be mindful of adjusting WAR in certain instances.
Its the headcount
not the plate appearances count. 5 or 10 guys is probably not a sufficient size. Also, you should be using some kind of third variable to prove that you are making an apples to apples comparison.
You used the same players and how they performed in different lineup positions (good!), you didn’t clarify their ages (bad!). What if those players started their careers in the 1 hole and moved to the 3 hole after they started to slow down. Which might mean that the 4 point drop you saw in their weighted scoring percentage is entirely due to speed differences, and not lineup position.
I don’t know what the third variable would look like (seasonal age? steal%?), but its called the independent variable and it needs to have the same value for each bucket to make a comparison, then you can calculate conversion rates between independent variable buckets AND THEN you’re a giant nerd who has some hard data to start basing lineup effect theories on.
And yes, we should adjust WAR for baserunning when comparing extreme players.
your adjustment would support my theory though.
as they get older and move down in the lineup, they should become worse baserunners, but in my limited dataset, they don’t.
to do it properly would be a huge pain in the ass. you’d have to take every guy (100? 200?) and correct for an expected age decline. i don’t have computer programming skills and a raw database to do it easily. i’d have to pull each guy out separately and set up a huge spreadsheet. it would take a couple of days.
i honestly don’t care about reinventing the wheel. i’m not going to present at a sabr convention. as larry said, this has already been done. it would be nice to have a historical database though, which i don’t believe exists.
for crissakes
i see it. right here. for projections it includes a variant of bill james speed score. but that includes some defensive stats, so i don’t know WTF is with that. uzr already includes defense.
the historical data appears not to include it. how many runs was lou piniella’s baserunning worth in 1978?
face in hands.
all right. let’s walk through this. click on the link:
http://www.baseballprojection.com/war/p/pinil001.htm
look at the column for bsr. click on the link for “stat definitions” at the bottom. some like to use a few of the other categories to detail baserunning, like roe.
aha. thanks
This is estimated from a regression formula for seasons without play by play data.
presumably that goes back to the 50s? if so that’s pretty cool. thanks for not telling me this yesterday as i learned some interesting stuff that i wouldn’t have otherwise.
i love the rally WAR
pierre was better than thome last year.
how did i know you were going to say that?
you are correct. so why does everyone have their panties in a bunch over the loss of thome?
He was. Today.
"Enough is enough. This has to change. This was not supposed to happen five or 10 years ago, it is supposed to happen now. Today is my day, this is my time, and this is my moment."
Every time I see this on the right rail I read "Log Jammin'" instead of base clogging.
But anyway it should be called “larry teaches Hatchetm and coffeepac a thing or two”
by e-gus on Mar 30, 2010 3:54 PM CDT reply actions 1 recs
Money he's got.
He likes the blowjob. Or so I’m told. You’re his favorite. Or so I’m told.
"Enough is enough. This has to change. This was not supposed to happen five or 10 years ago, it is supposed to happen now. Today is my day, this is my time, and this is my moment."
I can separate his message from his madness.
"Enough is enough. This has to change. This was not supposed to happen five or 10 years ago, it is supposed to happen now. Today is my day, this is my time, and this is my moment."
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on South Side Sox
Related FanShots on South Side Sox
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
FanShots
Quick hits of video, photos, quotes, chats, links and lists that you find around the web.
Recent FanShots
In Association With
Big League Stew - MLB - Yahoo! Sports
- Local radio host honors Cubs’ Tony Campana with parody song
- Roy Halladay bobblehead with glove on wrong hand selling on MLB.com
- Will Smith shows why making your big-league debut at Yankee Stadium is less than ideal
- Steinbrenners selling the Yankees? Family denies New York Daily News report
- The Juice: Brandon Phillips flip sets up Todd Frazier walkoff for Reds
