Navigation: Jump to content areas:


Pro Quality. Fan Perspective.
Login-facebook
Around SBN: The Most Dangerous Division in Sports

Overwrought Storylines

Do not ask questions about this. I don't know this.

I am lucky. Unlike Jeff, I am not compelled to listen to the national baseball media. I escape from the echo chamber. I can select who to follow and what to read at my pleasure, which insulates me from a lot of dreck. For example, I had no idea Ron Washington's decision to pinch hit with Esteban German instead of Yorvit Torrealba was at all a controversial decision until after the game when I carelessly ventured into some post-game recaps across the interwebs.

Jeff pretty much covered the reasons why German over Torrealba never raised a flag in my brain and though this post isn't about repeating those opinions and re-hashing the whole thing, I did want to throw another point out there. As far as the German was so cold because he hasn't faced a live at bat in weeks thing, isn't that an unquantifiable argument and wouldn't the manager be best equipped to judge that? What we know is what is publicly available. And the publicly available data indicates that German is probably the better hitter. The coaching staff knows (presumably) how German and Torrealba are feeling, how the two have looked during practice and a whole host of other intangible factors. As an outsider, it's foolish to propose an argument that could only be persuasively argued by an insider. Next time you want to, bring data. Discuss the history of players hitting after long layoffs. It's still not a killer point, but at least you would be on your own turf.

All that being said, a lesson to be taken here perhaps is yet another example of the folly of making snap judgments and reactions against the usual or expected. People expected Torrealba to pinch hit and not doing so was different. Nature urges us as a species to be immediately suspicious of what is different. Fight that urge. It is difficult to fight with mere willpower, but knowledge is a powerful aide. Knowing, for instance, the depth of the pinch-hitting penalty helps lessen the impulse to scream about a marginal pinch-hitting decision.

There's a bushel of other coping strategies that are useful, such as trying to play devil's advocate against yourself before making a conclusion, but for now let us stick to humbly trying to expand our baseball-related fact bank. Here is a place to ask questions, any question you like. I'm stuck indoors with a head cold, so I'll be around for a couple hours. Others here are well versed and can answer questions as well. Come with an open mind and speak up. What would you like to know more about?

Comment 40 comments  |  0 recs  | 

Do you like this story?

More from Lookout Landing

On Game 1 Of The World Series

Oct 2011 by Jeff Sullivan - 14 comments

Comments

Display:

Why has the league not imposed penalties for the Cardinals and Rangers both having the same player on their rosters?

You’re not fooling anyone, Moz/Daniels, we all know that Darren Oliver and Arthur Rhodes are the same person.

Purple Row - For all of your Colorado Rockies-related needs
Learn about Batting Metrics
Learn about Pitching Metrics

by Andrew Martin on Oct 20, 2011 2:54 PM PDT reply actions   1 recs

How much in WE is lost by an average manager over the course of a year due to suboptimal decisions?

Including everything from lineup to bullpen management to pinch hitting and bunt/hit-and-run decisions. Basically anything where the manager has time to think and make up his mind before making a decision.

M's fan newly relocated to SF My homepage

by lailaihei on Oct 20, 2011 2:55 PM PDT reply actions  

I doubt anyone has ever looked.

That’s an incredibly broad and subjective statement that would be intellectually impossible to actually accomplish.

by Matthew on Oct 20, 2011 2:56 PM PDT up reply actions  

I think certain aspects wouldn't be too hard to calculate given the right data set.

Especially:
- IBB (run expectancy before and after, taking into account other batters in the order)
- Pinch hitting (especially NL, leaving a mediocre pitcher in to hit even though they will be going through the order the third time in the next half-inning)

M's fan newly relocated to SF My homepage

by lailaihei on Oct 20, 2011 3:14 PM PDT up reply actions  

How do you know the manager runs out a suboptimal lineup

because he makes a suboptimal choice or because he’s compensating for a player or players not feeling well that day?

by Matthew on Oct 20, 2011 2:57 PM PDT up reply actions  

For this, I think you'd take the average lineup and assume days off etc are optimal.

Like Elvis Andrus batting 2nd all season long is suboptimal.

M's fan newly relocated to SF My homepage

by lailaihei on Oct 20, 2011 3:01 PM PDT up reply actions  

Oh Man

Why did Washington pinch hit German instead of Torrealba?

by jose luis on Oct 20, 2011 3:19 PM PDT reply actions  

I have so many questions.

But I feel bad asking most of them, since the answers are out there, I’m just too short on time/lazy/don’t know where to look to find them out.
I’d love to know more about the more advanced metrics and stats that are cited here for example, but I know that stuff’s out there, but I don’t have the time to really sit down and digest it.
So I guess what I will ask instead is what are the generally accepted metrics used in LL to evaluate hitters and pitchers? Fielding/baserunning is pretty much a crapshoot still, yes?
I guess it’s also harder for me at times since I never grew up with baseball and my experience with it is so limited not just in time/familiarity but in depth of understanding, especially outside of the more traditional triple slash/stats etc.

by Aussie Mariner on Oct 20, 2011 3:28 PM PDT reply actions  

It depends. (of course)

If you want a single number then
for hitters, wOBA is the standard, but OPS is a fair enough stand in when wOBA is harder to find.
for pitchers, any of the FIP, xFIP, tRA, SIERA, etc are good enough barometers.

but really, it’s about understanding what’s good, what’s bad and what’s irrelevant for hitters and pitchers. And some of that’s complicated because it’s not clear cut. It’s good for pitchers, on the whole, to get more ground balls than fly balls, but ground balls aren’t unilaterally better than fly balls. It’s good for hitters to draw walks, but not if the strategy they employ to get walks causes lots of called strikeouts and no power. Stuff like that.

Basically, you want hitters that can make contact on pitches in the zone, can hit for power when they do make contact and can lay off the pitches outside the zone. You want pitchers that can throw strikes, but also be able to miss bats and it would be great too if they kept the ball on the ground.

Help?

by Matthew on Oct 20, 2011 3:34 PM PDT up reply actions  

ERA+ is bad because it relies on ERA

but FIP- or tRA+ (or conversely, OPS+/wOBA+) are fine enough measuring sticks

by Matthew on Oct 20, 2011 3:42 PM PDT up reply actions  

Great, thanks.

At least now I have a list of things to go looking for when I finally have time to.
Last related question (I think): Is there a particularly good site(s) that explain these for newbies that you would recommend? Looking more for what it means/goes into it than the actual statistics/mathematics behind it.

by Aussie Mariner on Oct 20, 2011 3:48 PM PDT up reply actions  

Thank you.

And thanks so much for this thread, great opportunity and greatly appreciated.

by Aussie Mariner on Oct 20, 2011 4:14 PM PDT up reply actions  

Of all the falls under the penumbra of Sabermetrics, what can safely be considered dogma?

A long time ago a certain well-known baseball blogger wrote that batting order hardly matters at all. Today that blogger wrote that Josh Hamilton should not be hitting third in the World Series.

OPS+ used to be given a lot of weight, now it’s not. The same goes for UZR and quite a few other statistics.

What metrics and principals of the Sabermetrics movement would you, Matthew, feel comfortable saying will still be still be found to be as valuable in the future as they are today?

by katal on Oct 20, 2011 3:36 PM PDT reply actions  

Wouldn't that make it more important to have your best hitters getting more ABs?

I would agree that it wouldn’t matter much given a 162 game season and all, but for a short one, my instincts tell me it would be of great importance to have your best hitters getting more ABs.

by d0nkey on Oct 20, 2011 3:47 PM PDT up reply actions  

That's what I'm saying.

If Hamilton is struggling, then he shouldn’t be hitting third in a short series. My original thought was misguided.

by katal on Oct 20, 2011 3:53 PM PDT via mobile up reply actions  

To build on katal's question...

Matthew, could you talk a bit about

a) your current feelings about UZR’s accuracy and the general consensus of the analysts you respect on it?

b) the process you use now to come to a reasonable conclusion about a player’s defensive ability (i.e., which metrics do you favor the most and what, if any, non-statistical sources of information do you use)?

Thanks.

by Decatur on Oct 20, 2011 4:11 PM PDT up reply actions  

Overall, I dislike UZR.

I think it could be a useful tool, but I no longer trust it’s output and being a black box, there’s no way to restore my faith in it. I think there’s too much dogma around it as well. I’m not sure there’s anyway to adequately quantify a single player’s defensive contribution except over a vast number of years.

by Matthew on Oct 20, 2011 7:19 PM PDT up reply actions  

I personally give none

but I can see how if done comprehensively, in a PECOTA-style way, that could be somewhat useful.

by Matthew on Oct 20, 2011 4:06 PM PDT up reply actions  

I always saw player comps as a one-stop shop for saying what we expect or hope a player to turn into.

Like instead of quoting a triple slash line, describing power ceiling, and talking about strengths and weaknesses of defense it’s easier to just say “Dustin Ackley’s reasonable ceiling is a left-handed prime Brian Roberts with a little more plate discipline” without meaning that because Ackley is similar to Roberts, he should turn into Roberts.

M's fan newly relocated to SF My homepage

by lailaihei on Oct 20, 2011 4:43 PM PDT up reply actions  

I am writing a set of baseball biographies aimed at the 6-12 year old set.

Each player will get a short bio, a picture and some basic stats. Do I include pitcher Wins in the stats because they’re ubiquitous and inescapable or do I not include them so the kids don’t have to unlearn their value like we had to do?

by mkd on Oct 20, 2011 4:27 PM PDT reply actions  

I would not.

However, if they do ask, you can just say that baseball is a team sport and wins are a team effort. Nobody keeps track of quarterback wins, because it is a lot more complicated than that so keeping track of a pitchers wins is somewhat silly.

How come you can do all this other great shit, but you can't lie the fuck down and sleep?

by JAH on Oct 20, 2011 4:46 PM PDT up reply actions   1 recs

We hear a lot about needing a large sample size, or better put an adequate sample size,

for assessment of many different aspects of play. I’m curious about the other way around. How do results from an adequate sample size impose themselves on the moment?

If, for example, a manager has a choice between a left or right handed bat against a right handed pitcher, over a large sample size the left handed bat is the better choice. But what happens for the individual AB? Particularly, say, if you have a right handed bat that’s a much better hitter overall than the left handed bat?

It seems, Matthew, like you are touching on this with managers having more knowledge to work with than what we may know about. Unfortunatley, things like this are judged many times simply on result, especially in the playoffs, when a good move with a bad result may be judged as a bad move.

by PackBob on Oct 20, 2011 5:32 PM PDT reply actions  

Had to go out.

It seems to me that if left-handed hitters do better against right-handed pitchers than right-handed hitters, based on a large sample size, it’s most applicable again over a large sample size, less applicable for a single AB or a game. So if you have a pinch hitter with a single AB, it’s much less predictive of an outcome than if you had a bunch of pinch hit ABs. Over time, with a lot of pinch hit situations, a left-handed hitter would come out ahead. What I’m having a hard time grasping is what exactly this means for a single AB.

For a single AB, it just seems like there could be other factors that could override the large sample size conclusions. But also, do the odds change for a single AB compared to a lot of ABs?

I apologize ahead of time if this still doen’t make sense.

by PackBob on Oct 20, 2011 11:57 PM PDT up reply actions  

It means the same thing for a single AB as it does for 100 ABs.

Odds are odds, they don’t change based on the number of times you’re going to repeat the test. A 50/50 coin is a 50/50 coin whether you toss it 1,000 times or once.

However, when dealing with a simple sample other factors can certainly play a larger role because you don’t have the needed additional samples to help cancel out extraneous factors. That’s the whole point beyond “small sample size”. It means we haven’t seen a large enough number of tests to conclude that the differences seen are “real” and not just some combination of “other factors”

by Matthew on Oct 21, 2011 10:45 AM PDT up reply actions  

Something I've wondered about lately in regard to l/r splits on certain pitch types.

Cutters, as far as I know, tend to have little to no l/r split. Sliders on the other hand have a very large one. Aren’t cutters and sliders essentially the same type of movement, only with different speeds?
The same goes with changeups and two-seam fastballs. Two seamers have a large l/r split (I believe), whereas changeups have a reverse platoon split, being the primary weapon for pitchers to use against hitters of the opposing hand. Both move down and away from the hitter (if the hitter is of the opposite hand to the pitcher), yet two seamers get crushed.
So my question is: is there any known reason for these differences, despite the similar movement? Is it just the difference in veolicty?

by Kaorikaze on Oct 20, 2011 10:53 PM PDT reply actions  

Comments For This Post Are Closed


User Tools

By reading a game thread of your own volition you agree to accept all liability for any and all damage done to your delicate sensibilities.

FanPosts

Community blog posts and discussion.

Recent FanPosts

Small
Starlin Castro's fit with Seattle
Kawasaki80_small
Lists! So many lists!
M_s_hat_copy_small
OT -- May 22nd In Memoriam
Ichiro_small
Why do managers and media members hate walks?
Wbc_029_small
Friday Morning Music Thread
Small
Dustin Ackley BP swing vs game swing
Beastquakerwallpaper_small
More on the Struggles of Smoak
Randy2_for_sbn_small
Albert Pujols 2012: Three Retrospectives
Small
On Batting Orders
Niehaus_small
More on Dustin Ackley and the strikezone

+ New FanPost All FanPosts >

Yahoo_full_count

Sexy People

Wbc_029_small Jeff Sullivan

Small Matthew

Claw_small JY