THE BOOK--Playing The Percentages In Baseball

Related blog entry from last year:
http://www.insidethebook.com/ee/index.php/site/comments/the_2007_scouting_report_by_the_fans_for_the_fans/

Someone left me a message, but did not provide an email.  Therefore, I will respond here:

Gabe Gross has played on both the Brewers and the Rays and appears on both ballots since he satisfies the minimum requirements for each team (played at least 90 innings on the field for the team).

I will of course not show him on only his current team, since the fans of the old team has in fact seen him, and are more than capable to evaluate him.

Lots of ballots already (over 200).  Thanks!

About 10% of the ballots are accompanied with comments, and three felt the need to point out Betancourt:

1. Betancourt really is a terrible defensive shortstop.

2. Betancourt has become a horrid shortstop. I’m sure you know this, but just to reiterate: he’s HORRID.

3. Yuniesky Betancourt was an elite defensive SS in 2005. I’m utterly convinced that, at this point, he would bottom in the bottom third defensively at 1B. I have no idea what happened, but I have to think that this sort of total collapse in defensive value, at this early an age, without injury as a cause, is unprecedented.

No one else (so far) was mentioned multiple times by readers.

I don’t know anything about Betancourt other than the few readings, with this one from Michael Lewis the most notable:

http://www.vanityfair.com/politics/features/2008/07/cuban_baseball200807?printable=true&currentPage=all

He seems to have gone through alot of personal sh!t, and who knows if right now he’s still going through it right now. 

If these three fans are indicative of the Mariner fans in general who will submit the ballots, then we’ll have someone who will have gone from one of the best fielders in the league to below average, in just 2 years.

Other than injury, that sounds fairly unprecedented.  Which probably means something is going on with him personally.

Betancourt’s STATS, Inc. ZR is the highest it’s ever been; I don’t know about his RZR numbers (the raw figures are very difficult to compare year-to-year, although I guess you can get away with it for infielders.)

John Dewan has Yuni at -3, -10 and -18 (plays, not runs) in 2006, 2007 and 2008 to date, respectively.

Watching Betancourt play now, it really is remarkable how far he’s fallen.  When he came up, he was routinely getting to balls that I didn’t think anyone had a shot at - he covered a ton of ground. 

Now, if it’s not hit right at him, he’s not fielding it, and even if he does field it, good chance he’s going to throw it into the seats.  He never (and I mean never) gets to anything hit remotely close to the second base bag. 

It’s really been amazing to watch.

Phil/5: good stuff.

Those trend numbers that Phil posted, absent scouting information, is virtually meaningless.  As far as a forecast or true talent evaluation would expect, you’d need to weight all of the numbers, with a bit more weight to the most recent.

HOWEVER, given scouting information, and the (intelligent) observers’ view that something unprecedented has happened, then you have to give far more weight to the more recent numbers.

At the same time, you really need to know why he is the way he is.  Is it a long-term transient effect (like a powerful hurricane that’s been around for 4 months), or is it a true change in talent level (like an earthquake that has shaken the foundation to your house)?  Only Betancourt knows for sure.

Current counts, after 24 hours, is 457, which is great!  Here are the team counts.  If your team is at the bottom, then I need help there.  Mariners and Redsox fans, as is the case every year, are the ones with the most fans keen on this project.

n Team
97 SEA
49 BOS
34 CHA
33 TOR
30 TBA
23 NYA
22 CHN
18 ARI
14 TEX
13 LAN
12 SLN
12 CIN
11 BAL
11 MIL
8 MIN
8 CLE
8 NYN
8 KCA
7 SDN
7 ATL
6 SFN

5 ANA
5 FLO
5 OAK
5 PHI
2 DET
2 PIT
1 HOU
1 COL

Seeing I’ve got so much data on the Mariners, and I don’t expect this to affect anything, here are the basic results for the 7 categories for YuBet:
2.8
2.5
2.9
3.0
2.5
3.6
1.9

The average is usually around 3.3 for each category, so we see that he’s pretty much below average across the board.  Expect to see him come in around the low 40s or so, which is an enormous drop, some 25 or so runs from his heights of 2 years ago.

Note that I have made an update at the top of this page.

For those interested in the semi-technical reason for the bug, I will try to explain.  Hopefully, anyone whose confidence is shaken should get it rebalanced.

I’ve been using the exact same process that I’ve always been using, and that is to dump my database file into Excel for a final sort.(*)

Excel has some “weird” sorting issues when it comes to formulas, and it was there that I got hit.  For the other 19 teams, the player with the most games had ID = 1, and then I simply sequentially add 1 to that ID.  Works fine.  But, for the 11 teams in question, the ID of the first player was 0.

All that is really not even a problem, since everyone has a unique identifier.  And the HTML is also fine, since again, all the fields have a unique identifier.

It was in the perl script that I use to loop through the data.  I start at a field I call “cat1”.  However, because the first record had a value for catX below cat1, the perl script never captured the data.

What is very frustrating for me is how lazy I got on this.  In every other year, I would quality check each page one at a time.  I’d put in a data in the first field and in the last field, and then check the output file.  I expect to see data in the first and last field of the file for each record.  I got lazy.  I tried 3 teams, and then decided that was good enough to publish.  Why I did this, I don’t know.  Had I tested each page, I would have immediately seen a problem.

The bummer for me is that I got alot of votes on some of those teams (Ichiro, Overbay, Iwamura, and to a lesser extent Cano, Kinsler, Glaus).

I am disappointed in myself, no question.  I’m usually really good at this stuff, and this time, I just blew it.

Anyway, it was isolated, there was a systematic reason for it, and it’s been resolved.  Hopefully, this should give anyone who is wavering alot more confidence.

(*) I actually had intended to no longer do this, and I had a good program for it for my college balloting on fielders.  But, I wasn’t totally ready to convert that one for MLB purposes, so I decided to stick to the “tried-and-true”.  No more.

We’re up to 630 ballots (without double-counting the resends) after 2 days, which is fantastic.  Last year, I was at 1143 after 1 week.

So, I’m very pleased with the turnout.  Every day, I’ll post the 5 teams I would most love to get more participation from.  If you are a fan of one of these teams, please spread the word.  Ballot counts in parens:

Nationals (2)
Astros (3)
Pirates (5)
Rockies (5)
Marlins (6)

That article by Michael Lewis is quite interesting. Thanks for linking to it. I’ll put up a link at lookout landing.

It is fascinating and certainly gives you the human perspective that you rarely find elsewhere.  More editors should give their reporters the leash Lewis is given.

***

Now over 800 ballots after less than 3 days.  Very impressed!  Thanks, and keep spreading the word.

That worked out great!  The Marlins are no longer on my APB list.  Thanks to: http://tealandblack.net/forums , I’m now at 16 ballots and counting, putting them right in the middle.

As a result, the Royals come into my bottom 5:

Astros (3)
Nationals (4)
Rockies (5)
Pirates (8)
Royals (8)

If you are a fan of one of these teams, please, spread the word.

Approaching 1000 ballots and counts.  On my hit list is now:

Astros (4)
Nationals (4)
Rockies (6)
Royals (9)
Giants (10)

Pirates, Phillies, Padres and Angels fans are next on the list.

We’ve just cleared 1000 ballots after one week, which is a great turnout.

***

Thanks to…
http://mvn.com/mlb-astros/2008/08/21/your-turn-to-scout-ml-ballplayers/

http://www.crawfishboxes.com/2008/8/25/601077/helping-the-a-worthy-cause

http://www.crawfishboxes.com/2008/8/25/601115/houston-astros-at-new-york

...The Astros are now out of the bottom 5.

I’m looking for help on these 5 teams:
Nationals (4)
Rockies (6)
Royals (9)
Giants (10)
Pirates (11)

Astros, Phillies, Padres and Angels fans are next on the list.

I’m going to go after one team a day, like I did for the Astros.  Today, I’m going to look for Nationals fans.  Unfortunately, our biggest supporter no longer blogs:
http://dcbb.blogspot.com/

Anyway, this is a Nats drive today.  You come across a Nats fan, send him to the project.

Thanks…

I have done a preliminary scouting report.  I will be somewhat vague with the results so as to not influence future voters, and where I’ll be specific is in players not in danger of having their totals changed.

Betancourt has dropped out of this world, from an 86 in 2005, and 82 in 2006 to 37 today.  His offense numbers are a bit down across the board, his clutch numbers are in the minus.  I have to believe that something is affecting his concentration.

Ichiro, Beltran, and Rolen are top 10 players, just as they were in 2003, the first time I ran this.

Once again, John McDonald gets alot of love from Blue Jays fans.  In fact, Blue Jays boast the best infield defense in all of baseball (when Mac is playing).  And Wells and Rios are plus defenders in the OF. 

Rick Ankiel has one of the strongest arms in baseball.  The Chris Duncan experiment is simply not working.

Adam Everett.  RIP.  He still has all his running and catching traits intact.  However, he has no arm strength at all and is very poor in his accuracy.  That double-whammy puts him in the bottom 5% in MLB for his arm (almost as bad as Eckstein). This is the kind of thing that is very instructive.  Just by looking at the change in his trait numbers, you know that Everett has a throwing issue, related to some injury. Unless his throwing arm is going to heal, there is simply no reason to think that Everett’s true talent is anything close to his other-worldly UZR numbers of the last few years.

***

Keep the ballots coming…

The following teams have the fewest number of players who received at least 5 ballots:

Nationals (3 players)
Phillies (5)
Rockies (6)
Giants (9)

None of those teams have any players with at least 10 ballots.  So, those are the teams I have a great deal of concern with. The Royals have 14 players with at least 5 votes, and none with at least 10.

I tried working the Nationals angle yesterday, but I’m getting little response.  It’s a real shame.

Anyway, today I’m going to step up efforts on the Phillies.  If you are a Phillies fan, spread the word.

Thanks…

Well, that was fast!  Got several Nats reponses (will have to check my logs tomorrow as to figure out from where).  So, they are not on my “gotta worry” list anymore.

Here is now my must-get-fans list:
Rockies (6 players @ 5+ ballots; zero at 10+ ballots)
Phillies (8; 0)
Giants (9; 0)

Let’s hope the Phillies campaign works as well.

After that:
Royals (14; 0)

Nationals (13; 3)

Twins (12; 4) ... just need one post from Aaron Gleeman from his Friday links

Padres (14; 4) ... maybe I’ll ask DePodesta on his blog?

Pirates (18; 5)

Braves (14; 6)

Every team has at least 9 players with at least 5 ballots, so we’re in decent shape right now. 

Our next target is to get as many players with at least 10 ballots.  Here are the laggards:

Royals (0 players @ min 10 ballots)
Rockies (0)
Phillies (2)
Padres (5)
Pirates (5)
Nationals (6)
Braves (7)

If someone can get a hold of Joe Posnanski, I think one mention by him would move the Royals alot.

I think I’ll try to get out the vote from Rockies fans today.

As usual, anyone who can spread the word on these teams has my gratitude.

Tom,

I’ll post something about this to the Rocky Mountain SABR site, and see if that can generate something for the Rockies.

I’ve posted on a Braves board and a Dodgers board. Hopefully you’ll be getting more reports on those teams soon.

Thanks guys, I appreciate it!

I’ve also prepared an article for Hardball Times.  Just waiting for some fact-checking, and then I’ll send it off to studes.  Hopefully, exposure there will create more interest.

Royals, Rockies, Phillies are low.  Any help there would be appreciated…

Here are current ballots counts.  I have 1360 ballots so far.  The numbers below represent the number of “full” ballots.  I take the 13 players with the most ballots for each team, and average their ballot counts.  It makes more sense for me like this, since most ballots are not completely filled out.  That gives me a total of 821 “full” ballots. 

My goal is to get at least 18 full ballots per team.

90 SEA
83 BOS
48 DET
48 SLN
47 BAL
45 TBA
40 CHN
33 SFN
31 LAN
31 CIN
31 TOR
29 NYA
28 CHA
23 ARI
20 OAK
18 FLO
-- need a bit of help starting here and below
17 ATL
17 CLE
16 MIL
16 NYN
15 HOU
15 MIN
13 ANA
13 WAS
12 TEX
-- need lots of help starting here
10 KCA
9 PIT
9 SDN
7 COL
7 PHI

Thanks to Poz, the Royals are now off the most-wanted list.  I’m always fortunate when one big blogger can use their muscles here.  Here is the updated list:

93 SEA
85 BOS
50 SLN
50 DET
47 BAL
45 TBA
44 CHN
35 NYA
34 TOR
33 LAN
33 SFN
32 CIN
30 CHA
25 ARI
20 KCA
20 OAK
19 FLO
19 CLE
18 NYN
-- need a bit of help starting here and below
17 MIN
17 ATL
16 HOU
16 MIL
14 ANA
13 WAS
12 TEX
-- need lots of help starting here
10 PHI
9 PIT
9 SDN
7 COL

I got a flurry of ballots following the DePodesta post, enough that the Padres are now in the middle of the pack in responses, and other teams moved up a bit.  (Looks like it’s not only Pads fans that go there.)

Now, if someone can prevail upon Baseball Prospectus and BeerLeaguer.com, I’d be most appreciative.  For whatever reason, I can’t get a hold of anyone there.

***

93 SEA
89 BOS
50 DET
50 SLN
47 BAL
45 TBA
45 CHN
39 NYA
34 TOR
34 CIN
33 SFN
33 LAN
30 CHA
25 ARI
24 SDN
21 CLE
21 KCA
20 FLO
20 NYN
20 OAK
18 MIN
-- need a bit of help starting here and below
17 ATL
17 MIL
16 HOU
14 ANA
13 WAS
13 PHI
12 TEX
-- need lots of help starting here
9 PIT
8 COL

Just giving you the bottom teams:
16 HOU
14 ANA
13 WAS
13 PHI
12 TEX
Lots of help needed here:
10 PIT
8 COL