THE BOOK--Playing The Percentages In Baseball

Friday, January 04, 2013

Simple Hall of Fame voting change

This is really simple.  Change to MVP-style voting, where you list players 1 through 10.  First place guy gets 14 points, second gets 9, third gets 8, down to tenth gets 1.  You don’t even need to fill out every slot if you don’t want.  Points are tallied, and the top two votegetters are elected to Hall of Fame. That’s it.  This is a real easy change, and the voters are used to this kind of voting anyway.

***

Then you have the issue of returning players.  You could make it that the top 20 returning players (those who end up #3 through #22) get back on the ballot the next year, plus the next year’s rookie class, so you always have some 30-40 players to choose from.  Not really much different than what we have now, but instead of the 5% rule (which can’t apply here any more), we just go with a straight count.  There’s some flexibility here.  We can look at the last 5-10 years to see how many returning players there have been and use that as a guideline.  Someone else can look into that.

***

Drop off players?  You could drop off players after 10 years on the ballot.  After that, you still have a veteran’s committee, maybe have them around every 5 years, and THEY vote in exactly one guy, in the same MVP-style voting.  This way, all those guys that drop off the ballot have one last chance to get in.

***

This means that every 10 years, 22 players get elected, which is about the “right” number.  How do I know that?  Because that’s what the Straight Arrow readers said.  When asked in a vaccum, it was about 21 players.  When asked specifically for players born 1962-1971, it was 23.

Tuesday, January 01, 2013

BBWAA poised to make historical non-vote

Based on the unofficial Ballot Collecting Gizmo by the unmatched Repoz, five players are easily expected to clear the 50% support line for the Hall of Fame, and two more (Bonds, Clemens) are on the cusp.  None of them are expected to get to 75%.

The last time no one was elected by the BBWAA was on the 1996 ballot where Neikro, Sutton, and Perez all received between 60 and 70% of the votes.  In 4th place was Garvey at 37%.  Before that, on the 1971, 1967, 1964, 1960, and 1958 ballots, there were a maximum of four players to receive at least 50% of the votes, without anyone getting elected (that year).

We have to go all the way back to the early voting, in 1950, when seven players got at least 50% of the vote, but no one was elected.  And again in 1949, with six players with at least 50% of the vote, and no one elected.  I’m sure Bill James’ Politics of Glory explains what happened in those early days.  In 1949 for example, very obvious Hall of Famer Mel Ott made his first appearance and only received 61% of the vote, and again in 1950 only received 69%.  Jimmie Foxx was on the 1950 ballot for the 6th time and only received 61% of the vote.

The 1950 ballot is interesting because the top 23 votegetters eventually got into the Hall of Fame one way or the other.  And in all, 48 of the players on the ballot eventually got in.  And in 1950, voters averaged 8.8 players listed per ballot.  So, plenty were going 10 players deep, and still, no one was getting elected.  I suggest that it’s a messed up process that if you agree that 48 players deserve to be in the Hall of Fame but that you can’t even elect one.  (It’s probably even more messed up that 48 players got in, but that’s another topic.)  But every player that finished in the top 23 eventually got in.  Politics of Glory indeed.  The voting process in 1950 seemed as dysfunctional as Congress.

Here we are for the 2013 ballot, a group that will have at least ten eventually make the Hall of Fame, and yet, they can’t agree today on which ones will go in, and so, none will get elected.  In their quest to make sure no unworthy players get the glory, they will make the worthy one wait and wait and wait. 

Is this really the best process that we can conceive?

And the BBWAA is in a terrible conflict of interest, not even reporting on this news, because they themselves are the ones creating this news. 

Friday, November 16, 2012

Imagining a world without Miguel Cabrera

There were four main contenders for the MVP: Cabrera, Trout, Beltre, and Cano.  There was very little to separate Cabrera, Beltre, and Cano, other than “RBI”.  They all get the “best player among teams which made the post-season”.

Except when it comes to the MVP, the BBWAA implicitly annoints only ONE such player with the exception.  So, if you have Cabrera, Trout, and Beltre, the exception can only apply to either Cabrera or Beltre, but not both.  This is how Trout ended up as #2.  In a Cabrera v Beltre choice for the exception condition, the BBWAA went with Cabrera.  Cabrera got all the “best player among teams which made the post-season” bonus points.

But, what if Cabrera had a bad season?  We are therefore left with our three contenders: Trout, Beltre, and Cano.  Suddenly, the fight is between Beltre and Cano for that coveted exception condition, which is worth major points.  Trout will once again be resigned to finishing #2 on a majority of the ballots.

Follow me here: if you have Cabrera, Trout and Beltre, then Trout is deemed more valuable than Beltre.  But if Cabrera is not in the cards, then Beltre is deemed more valuable than Trout!  (Or such is my supposition.)

Anyway, the BBWAA so nicely gives us each ballot.  What I did therefore is removed Cabrera from the ballots.  If Trout finished #2, then whoever finished #3 moved to the first slot on that ballot.  That person got the “best player among teams which made the post-season” exception, and without Cabrera to claim it, whoever finished behind Trout got it.  Otherwise, I move everyone up 1 slot.

And what happens?  Well, Beltre ends up with 12 first-place votes, Trout is stuck at 6, and Cano got 5.  In terms of MVP points, Beltre gets 297, Trout at 282, and Cano at 198.

In a world without Cabrera, it’s very possible that we’d be seeing Beltre as the MVP and not Trout.  Such is the power of the “best player among teams which made the post-season”.

Thursday, November 15, 2012

Most Outstanding Player Leaderboard

Sam/Kevin are asking if we should have the single-number leaderboard.

You know, like we’d have in tennis, whereby players earn points for each tournament, or they earn dollars for each tournament (ideally they’d be proportionate, but I guess some tournaments may be worth more or less than the money they generate).  Player of the year would be whoever accomplished the most.

Molson’s used to (still?) sponsors the three-star selections for each NHL game.  Molson’s player of the year was whoever got the most three-star selections.  That is, they simply tally who accomplished the most, or rather, tallied who had the most games where he accomplished more than almost anybody else.

So, WAR takes a more granular viewpoint.  It awards “points” for each player’s accomplishment.  Perhaps we should make it so that the winning team ends up with more “points”, or perhaps you give more points proportionate to your margin of victory (winning team gets 60 points if they win by 1 run, 70 if they win by 2, 80 by 3, and 90 by 4 or more… something like that… losing team gets the leftover so that the total number of points awarded each game is 100).  You can even get into negative points if you like, so that someone who goes 0 for 8 in an extra inning game actually is a minus player, and worth less than the guy on the bench who didn’t play.  Maybe.  Or if you want, any player that plays gets at least 1 point for playing at all.

But, whatever.  You can figure out whatever system you want, and then you tally things for every game, and then, voila, at the end of the year, it’s obvious who accomplished the most.  Why do we instead wait for all the numbers to come in over 162 games, and then we have to go through each player’s 600-700 plate appearances and 800-1000 batters faced to figure out how much each player accomplished? 

Actually, we don’t do that, do we?  We just go on memory of what happened.  Hey, he got 145 RBI!  Forget about the other 550 plate appearances where he didn’t get an RBI.  It doesn’t matter what he did there, or rather, my memory will simply ignore it.  That is, I am going to take a SAMPLE of all his PA, and it’s going to be biased sample based on all the stuff he did good.  Let’s look at Federer’s aces and ignore all his unforced errors.

Instead, we get into debates as if we haven’t learned anything since we hit puberty.  How many times can we argue about RBIs without sounding like the politicians we mock?

So, that’s my challenge to you: create a point system, much like James does with Game Score, but do it for the whole team, and tie it in to whether the team won or not.  Can’t be too hard, can it?  (Of course, it’ll look suspiciously like WPA or RE24, so that’s going to scare people.  They key is to not use those terms.)

Wednesday, November 14, 2012

Can you make the case that Mike Trout was not one of the 10 best performers in the AL this year?

I love the IBA that Greg Spira (RIP) founded, and is now in trust with BPro.  It appeals to my Wisdom of the Crowd sensibilities.  That you can have an honest disagreement as to whether Mike Trout or Buster Posey or whoever had the best or third-best performance, that you can try to make that case, but that any single individual voter will simply be subsumed by the crowd.  It requires some level of knowledge, and some amount of independence.  The Crowd however also has its share of a$$holes.

So in the AL, there were 654 ballots cast for MVP, but there were 11 ballots where Trout was not on the top 10.  Every ballot had at least five selections on it.  It’s unclear whether Trout was missing on those with partial ballots, or whether someone listed 10 names, and didn’t put Trout on it.

When Bill James was doing the same thing about 25 years ago, asking his readers to rank players at each position 1 to 26, he would manually go through them, and if he saw some obvious shenanings, like putting Don Mattingly 19th or something, so that Eddie Murray can sneak through overall, he’d reject that ballot.  I do the same thing with the Fans Scouting Report (there are TONS of Jeter and Pedroia haters/lovers).

But when you look through the players, it’s not just one or two players or ballots.  This is pervasive across the board.
http://www.baseballprospectus.com/other/iba2012/
http://www.baseballprospectus.com/article.php?articleid=18879

Kershaw appeared on 460 of 575 ballots, where you can list 5 pitchers.  Was that RA Dickey fans who tried to knock out his competition?  Is there a bias among those ballots that didn’t have Kershaw on it?  And similarly, did Kershaw fans put him #1 and knocked Dickey completely off? 

Trout appeared on 585 of 594 ROY ballots. 

I think there should be some level of “cleanup” required in the IBA process.  I think it’s a blackmark to simply weight every ballot equally.  At least with the BBWAA, they only hand out ballots to voters they deem as “qualified”.

Anyway, I know there are others out there who don’t agree with my position, that I shouldn’t clean up the Fans Scouting Report, and simply hope that the weight of the sample will push the signal over the noise.  And similarly, they’ll see the IBA in the same view, that we have hundreds of ballots, so if a few a$$holes have their voices heard, they’ll be rendered into whispers by the time the crowd’s yell is heard.  That you have to go looking for those whispers.

Where do you stand?

Tuesday, November 13, 2012

Raines30.com updates

We have engaged with a web editor, who is taking control of the Raines30.com website.  If you have any suggestions, links, etc, that you want to see added, or any kind of ideas, feel free to post them in the comments section.

Monday, November 12, 2012

BBWAA makes ballots available

Not that exciting for the AL ROY, but, it’s there.  Here is how WAR would have chosen the five players in question:

WAR Player
11 Trout
4 Darvish
4 Parker
3 Cespedes
2 Chen

Except for the huge K advantage for Darvish, not really much to choose Darvish over Parker.  Looks like BBWAA might be closet FIP fans and not know it. Above numbers are rounded and from BR.com.  Below numbers are Fangraphs:
10 Trout
5 Darvish
4 Parker
3 Cespedes
2 Chen

How would the BBWAA react if there was only one MVP award to hand out?

I present to you the greatest single sentence ever uttered on my blog, courtesy of the Straight Arrow poster named Neil S:

Seeing the entire MLB as one league with one MVP, one Cy Young, etc. would also add a new wrinkle to the AL MVP discussion - no Triple Crown for Cabrera.

Friday, October 26, 2012

Sporting News - Come so far, and so far to go

We’ve come so far, and yet, we’ve got so far to still go.  Cabrera ahead of Trout, Braun and McCutchen after-thoughts, and Robinson Cano completely shut out.

***

“as selected by a panel of 203 major league players.”  Except if you add up the counts, there were 205 votes.

Thursday, October 25, 2012

Most Career PA, age 30+, born since 1931, limited to players with a maximum of 3 WAR total

Bil Buckner leads the way.  Juan Pierre has a chance to crack the top 15 this year.  And how much longer can Geoff Blum go on like this ?

http://bbref.com/pi/shareit/kVtkp

Tuesday, October 23, 2012

Globe Gloves: Fans Scouting Report 2012

The fans have spoken, and Brendan Ryan was selected as best fielder of 2012.

Position-by-position:
C: Yadier Molina with a commanding lead, followed by Matt Wieters and Salvador Perez

SS: Ryan, followed by the future in Andrelton Simmons, and a tight race for third with Elvis Andrus, Tulo, Brandon Crawford, and Jimmy Rollins

2B: Brandon Phillips followed by the future (if he can stick around) in Freddy Galvis

3B: Adrian Beltre way ahead, but more future-following with Adeiny Hechavarria at some infield position

CF: A tight race that is essentially a tie among Peter Bourjos, Franklin Gutierrez, Mike Trout, Austin Jackson, Carlos Gomez, with Michael Bourn and Craig Gentry close by

Corner OF: Ichiro, Heyward, Gose, Alex Gordon, Starling Marte

1B: Basically a tie among Brandon Belt, Pujols, Kotchman, Votto, Teix, Barton, Helton

***

In the tools category, and excluding the above players, honorable mention to:
Instincts: Ryan Zimmerman
First Steps: Ben Revere
Speed: Stubbs, Revere, Campana
Hands: Polanco, Darwin Barney
Footwork: Scott Rolen
Arm strength: Bryce Harper, Rick Ankiel
Arm accuracy: Ankiel

***

Final ballot submissions accepted through to the day after the World Series.

http://www.tangotiger.net/scout/

Monday, October 22, 2012

What did Cespedes do that Trout didn’t do as well or better?

This is what Sam is asking, and he can’t find it.

Friday, October 05, 2012

Kris Medlen v CC Sabathia

In 2008, Sabathia came over to the Brewers and had one of the best half-seasons imaginable.  Kris Medlen’s numbers look remarkably the same.  Sabathia faced 516 batters to Medlen’s 520.  Sabathia’s SLG allowed was .289 to Medlen’s .286.  Strikeouts are 128 for CC to Medlen’s 120.  Walks are 25 to 23.  Hits allowed are 106 to 103. ERA+ 255 to 256.  Even the W/L record are a close match, 11-2 v 10-1.

The major difference is that CC did that on 17 starts and 7 CG, while Medlen as 12 starts and 38 relief appearances.  In that respect, Medlen’s best comp may be the young Liriano, 16 starts, 12 relief appearances, 12-3, 2.16 ERA, 121 IP, 144 K, 32 BB.  (Man, what a loss of a great potential career.)  Liriano had no votes for the Cy Young, even though Kenny Rogers did.

Roy Halladay in 2005 was also very high, 12-4, 2.41 ERA, 141 IP, 108 K, 18 BB.  Jon Garland got votes, but Doc didn’t. 

In these two cases, Liriano and Doc, guys with an ERA 1.00 higher, but nearly 100 IP more, got the votes.CC on the other hand did get votes. 

Can Medlen get votes being 1.00 ERA lower than Dickey/Kershaw, but 100 IP less?  With 5 names on the ballots, rather than 3, maybe he’ll get a few 4th and 5th place votes.  Medlen may end up finishing ahead of Strasburg.

Thursday, October 04, 2012

Miguel Cabrera v Robinson Cano

9 more singles for Cabrera
13 more HR for Cabrera

9 more 2b+3b for Cano
5 more bb+hb-ib for Cano

Net: 16 runs for Cabrera on offense

Is it a stretch to think that Cano might be 16 runs ahead of Cabrera with the glove?  Maybe 10 at least?  Overall, not much daylight.

Who the heck would choose Cano’s season over Trout?

Tuesday, October 02, 2012

Team player of the year

Oh, I like this idea.  For a team’s “player of the year”, it’s not the guy that was the best on the team, but the guy that best exemplifies or otherwise describes the team’s performance that year.  Felix Hernandez for example would not be the Mariners’ player of the year.  Maybe it’s Michael Saunders, or maybe it’s Franklin Gutierrez.  Is it Brandon League or Tom Wilhelmsen? John Jaso, Miguel Olivo AND Jesus Montero?  It all depends how you as a fan happened to appreciate how the Mariners played this year.

So, I’d love to see a sentence of two from you guys, on the teams you follow, and why your guy best exemplifies what your team went through this year.  What’s the lasting image of your team in 2012, and who’s the face of that?

Monday, January 09, 2012

Hall of Fame results 2012

There’s three numbers here: the 1st (to left of name) is how much they actually got, the 2nd one is how much was predicted by

Jay

Chris Jaffe, and the third is how many they got last year.

Players with ! jumped over 10 points from last year.

Actual Name Predicted 2011
86 Barry Larkin	82	62 !
67 Jack Morris	65	54 !
56 Jeff Bagwell	54	42 !
51 Lee Smith	52	45
49 Tim Raines	52	38 !
37 Edgar Martinez	39	33
37 Alan Trammell	32	24 !
23 Larry Walker	27	20
20 Mark McGwire	24	20
24 Fred McGriff	24	18
15 Dale Murphy	19	13
18 Don Mattingly	18	14
13 Rafael Palmeiro	15	11
10 Bernie Williams	12	XX

The BBWAA also provided links to their members’ articles.

HOF Tracker

Sounds like Repoz and this person need to meet.

Thursday, January 05, 2012

Discussing HOF candidates

Fun read.

Sunday, January 01, 2012

HOF ballot collector

By the incomparable Repoz.

Tuesday, December 27, 2011

Poz’s implicit rule of thumb for HOF

His guys and explanations:

Larkin, Trammell
Raines,Walker
Edgar, Bagwell, McGwire, Murphy

He went for seven of the top eight in WAR on the ballot, swapping out Palmeiro for Dale Murphy.  He shamelessly considers Murphy his pet project, and I don’t mind that.  He’s got that Saberhagen / Guidry / Cone, high peak, short career to him.  And his rejection of Palmeiro:

He compiled huge numbers (3,020 hits, 569 homers) in a huge numbers era, but in my opinion he was never a great player — he only twice finished in the Top 10 in WAR, and never higher than fifth.

It’s interesting if you compare Palmeiro to Raines, and sort them by highest WAR.  Best WAR: 7.5 for one, and 7.4 for the other.  2nd highest WAR: 6.8 for one, and 6.2 for the other.  Third-highest: 6.0 for both.  4th highest: 5.5 and 5.7.  It’s at that point that Palmeiro loses it.

Poz has basically drawn the line in the sand that a “high WAR” player needs at least 5 WAR.  Raines has six of those, and Palmeiro has four.  Dale Murphy also has six of those (and his drop to his 7th highest WAR season is a doozy: only 2.9 WAR).

Alan Trammell has 7, Barry Larkin has 7, Edgar has 9, Bagwell has 9, McGwire has 7.  That seems to be his line.  If you have at least 6 seasons of at least 5.0 WAR, you’re a great candidate for the Hall.  If you have at most 4, then, eh.

Which is fine.  But look at Larry Walker: 4 seasons of at least 5 WAR, just like Palmeiro.  Four seasons of 4.0 to 4.9 WAR… just like Palmeiro.  Another six seasons of 2.0 to 3.9 WAR… just like Palmeiro.  Three seasons of 1.0 to 1.9 WAR… just like Palmeiro.

If you add up all those 17 seasons, Larry Walker has 67 WAR and Palmeiro has 66.  The main difference is that Larry Walker did all that with over 3000 fewer plate appearances.  In every one of those 17 seasons, Palmeiro was a full-time player, where he had at least 600 PA every season, except for the 1994 reduced-season (where his 492 plate appearances was done in 111 of his team’s 112 games).  Larry Walker on the other hand exceeded 600 PA only one season (though we should make a similar allowance for 1994).

That therefore seems to be the Poz implicit rule:
- at least six seasons of 5 WAR (I’ll call that the Guy Lafleur rule)
- or at least eight seasons of 4 WAR, but done in limited playing time (awesome when actually on the field)

It’s a good implicit rule of thumb.