THE BOOK--Playing The Percentages In Baseball

Is “Mailbags on Parade” supposed to be a takeoff on the song by Rage Against the Machine “Bulls on Parade”? If so, cool. I’m kinda old now, but at least I’m an old rocker (guitar whiz, composer, recordings, etc.) The live version of Bulls on Parade from their 2003 live DVD is awesome.

Or, maybe it has nothing to do with that.

Tom, I think you’re combining the people who comment here who go by “dan.” Not that it matters at all, but there’s at least two commenters who use it. One has a blog (clutchiness) and one doesn’t (me).

If I were a true artist, I’d let you believe whatever you like!

However, it is a takeoff on Al Michaels opining of the ESPN “analysts” as: Gasbags on Parade.

***

Yeah, dan, I don’t understand why people don’t use unique handles.... that’s what a name is for!  Posting as “dan” really makes it impossible to do searches, etc.  I can look for:
tango baseball
and I know I’ll find things almost only about me.

...until Dancing With The Stars adds a former MLBer.

I have also posted several times on this blog, but never (or seldom) under the same name twice.  Not sure if that makes it “better” or “worse.”

Anyway, cool idea.

If I had anything important to say than I’d probably think of something more unique

I have added a link at the top right that says “Contact Us”.  This way, you don’t have to look for this thread.

Once we start answering questions, I’ll let you know where to go.  But, here’s the sneak peak (questions, no answers):
http://www.tangotiger.net/wiki/index.php?title=Mailbags

I’ll have that link in the “Contact Us” page as well, once we answer questions.

Though, maybe instead of “Contact Us” I should put “Mailbags” or “Mailbags on Parade”.

I saw Rage’s last Philadelphia concert, which is particularly fun considering their song “(expletive) the Police” is directed towards Philadelphia cops.

Saw their reunion show at Rock the Bells this summer, too, and met actor Paul Dano; boy was that awkward.

Love this feature Tango.  I also love typing my name into Google and finding that someone registered me on the website manpaper.com (where people can post nude photos of celebrities), due to me being on IMDB.

Gotta love the wiki.  Makes like easier.

Anyway, mailbag Q&A is now up-to-date.  When there’s a new Q&A, I’ll let you know.

Mailbag now updated.

I added my thoughts within MGL’s reply.

More replies.

More…

Are you fielding reader mail for Bill James now, too?!?

That was actually posted on the Freakanomics blog, and James didn’t answer those.

More questions answered…

In regards to Rule Changes - DH, the DH is covered by Rule 6.10.  It is a League decision as to if the DH is allowed in home parks.  However, a team may choose not to use the DH.

Mailbag updated…

Mailbag updated (and questions outstanding too).

MGL and AED added as well to an older question here:
http://tangotiger.net/wiki/index.php?title=Mailbags#Betting_accuracy

Mailbag was updated on “2 June” (search for that).  MGL has 3 responses…

You (not sure which of “you” since the response wasn’t labeled) said the following in response to a question about whether defense is better behind quick-working pitchers:

They found no correlation between error rate and time of game, which did not surprise me in the least. If a more rigorous study was done (you’d probably have to have someone make a list of fast and slow working pitchers, because there is nothing in the data that tells you that other than time of game), I would be surprised if there were any significant correlation between the speed at which a pitcher works and the quality of the defense (however you want to measure it) during that game.

In fact, the PITCHf/x data for 2008 includes a time stamp down to the second.  Without knowing for sure what event has its time recorded, I would guess this time stamp corresponds pretty well to when the pitch was thrown.  I could inquire of Sportvision/MLB if someone wants to know.

For example, Octavio Dotel’s pitches to the Royals in the 14th inning last night are labeled as follows:
http://gd2.mlb.com/components/game/mlb/year_2008/month_06/day_04/gid_2008_06_04_kcamlb_chamlb_1/inning/inning_14.xml
to Buck
080604_230645
080604_230705
080604_230728
080604_230748
to Gathright
080604_230823
to Pena
080604_230908
080604_230931
080604_230958
080604_231025
080604_231045
(coaching visit to mound)
to DeJesus
080604_231152
080604_231212

Also, there is a question about bunting with two strikes.  That made me realize that I had probably never published my stats on ball/strike types broken down by count from 2007.  I had intended to do that but got distracted by other stuff.

I don’t have all the pieces of data in front of me here, but the bunting frequency by strike is as follows:
0 strikes: 1.3% of pitches result in bunt attempts
1 strike: 0.9%
2 strikes: 0.13%
Those numbers include pitchers and position players together.

The success rate, if you define success as base hit or sacrifice, is around 40% for all the counts, possibly slightly lower for 2 strikes but not by a large amount.  I would need to refine the success rate numbers a little bit by looking for bunts in the fielder’s choices and reaching on error and so forth.

I contacted Sportvision re #21, and they said that the time stamp that is on the pitch data for this year is the time when the tracker first detects a pitch in the air.

Mailbags has been updated…

Updated.

***

Question: as the number of questions expands, do we want to sort the list by
a) timestamp of questions (current setup)

b) timestamp of answers (harder to do)

c) category (easy enough to do, but you won’t know what’s the latest question being answered unless you look for it)

Can you let the guys with the questions post their name and home town if they like?  It makes it more “personal” and I think that people like to see their name in lights.

Timestamp of questions.  And I still think we need to make the question and answer link more prominent and “professional” looking at the top of the blog home page.

It should be part of the “The Book, Articles, Blog, About Us” Banner.

It should be “The Book, Articles, Blog, About Us, Ask Us a Question, Read our Answers” or something like that.

I kinda like the banner as-is.  If someone has (strong) thoughts either way, feel free to post…

Question 32 asked and answered.

Question 31 in mailbags has additional answers by mgl and me.

Minor update…

Several questions updated, several more pending answers…

More questions answered…

Tom, I have a question about the “ Calculating pitcher WAR” question. How exactly do you weigh them? Is it something like FIPx + ERAx = Y?

Also, the http://www.tangotiger.net/Marcel/ link is dead.

Thanks for the alert.  It’s case sensitive, so you should do this:

http://www.tangotiger.net/marcel/

As for how much to weight each, it depends on how much data I have.  The more data, the more I weight towards ERA.  Basically, just do a 50/50.

Ah, I understand what you’re saying. The less data you have on a pitcher, the more likely his ERA is “flukey”, for lack of a better word (even though that’s not a word) Take Gavin Floyd as an example.

Andy answered something here:
http://www.tangotiger.net/wiki/index.php?title=Mailbags#Regression_toward_the_mean

Thought my tally at 42 was a little low until I saw myself without the “J” initial at an additional 35.  That brings me (mental math) 77 and counting.

Guys, I asked a long-winded question regarding fast-pitch softball (which I may be coaching in the future) and radical defensive positioning, assuming a 200 ft. fence.  Did you guys get it?  I’d be very curious to see what all three of you think. I’d be happy to send it along again.
The crux of it was - as radical as this sounds - what would the effect be of playing two outfielders at all times and playing with an extra infielder at all times.  I realize that we don’t know the personnel, but I’m trying to weigh how many singles my defense would take away, and how many extra doubles/triples I’d allow.

(The benefit being that, with an extra infielder, a fast-pitch softball team could completely take away the short game, and perhaps affect the other team psychologically as well).

If anyone else has any thoughts on this as well, I’d be appreciative.  Assume that I have a less talented team and have to think outside the box to compete with superior talent.

I remember the question.  Let me dig it out tomorrow, and post it…

Thanks Tango!  I very much look forward to reading your thoughts.  I think the 200 ft. fence is significant, because it means that the ball can only roll so far, and I don’t think you should have to worry about giving up an inside-the-park HR if you play the ball correctly, even with 2 OF.

(Although, fast-pitch softball also has shorter basepaths, of course).

My theory with the extra infielder was that you could play her as a buckshort/"true" 2b, practically standing on the bag - a few feet to the pull side.  That player becomes responsible for all plays at second, which frees up your actual 2b to play at the lip of the outfield grass, and your shortstop to play in the hole, where so many softball hits land.  You could also play your third baseman extremely close to the hitter, and put that player and the P in charge of all bunts, allowing your 1b to play back and cover first at all times.

I’m trying to weigh how many extra triples/doubles you’d allow, and if the amount of singles you take away could still make it a an effective strategy.

I wasn’t sure where to put this or who to email, so I figured this thread would get my quickest response. In the Quick’N’Dirty Marcel spreadsheet from THT, I have a question as to where a number is coming from.

On the Marcel tab itself, under the section of data at the bottom, there is a row titled “pro-rated lg avg.” Every single cell on that row has a formula except the one for the PA column. This has a number entered directly. I feel like there should/could be a formula there. Problem is, I don’t know where it was derived from.

Any clue? Once again, I apologize if this should have been emailed and not posted here.

Mailbag updated…

Mailbag updated…

Regarding the softball question… it’s impossible to answer “sabermetrically” with the information given.  Knowing the run environment is sort of a help, but you’d also need to know the BA/OBP/SLG of your league as well, I’d think.  If people are scoring 20 runs a game, the difference between a double and a triple is negligible.  Though I’m guessing if you’re talking about bunting, it’s probably a low scoring league.

Still… do you have Brandon Webb pitching for your team?  What’s the GB/FB ratio?

Don’t forget you’re not just giving up a single, you’re also not getting an out.

In a low-scoring environment, I’d be more willing to allow a few more singles and a few less doubles and triples.

I wasn’t sure where to put this or who to email, so I figured this thread would get my quickest response. In the Quick’N’Dirty Marcel spreadsheet from THT, I have a question as to where a number is coming from.

Josh, you can contact the author of the spreadsheet, Sal Baxamusa, at the email address listed on his author info page at THT (click the link on my name).

I think you guys should link to the mailbag in your article, and also have it more prominently featured on your webpage. I’m sure most people don’t even look at the top tables in your blog. I know I didn’t.

People are conditioned to ignore the nav bar, or not.  They have their routines, or not.  They like to see the offsetting color, or not.

As it is, I make it a sticky every few weeks.  The only thing I know that works for sure is that people will always look at the first blog entry.

I’m not sure what is being suggested.  Can you clarify?

Mailbag updated…

I’m just suggesting you make it a little more intuitive to access or prominent. Maybe I’m projecting, because I usually just skim over everything I don’t find immediately interesting, and had difficulty finding it. The first time I saw the post a dew days ago, there was no link to the actual mailbag in the article so after just skimming the important phrases around the hyperlinks, I assumed it was only a supplication for responses.

Second time I actually read the post and then couldn’t find the mailbag. The link is tiny and looks more like a header or adblocked google advertisement table over the search bar, and the “You Ask":"We Say” split links look like a single link upon first glance, so clicking on the “You Ask” was again, confusing.

I guess this is kind of an annoying criticism but it’s really just because I appreciate the work you do, and was surprised at how much I was missing in the mailbag.

Ah, that’s better.  Thank you for your constructive criticism.

I’ll make some changes tomorrow.

Question:

Where are the Fans Scouting Report results? I can’t seem to find them on your page.

http://tangotiger.net/scouting/

traced: thanks for your suggestion, and I have implemented changes.  Let me know if that makes things better.

Mailbag answers updated, with a HUGE correction by me on the pinch runner.

Thanks Tango, I think it’s definitely better.

Mailbag updated as of:
07:08, 31 July 2008 (PDT)

That pretty well sums it up Tom, thanks

Mailbag updated…

Mailbag updated…

Lots of updates since Aug 12, starting at a very lengthy response here:

http://www.tangotiger.net/wiki/index.php?title=Mailbags#Free_agent_dollars_-_allocation

There’s some stuff in there that is new to everyone, so this is recommend reading.

I used weighted average error instead of r-squared, but I took a look at the accuracy of various ERA estimators:

http://mvn.com/mlb-stats/2008/08/15/creating-a-dynamic-fip-with-baseruns/

Good job.  RMSE and its sisters is better than r-squared for our purposes, since r-squared will alter the slope to get a best-fit.

Q67 has been expanded.  Q82 added and answered…

More updates…

Tom, you should update questions #33 and #46 by posting the link to statcorner.com, since that is exactly what those people were asking for.

Mailbags updated with answers. 

(Hyl: thanks for that.  I’ll update later today.)

Mailbags updated…

MGL, do you still work for the Cardinals?