THE BOOK--Playing The Percentages In Baseball

12.1 A “read more” link in the RSS feed, so I don’t have to click two links to read the entire post + comments.

12.2 Basketball. We need more smart folks. Come on in, the water’s great. Regression to the mean is a topic that needs addressing.

4,5,12a,12b

12a. Park factors primer
12b. Similarity score primer
No hockey please!!!

vr, Xei

Xei/2: even if I wanted to, I couldn’t… I’m Canadian.

I find 5 and 6 tremendously exciting.  7 has promise, though I’m not sure what data you’re talking about.

Hawk/4: regarding 7, basically any of the splits data you see at baseball-reference, and more Retrosheet-splits that we can think of, plus merging the Japanese data, and any minor league and Negro League data we can get our hands on.  Plus also including as much sabermetric metrics and counts that we can think of. 

The end-result is that there should be little use for Retrosheet event files on a grand scale, because we’ll have captured all the information we want out of it into the database.  Any suggestions I made here, I’d incorporate:
http://www.baseball-reference.com/blog/suggestions

The three I would really like to see, in order of preference, are six, two and one.

As far as number twelve goes, permit me to be greedy here. Someone somewhere has to do something with tennis. There are 108 ranking systems, at the least out there for college football (click my name), but none as far as I can tell for tennis beyond the two tours’ official rankings. Of course, those systems are woefully flawed. The problem with tennis is that there is absolutely nothing out there as far as data to build on, so it’s a huge burden for someone to start.

I’d be more than willing to do some bitch-work formatting Community Forecasts if you could get a split database together by the end of the month; it’d be insanely useful.

#7 is really cool, as is #5. 

On #8, maybe there could be some kind of bribery scheme.  “Nobody gets the 2007 scouting report results until there are 100 articles!”...something like that. 

Just kidding, although playing on people’s guilt might not be such a bad thing

Not really on topic, but since someone already brought it up ... how about eliminating the “read more” entirely?  I hate having to click twice.

In order of what I would most like to see done:

1. Community Forecast - it’s a great idea, it’s something I did at my teams forum. Are we talkin just having a ballot where people input there projections for players OPS, or the 3 split stats? That’s what I did, and it’s worked well. I’m excited to see how well fan’s forecasts correlate with actual performance.

6. WOWY fielding. This system is awesome. More work should definitely be done.

3. Clutch project

Hmmm… the “read more” is done intentionally, so that you can navigate through the main page without going through too many clicks.  That is, the tradeoff is the two-clicks for reading a long post against all the extra page down clicks to go to a particular post further down (which you would do occasionally.

I would actually prefer that YOU change your blog so that it would behave like this!

I’d personally like to see anything related to hockey.  Not even so much as the number-crunching, but the theory.  What data is out there, what matters, etc.  And if it’s really possible to have an intelligent analysis of players beyond goals, assists, and +/-, with the data available.

I don’t get the two clicks complain.  I never hit the read more anyway, just click on comments.  It gets you the whole post, and the comments.

@6: If you’re reading through a feed reader (or at least, though my feed reader) you don’t see a comments link. You just see a “read more” link, which shows the full post, but you have to then hit a “comments” link to read the full post+comments. This two-click dance step has bugged me so much that I actually downloaded two other feed readers just to see if I could avoid at least one of those clicks—neither helped. I know this is a minor point, but it really has bugged me in the past.

(I am reading this through GreatNews, one of the others I tried was RSS Owl, and I forget the third. Are you reading this through a feed reader? Because I would love to know which one allows a single-click -> full post+comments.)

Of course I meant @13 not @6.

You have no idea how much the Retrosheet Fielding Database will help. It might be the first step in my dream of a CAD database, which you acknowledged would be a bear.

Two ideas:

Groundball/Flyball. In The Book, there’s a chart of wOBA by breakdown, showing that like-against-like favors the pitcher, but it doesn’t show a breakdown of how. Is it singles or extra base hits, or strikeouts?

Hit-and-Run: How good a strategy is it?

6 and 7 are the two biggies for me

I’ll echo #6 as something I’d love to see, though Rally’s excellent retrosheet fielding numbers make this not quite as important as it would have been a few months back.

Possible related project idea as a #12: historical position adjustments, based on either your or Rally’s retrosheet fielding data.  I’d be very interested to know the extent to which changes in defensive skill among positions has tracked changes in offensive performance among positions.  You could do it decade-by-decade and have fabulous samples to work from. 

This would be a real boon for folks like me who are starting to dabble in historical player valuation .  It’d also be nice just to have the position adjustments calculated on a system other that UZR, in case UZR handles some positions differently than others.  ...  I could do this too at some point, but the night class I’m teaching right now is really killing my baseball analysis time. -j

#11- I’d love to see what you can do with the pitch f/x data, you always seem to have useful suggestions for Mike Fast and that crew.

And I personally despise hockey, but I’d be interested in seeing a little bit of the analysis that you do with it (in addition to the scouting report).

@Xeifrank/2:

MGL just did something on park factors, although it seemed rather technical and beyond my interests (I skimmed it)

http://www.insidethebook.com/ee/index.php/site/article/mgl_component_park_factors/

I second JinAz’s request for historical position adjustments.  1, 4, and 6 are also high on my list.  Thanks.

I forgot one. I added #13.

Count me in on the WOWY love. Great stuff. Thanks.

It’s too bad it’s likely impossible to track down data to study defensive strategies (historically, that is). That is, “no doubles"/guarding the lines, playing the infield in, etc. I’d love to see a “The Book"-like chapter on these tactics.

Added 14 and 15.

Ooooo...#15 sounds fantastic…
-j

#’s 13 and 15 would be at the top of my list as well.

I don’t supposed we can demand that you finish all of one through fifteen in the next month?  Nah, didn’t think so.

Tom, put me down for more hockey analysis. I’d love to read your thoughts on the subject, some sort of pre-trade deadline or playoff thread would be great.

6, 7, 9 and 13.  I get frustrated when I try to find old work of yours but can’t easily.  Perhaps this is something to incorporate into the wiki?

Ed/1: the RSS feed will now give you one-click access to the comments

Ok, based on responses here, the following will be my schedule.  I’ll do my best to follow-through.  (Text in parens refers to original item number at top of blog.)

TIME-SENSITIVE
T1. By Feb 11. College Scouting Report, 2008. (Item 5)
T2. By Mar 07. Clutch Project, 2008.  (Item 3)
T3. By Mar 14. Fantasy Forecast, 2008. (Item 4)
T4. By Mar 21. Hockey Scouting Report, 2008. (Item 2)
T5. By Apr 01. Community Forecast, 2007. (Item 1)

PRIORITY OF REST
P1. With Or Without You, Retrosheet Fielding Database. (Item 6)
P2. Baseball Database. (Item 7)
P3. Primers - Tango Archives, Wiki contributions, Critical Reviews (Items 8, 9, 10)
P4. Compare Hit Location Systems - BIS, STATS, HitTracker.  (Items 13, 14)
P5. Fans Scouting Report Database.  (Item 15)
P6. PITCHf/x.  (Item 11)

This post is mainly for my purposes.

T1 done.
T5 done.

Say, Tom, would it be possible to compile traditional fielding stats from Retrosheet by pitcher handedness, batter handedness and base/out? It’s a little late to the game, but it should be relatively easy. A side part of With or Without You.

I’m already tracking that in my WOWY, and it’ll be a part of it.

Post mainly for my purposes.

T2 done.
T4 Delay to playoffs.

T3 do next.