THE BOOK cover
The Unwritten Book is Finally Written!
An in-depth analysis of: The sacrifice bunt, batter/pitcher matchups, the intentional base on balls, optimizing a batting lineup, hot and cold streaks, clutch performance, platooning strategies, and much more.
Read Excerpts & Customer Reviews

Buy The Book from Amazon


SABR101 required reading if you enter this site. Check out the Sabermetric Wiki. And interesting baseball books.
MOST RECENT ARTICLES
MAIL : You ask | We say

Advanced


THE BOOK--Playing The Percentages In Baseball

<< Back to main

Thursday, June 17, 2010

The next Ensberg challenge

By Tangotiger, 09:49 AM

He asks:

Scenario: You are the runner on 2nd base and there is 1 out.  The score is 3 to 3 and it is the 4th inning.  A fly ball is hit to right center field and the right fielder will attempt to catch the ball.  As a runner you have average Major League speed.  If you choose to tag, you will be safe at 3rd base, but you will have to slide. The next batter up hits in 7th in your lineup and he is a career .250 hitter with 10 HR’s and 60 RBI’s. What do you do and why?

My answer:


Here are the possible outcomes:
a. RF catches ball, you tag up, you make it to 3B, 100% of the time
b. RF catches ball, you are somewhere between 2B and 3B, you need to get back to 2B, and will be thrown out there proportionate to how close you were to 3B
c. RF just misses getting the ball, you tag up, you make it to 3B, 100% of the time
d. RF just misses getting the ball, you round 3B, the RF throws it in, and you attempt to score proportionately to how close you were to 3B when the ball dropped in, and you will be thrown out a fixed percentage of your attempt rate (say 10%)

Now, give me about 10 minutes to look into my WE charts, and I’ll give you my answer.

UPDATE:
Presuming bottom of the 4th, the chance of winning to start that plate appearance is .5832.  Let’s do it:

a. Tag up: Runner on 3B, 2 outs makes it .5466 win probability

b. Take a lead:
(i) Runner on 2B, 2 outs makes it .5412 win probability
(ii) DP makes it .5000

c. Base hit: Runners on corners, 1 out makes it .6384 win probability

d. Base hit:
(i). You stick at 3B, so same as c. above
(ii). You try to score, the batter-runner holds at 1B, and you are:
1. Safe at home: Up by 1, runner on 1B, 1 out: .6824
2. Out at home: Runner on 1B, 2 outs: .5291
(iii). You try to score, the batter-runner goes to 2B on the throw, and you are:
1. Safe at home: Up by 1, runner on 2B, 1 out: .6983
2. Out at home: Runner on 2B, 2 outs: same as b(i) above

Ok, you need to have all those numbers in your head when you have to make your decision!

The difference between a. and b. above is so small that on a play that you think is a sure out, that it doesn’t really matter if you tag up or not.  That is, being on 2B with 2 outs and being on 3B with 2 outs is almost the same thing.  (That’s because the SF is now out.) So, tagging up is not really an option, not unless you are 100% sure that the RF will make the out.

Now, the tougher part: you are not sure if the RF will make the out.  If you take a big enough lead, you will score easy, and your win probability is around 68% to 70%.  If you take a good lead, but not big lead, you might get thrown out at home, in which case the win probability is 53% to 54%.  If you play “station-to-station” ball, the win probability is 64%.

So, let’s try to work it out, presuming it’s a 50/50 play (you are not sure if the RF will make the out or not).  You take a good lead (say halfway), but not big lead.

Batter is out, you make it back 45% of the time, and are doubled up 5% of the time.
Basehit, you score 40% of the time, and are thrown out 10% of the time.

That’s a .595 win probability.

If you simply played station-to-station (half the time a., and half the time c.), it would have been .593.

As you can see, you have to take a lead such that you have at most a 10% chance of being doubled up on a catch, and at most a 20% chance of being thrown out at home.  Sounds therefore that you should be about one-third to half-way to 3B.

That’s on a 50/50 play.

#1    Tom N.      (see all posts) 2010/06/17 (Thu) @ 10:13

I think assumption b is a little off. Under this assumption, someone who is 10% of the distance between 2B and 3B will be thrown out 10% of the time. But 10% of the distance is only 9 feet. I would think the runner would get back 100% of the time if he’s only 9 feet from the bag, especially if the assunption is that the ball is hit deep enough so that he could get successfully tag up 100% of the time.


#2    Tangotiger      (see all posts) 2010/06/17 (Thu) @ 10:41

When I mean “proportionate”, it doesn’t mean “linearly proportionate”.  I just mean that the farther off you are, then the more chance of being doubled-off.


#3          (see all posts) 2010/06/17 (Thu) @ 10:42

I think that your answer fits very well with the conventional wisdom.  When a ball is hit in the air to right field, the runner is taught to go halfway(ish) and judge whether or not the RF will make the catch.  If he is virtually certain to make the catch, you tag up and try to force a throw.  If you’re unsure, you stay halfway and await the outcome.


#4    Peter Jensen      (see all posts) 2010/06/17 (Thu) @ 11:34

Tango - Your analysis above is ignoring the lineup information that is given in Ensberg’s problem.  Since you have the lineup information, you’d get a more accurate result be computing the likely runs scored in the inning using lineup specific RE tables and then comparing the resulting WEs for those run numbers at the end of the inning.


#5    Tangotiger      (see all posts) 2010/06/17 (Thu) @ 11:42

I have a WE table for 4 RPG (and 3.5).  I can rerun doing those, but I would say I’m not going to get much different answers.


#6          (see all posts) 2010/06/17 (Thu) @ 11:51

I think you also discount the frequency of fly balls being caught vs. dropped.  You’re assuming it’s a 50/50 play, when as far as I can tell it’s typically like a 98% chance of an ordinary fly ball being caught


#7    Tangotiger      (see all posts) 2010/06/17 (Thu) @ 11:54

Bookworm: I never said anything about an ordinary flyball (and if Ensberg implied it, then the answer will be boring). 

I was talking about a flyball where the runner is 50/50 on whether the RF will make the play or not.


#8    bsball      (see all posts) 2010/06/17 (Thu) @ 12:41

Tango #5:
I think Peter was referring to the hitters due up next being 7, 8, 9 in the order and thus being below average (especially if you’re talking Astros from when Ensberg played). The run environment for the game might be 4 or more, but the 7, 8, 9 hitters may make the RE for the current situation lower than the tables you used and therefore make you want to try to score with higher % of being caught.


#9    Morgan Ensberg      (see all posts) 2010/06/17 (Thu) @ 14:16

That’s it!  We all have to get together for beers!  I love math!  But I will not reveal what I think just yet…

MorganEnbserg.com


#10    Tangotiger      (see all posts) 2010/06/17 (Thu) @ 15:09

bs/8: My original numbers used a 5RPG environment.  I’m guessing the 7,8,9 hitters would be a 4RPG environment.


#11          (see all posts) 2010/06/18 (Fri) @ 09:01

wouldn’t it matter if it was bottom of or top of the inning?


#12    Tom N.      (see all posts) 2010/06/18 (Fri) @ 09:02

Tango #2: ah ok, I misunderstood what you wrote. I thought you meant you would do it linearly.

btw, the fact that Morgan Ensberg is posting here is pretty awesome (I’m assuming that really was Morgan)


#13    Tangotiger      (see all posts) 2010/06/18 (Fri) @ 09:03

It’s the 4th inning, so probably not.

Usually the top/bottom parameter becomes an issue in the 7th or later innings.


#14    Davor      (see all posts) 2010/06/20 (Sun) @ 09:20

Information on fielder and his position has not been analyzed enough. Ball is going to right-center and fielder can throw to third in time “I” need to run some 85 feet (I’m safe at third with sliding, meaning throw will beat me a bit before I run 90 feet). I’m taking that right-center means that third-base angle in triangle 2B, 3B, fielder is small, maybe 10 degrees or so. That means that distance between the fielder (when he catches the ball) and 3B is almost the same as the sum of distance between the fielder and 2B and 2B and 3B. That means that on any ball caught less than 100 feet from 2B I’m out at 2B more often than not. For balls that are caught further out, more than 200 feet from 3B, distance between fielder and 3B is close to one between fielder and home, so if I try to advance on cleanly fielded ball from halfway towards 3B, I’m comfortably out.
Based on this, not waiting on (or around) 2B is advisable when either:
a) fielder with Damon-like arm is fielding the ball relatively close to 2B, maybe 50 - 60 feet out, and it’s less than 20% chance he’ll get it, or
b) ball is fielded very deep, and there is a chance that fielder loses a few seconds fielding the ball.
In all other instances risk of being doubled/thrown out at home is too great.


Page 1 of 1 pages


Name (required)
E-Mail (optional; WILL be published)
Website (optional)

<< Back to main


Latest...

COMMENTS

Feb 12 00:40
Clutch analogy

Feb 12 00:38
Reader Mail of the Day: Why do we need X years of fielding data?  And what about outliers?

Feb 11 22:08
Who is Jeremy Lin?

Feb 11 20:11
Fighting leads to goals?

Feb 11 19:55
Why do players get crappy caps?

Feb 11 19:12
Hero of the month: Brittney Baxter

Feb 11 17:59
MGL: Today on Clubhouse Confidential

Feb 11 10:29
Dwight Evans

Feb 11 02:12
Performance through the ages

Feb 10 23:01
For Your Soul