THE BOOK--Playing The Percentages In Baseball

Thursday, November 15, 2007

Scouts were right, Jeremy Brown can’t run

Dan Fox checks in with his minor league baserunning numbers.

Friday, October 12, 2007

“I’m into sabermetrics” says Joel Skinner

He didn’t say it in so many words, but he may as well just have:

Well, the best way to shut down the running game is to keep them off base in the first place, but if the Red Sox *do* get on base, we’re not going to do a damn thing differently, and if Boston wants to run into outs in front of Ortiz and Ramirez, God bless ‘em.

Which is exactly why you don’t want to put speedsters in front of those guys.

Friday, September 21, 2007

Runner movements on non-steals

Dan Fox gives us some leaders and trailers for runners on non-steal, non-PA events. 

In The Book, I looked at SB, CS, PK, BK, WP, PB, DI, OA (though maybe the last two didn’t survive the final cut… don’t remember) when looking at the basestealing chapter.  I reasoned that a Pickoff is very close to a Caught Stealing, so I had to include it.  I then reasoned that a Balk is the flip-side of the Pickoff, so I had to include that as well.  Whether WP or PB had anything to do with the runner, I figured it wouldn’t matter, since given a large enough sample, if it didn’t matter, it would end up being close to zero anyway. 

A few years ago, I also looked at it in evaluating catchers from 1974-1990.  By the way, I wrote an article for the THT 2008 Annual that updates this article for all the Retrosheet years (1957-2006, minus 1999).  And I also have a leader/trailer chart for pitchers too. 

Do catchers call for more fastballs with a runner on first?

And the answer is...

Wednesday, August 15, 2007

The New Speed Scores

Pizza Cutter gives us his speed scores.

I also used speed scores in The Book, though I didn’t lay out the exact process.  It is fairly similar to what Pizza did.  One thing he did was look at pickoff throws, which is cool.  He had an earlier article on the subject, and I’ll repeat it here: he figures it as pickoff throws per “time on base”, where “time on base” is simply “1”, regardless if he goes through 1 or 3 batters at the plate.  This doesn’t make any sense.  What you really want is pickoff throws per plate throws.  But, if that’s not easy to come by, use 3.75*batters faced (or use the simple pitch count estimator). 

Regardless, it’s fantastic stuff.

Thursday, May 24, 2007

Stolen Bases in college ball

A quick look at steals in college ball over at Baseball Analysts, with my comments reproduced here:

Read More

Tuesday, April 24, 2007

Baserunning and its dependence on…

I wrote the following in resonse to Matt’s article:

There are three major reason why the baserunning rates would differ:

1 - Parks.  The closer the OF is playing, the less chance you’ll have to take the extra base.  As well, turf parks force the fielders to play a bit farther.  As we know, parks changed considerably in the last 40 years.

2 - Run Environment.  The higher the run environment, the larger the cost of the out.  However, the value of the base does not rise to the same proportion.  So, with the breakeven point higher for SB and for baserunning/taking extra base, it will be less likely that a runner will try for the extra base.

3 - Actual players.  The Vince Colemans and Willie Wilsons do not fill our league like they used to.  It could very well be that this era is filled with guys who simply don’t have the speed to take the extra base.

If you were to create a chart like this:
http://www.tangotiger.net/destmob.html
for say a group of speedsters today (Crawford, Pierre, Ichiro, etc), and compare it to speedsters of yesteryear (Raines, Rickey, Coleman, Wilson), might they be the same?  Maybe.  But, you might have a disproportionate number of those speedsters back then.

In any case, all three are plausible.

Thursday, April 12, 2007

Stealing with one out to go in the game

Black Hawk goes through the numbers for us.  The LI was a 3.7, meaning everything (on average) that happens is magnified 3.7 times more than normal.  In this case, the stolen base event is only magnified twice as much.  Typically, the win value of a SB and CS is around +.02 and -.04.  In the Ackbar case, it was double that.  So, what he attempted didn’t have the high risk/reward that the batter had.

In the minor leagues, he was a terrible percentage stealer.  179 SB and 96 CS?  He attempted to steal well over 50% of the time that he could in the minor leagues.  This guy ain’t no Vince Coleman.  Steve Lyons is more like it.

I don’t see “Game Theory” really in play, because this situation (poor but fast basestealer, with one out to go in the game) is not common enough that it’s a play that keeps the opponent wondering.

Wednesday, April 04, 2007

Jim Rice: meet Jose Vidro

Dave gets into the nitty gritty at the top of this page while I check in at the bottom.

Saturday, March 31, 2007

Throwing Over to First

PizzaCutter takes a look at throwing over to first.  The league leaders in runners who had the most pitches thrown their way is a who’s who of fast runners.  A couple of nits:

Read More

Thursday, March 22, 2007

Double Steals

The always resourceful Dan Fox comes through again, this time on double steals:
http://www.baseballprospectus.com/article.php?articleid=6003&mode=print&nocache=1174595744

This part struck me:

Read More

Thursday, December 28, 2006

Learning Things Today

My favorite feature to read is Studes’ Ten Things I Learned....  The interesting tidbit is the WPA for stolen bases.  We see that the leverage index (LI) is higher when there is a basestealing event (in the 1.2 range).  This is a great use of LI.

In the same light, you may find Morong’s article interesting, as well as the Oswalt link he provides, which shows how many wins (not runs) Tim Raines and Rickey Henderson added.  After all, while bases, out, and runs are nice and all, it’s wins that is the true currency of baseball, sport, and the financial world.

Wednesday, October 11, 2006

Making Sense of Win Expectancy and Stolen Bases

Rob McQuown has a good understanding of how to use the win expectancy table.  I wouldn’t go as low as “ to about .250-.260.” when you get to that part, maybe down to .270 as a guess.  And in THE BOOK, I do show how the breakeven point can be anywhere from 60% to 90% depending on the inning and score, so the adherence to the “average” breakeven point obviously should not be listened to.  The important takeaways from the articles are:
1 - Understand the context, don’t assume average
2 - You should be able to use some educated guesses to tweak the numbers to fit a particular context

Wednesday, September 13, 2006

Baserunning

The always informative Dan Fox gives us his list of the top baserunners.  The list itself has few surprises (Beltran, Jeter, Damon on top).  What may be interesting to some is the limited impact baserunning has, on the order of +/- 5 runs per player.  What should we have expected?

Read More

Thursday, July 06, 2006

Stealing Home

I love Carl Crawford.  Stealing home just makes me like him that much more.  How smart was the play?

Read More

Tuesday, June 27, 2006

Baseclogging

Tim McCarver likes to talk about how slow guys with a high OBP clog up the bases for the subsequent batters.  Let’s try to work some numbers.

Read More

Sunday, June 11, 2006

The Value of the Stolen Base

Going over the Dave Roberts steal off Mariano Rivera in 2004.

Read More

Saturday, June 03, 2006

Run Value of the CS and PK

Several people noted the rather large difference in run values between the caught stealing and pickoff in The Book.  Here’s the reason.

Read More