THE BOOK--Playing The Percentages In Baseball

Sunday, October 21, 2012

Tigers game plan v Yankees

Jim Leyland said it was:

Our advance scouts did a great job, and Jeff Jones, our pitching coach, did a great job putting the plan together and our pitchers executed very well… We didn’t do anything fancy… We sped it up and slowed their bats down. Nothing tricky to it.

Except for Ichiro of course, which seems fine to me to make a blanket assertion that any rule you make won’t apply to him.  Here is the Yankee stats, courtesy of BR.com:

http://www.baseball-reference.com/postseason/2012_ALCS.shtml#post_batting_loser::14

The focus should not be on Cano or ARod, but on the whole team, a team performance was more than 3 SD from the mean.  I’d like to see an historical analysis on this front.  It’d be embarrassing if all this gets washed away with a focus only on ARod.

Friday, October 19, 2012

fWAR v rWAR

Great stuff here.

http://www.beyondtheboxscore.com/2012/10/19/3521634/wins-above-replacement-variants-fWAR-rWAR-WARP

The good news is that both BR.com and Fangraphs have implemented the roughly 57% / 43% breakdown between nonpitcher and pitchers that I advocate. 

More good news is that Fangraphs and BR.com have more pitching wins in the AL than NL.  (Whether the amount is correct, I haven’t checked.)

But, Fangraphs may have a problem with their nonpitcher wins breakdown between AL and NL.  It’s possible that the breakdown is as Fangraphs is showing, that the true difference between the two leagues is all on the pitcher side, and then some.  BR.com takes a neutral position, and simply says “I don’t know where the difference is, so I’ll just presume it’s even”.

So, I’d like to see more analysis here.

And, BPro’s numbers obviously went through a change for the last two seasons that didn’t get applied to the prior years.  I would consider this a significant problem.

Wednesday, October 17, 2012

Solution to Infield Fly

Here’s a novel one, and if you understand “fair catch” in football, then you’ll like this:

Let the batter “surrender himself,” and in so doing take the force off the runners on first and second (or first, second, and third). On any fly ball, let batter make a signal, if he so chooses, that he gives up the right to try to reach first, and is called out. (It’s analogous to three plays in football: Fair catch, kneeling down, and quarterback slide.) Obviously, this would only be desirable for the batter when the ball was almost certain to be caught, and a drop would only advantage the defense—i.e., the conditions when the infield fly rule applies. I like that this makes it a decision for the players. If the batter wants to take his chances that the ball will not be caught, and that this would advantage his team (e.g., when the ball is deep enough), that would be up to him.

That is, it’s the offense that decides if they want “relief” from the play.  Basically, you force the batter to abandon his baseline if he wants to have his flyball declared an automatic out.  It could be any kind of signal.  He can say take off his helmet, he can run toward the dugout.  Anything that everyone can agree is an obvious “fair catch” signal.

This is similar in spirit to the “advantage” rule in soccer, or the declining of penalties in football.

I love it!

I presume there are those among you with inertial reasoning who believe that a rule created 120 years ago is “perfect”. This thread isn’t for you.  Don’t bother posting here.

This thread is for those with novel ideas, to brainstorm ideas.  It’s not to shoot down ideas.  If you want to point out flaws, you must ALSO point out how to improve it.

Monday, October 15, 2012

Biggest rule changes in each sport in the last twenty-ish years?

I’ve been following sports since around 1976 or so.  And I’ll try to think about the biggest rule changes in each sport that I’ve followed to any extent.  This is just me off the top of my head, so I may miss several of the big ones.  So in the comments, start with the sport, and then put your nomination.  And it has to be an actual rule change, not a defacto rule change.

Baseball - lowering top of strike zone?  balk rule? HR video review?

Hockey - goalie crease/protection?  five-minute majors on serious infractions?  second referee? removal of center line for offsides? video review on all goal calls?  suspension for coach and player(s) for a bench clearing brawl?

Football - QB protection? starting point for kickoff? video review on challenge?

Basketball - zone defense?  3pt line?

Soccer - ?!?

Tennis - video review?

Golf - rules are perfect, except for the rules that were changed, in which case, THOSE rules are now perfect?

Boxing - fewer rounds?

Other sport?

***

Again, don’t comment on my list as if it’s not good enough.  I’m telling you it’s not good enough!  Just tell me what YOUR list is, and then we’ll come up with a more definitive list.

Thursday, October 11, 2012

Rerun: flexible Strasburg plan

I know the Nationals fans are suffering right now.  And they want to know if they had to be in this position.  The answer is “no”.  I posted this six weeks ago, so I am reposting for your commiseration:

I wrote a program that made Strasburg always pitch on Saturday using the Nationals actual schedule.

I ended up with the following number of starts (number in parens is number of starts every five days):
34 Gio (24)
33 Zimm (24)
33 Jackson (21)
26 Detwiler (9)
26 Strasburg (0)
10 “Sixth� starter

You’d need that sixth starter on these days:
12-Apr
19-Apr
17-May
28-Jun
19-Jul
24-Jul
3-Aug
9-Aug
6-Sep
27-Sep

I tried to maximize starts for the top 3.  It’s possible that if I didn’t maximize for Jackson, that maybe I’d be able to get more starts for Detwiler, and also need less “sixthâ€? starter.

The important part is that it MUST BE ON A SATURDAY!

If I made him a Sunday starter, I’d need 19 “sixthâ€? starter starts.  It was similar if I tried any other day of the week.  Saturday was the key.

In any case, if you are going to give Stras 26 starts anyway, whether over 162 games or whether over 140 games, you STILL need another 7 starts from the “sixthâ€? starter at a minimum.  So, 7 of the 10 is to simply pick up the missing starts that we are limiting Stras on, regardless of the setup. Plus another two games are double-headers like I said, in a stretch where they have 1 rest day, so they’d almost certainly need two “sixthâ€? starters then.  Indeed, two Lannan starts are during these double-headers. So, that leaves with just one start missing.

As it turns out, 10 of the 26 Stras-on-Sat starts would have been against Phillies, Mets, Braves.

So, the Nats could definitely have done it, by using Stras as a Saturday-night-special pitcher.

And I also said this:

Since most rotations are of the “5 gamesâ€? variety, that means that off days occasionally makes a pitcher start every six days.  Indeed, 12 of his 24 starts (after Opening Day) had him rest the extra day. 

And on some occasions, even two extra days.  That is, three times he pitched on a weekly basis.  Not that three games means anything, but taking those three games plus the game after the All-Star game: 27 innings, 35 K, 5 BB+HB, 20 hits, 3 runs.  That’s a 1.00 ERA if you didn’t notice the 3 runs and 27 innings dividing so easily.

So, there’s two things the Nationals could have done:
1. Pitch him against divisional opponents.  You get double-bang for the buck.  Not that it ended up mattering anyway because of the way they are running away with it.

2. But more importantly, if they pitched him on a weekly basis (i.e., with two extra days of rest), and limit him to some 90-95 pitches per start as they’ve been doing, that’s 25 or 26 starts of 6 innings each, or 150-156 innings.  You give him one start in each of the possible playoff series, and that’s 3 more starts, at maybe 7 innings each, or 171-177 innings for the season.  Heck, might even have room for two starts in the World Series.

Welcome to the empty feeling of an Expos fan, Montreal’s legacy to Washington.

***

Is it too much to think that maybe one day one team will look at some up-and-coming pitcher, or some back-from-rehab pitcher, and simply make him a Saturday-night-special pitcher?

Verifying the forecasts of experts

Back in the 1980s, I would buy every baseball magazine I could find.  My favorite, and the name escapes me at the moment (Sports Business Journal maybe?), had the “year in quotes” in the back.  I loved reading those.  A perfect snapshot to the year. 

And in the late 1990s, I would often read the Wall Street Journal (our office got a subscription), and I would be very interested in how the forecasters from the big brokerage houses (Merrill, Lehman, etc) would do against the S&P 500.  They each came out with their “top 10 picks of the year”, and WSJ tracked that.  And only one of the 13 or so brokers would beat the S&P 500.  It was then that I decided that we’re all full of sh!t when we try to predict anything, as if a single person is better than the marketplace.  (Wisdom of the Crowd.)

Ben does a fantastic job in putting together my two great interests, by taking pre-season quotes from players and management, and then seeing how they did.  Obviously, he’s cherry picking, but it’s a great lesson to take everything with a grain of salt.  He’s got every team covered, so read through them all.  (I hope no sub is required.)  I love the Carlos Lee (Astros) quote, and then followed by the Mets quote.

As a tangent, after reading the Marlins quote: is the “new park fad” over in terms of getting fans to come in droves to see a new park?  Or is this a Florida-thing?  I think I’ve seen a study on this, and if so, can someone post a link to it?  I’d like to see how the attendance has done in the first, second, and third years of new parks, relative to their old parks.  (I understand win% affects things, and if you want to adjust, every win adds about 2% to the attendance.)

Wednesday, October 10, 2012

Montreal honors Gary Carter

A stadium and a street:

Former Montreal Expos’ catcher Gary Carter, who died of brain cancer in February, will have a baseball stadium and street named after him in Montreal.

The baseball stadium in Ahuntsic park in Montreal’s north end will be named Gary Carter Stadium. And Faillon St., near Jarry Park where the Expos first played in Montreal, will be renamed after the Hall of Fame catcher who played for the Expos and New York Mets.
...

The city says the baseball stadium in Ahuntsic park is considered one of its nicest baseball diamonds used by teams at the regional and provincial levels.

Tuesday, October 09, 2012

Ben Johnson (and Kirk Gibson)

I love the ESPN 30/30 series, and I’m looking forward to the one on Ben Johnson, perhaps the single event that shattered any pretense of what the sporting world was to me.

How do I explain that? And then how do I explain that everything — and I mean everything — that came later doesn’t make a damn bit of difference because that moment was complete and self-contained and absolutely perfect? How do I explain why, when I close my eyes, I can still see how the light fell that day, and how the stadium seemed to shiver beneath my feet?

How do I explain how long I seemed to hold my breath?

9.79 seconds.

Then, the world exploded. And after that, the echoes. The endless, sounding echoes. Always the echoes.
...
Gradually, the details came to light. Johnson had a troublesome hamstring, and so he’d gotten himself shot up to run the race. (Which is not to say that he was unfamiliar with the ways of the needle, just that this was the proximate cause of the events in Seoul.) It was only a few weeks later that a Los Angeles Dodger named Kirk Gibson had a bad leg, so he shot himself up to be able to play in a World Series game against the Oakland Athletics. He hit a home run to win the game and people say it was one of the greatest things they ever saw.

Me? I’ll take that afternoon in South Korea, when the light fell just right and the shadows poured down the track before the runners did.

Saturday, October 06, 2012

Rule-explaining challenge #3

Explain why the infield fly rule needs to exist on popups, but not on liners and grounders?

Rule-writing challenge #2

Re-write the infield fly rule so that the rule is clear, and doesn’t have any absurdities to it.

Rule-writing challenge #1

Re-write the infield fly rule so that the rule is clear on its own, and consistent with what we saw, however absurd you may otherwise think.

Friday, October 05, 2012

Mike Trout is Tim Raines+

Why is it that the saber-followers outside of Detroit love Mike Trout? 

Tim Raines was a star player when the 1987 season started up, but he was without a contract because of collusion.  Teams offered him a discounted contract (Padres, Astros, Mariners I believe), and Raines said he was all set to go to a team outside of Montreal, but they pulled the offer back.  The offer was for about half a million LESS than what the Expos ended up giving him.  Weird right?  Expos offered him 1.6 million per year for 3 years, the other teams were offering him just a bit over 1 million, they pulled their offers back, and then Bronfman gave Raines even more to come back.

The end result is that Raines missed a month of the season because of the free agency rules in place.  His first game back was a game for the ages, on NBC’s game of the week, with Vin Scully calling that game, capped with a grand slam in extra innings.  I still remember sitting on my couch watching that game.  Raines had no spring training.

Anyway, the remarkable thing about that 1987 season is that Tim Raines, missing a month of the season, led the league in runs scored with 123.  As leadoff hitter, he drove in 50 runners on base (plus his 18 HR for 68 RBIs).  He stole 50 bases, got caught only 5 times.  His slash line was .330 /  .429 / .526, for an OPS of 955.

Mike Trout also missed a good share of April.  Trout is also a leadoff hitter.  Trout also led the league in runs scored (129).  He also drove in alot of runners, 53 (plus his 30 HR for 83 RBIs).  He also stole alot of bases without getting caught much (49-5).  His slash line is also impressive: .326 / .399 / .564, for an OPS of 963.

The two major differentiators are the age, with Raines at his peak, and Trout just getting started.  And the other is the fielding.  Raines was a decent fielder, good enough to play CF for one year when he was a bit younger.  But Trout is good enough to play CF until he’s 35.

Raines finished 7th in MVP that year.  That was a bad year for the voting though.  Perhaps an even better comp for Trout is Eric Davis, also an outstanding fielder.  Davis played fewer games than Raines, but had more WAR, yet finished 9th.  Tony Gwynn led the league in WAR but finished 8th.  Dale Murphy has his resurgent year, finished 3rd in WAR but finished 11th.  We had to go to Ozzie Smith’s playoff-bound Cardinals, and his 5th place in WAR to see someone deserving high up in the MVP (2nd, behind Hawk, who has his own collusion story to tell the kids).

Anyway, saber-followers love Tim Raines, and Mike Trout is more than a worthy successor.  Soon enough, he’ll be beloved by all followers like Rickey Henderson was.  We’ll get there in time.

Qualified leaders and Medlen

We all know about adding hitless at bats until you reach the qualifying level, for those below that level.  So, a guy with 200 hits on 450 AB, and 500 PA gets 2 extra hitless at bats to qualify as 200/452 = .442 batting average.

Pitchers don’t get anything like that because of the obvious.  I will state the obvious because I know that someone out there will want to state it anyway.  You can give up an infinte number of runs.  Nevermind that ERA leaders don’t give up infinite number of runs.  But, that’s the logic we’re supposed to follow.  F-ck that.  The logic I’d want to follow is more akin to “replacement level”.  And not just a reasonable replacement level, but a replacement level so low as to be embarrassing.

How about 1 ER per inning.  That’s a 9.00 ERA.  That’s embarrassing right?  Kris Medlen gave up 24 ER in 138 innings.  He was 24 innings short of qualifying, so I would add 24 ER and 24 innings to his totals, insofar as the leaderboard is concerned.  That’s 48 ER, 162 IP, for a 2.67 leaderboard ERA.  Kershaw led with 2.53, and Dickey followed with 2.73.  So, Medlen would be listed second in the leaderboard.

This would also help Mark Eichorn’s 1986 rookie year, the same year as Clemens’ coming out party.  You kids can look that up.  He was a reliever by the way.

Wednesday, September 26, 2012

Five rookie starting pitchers, and heading to the playoffs

Neyer on the A’s.

Reminds me when the 1986 Canadiens won the Stanley Cup with 10 rookies and a rookie coach.

When scab umpires roamed the fields of baseball

A great article by Larry Stone, as he brings us back to the MLB umpire strike of 1979. 

Monday, September 24, 2012

Re-writing leaderboards

Sean proposes two lists.

It’s interesting though that since the rule (now) says that you can only add the 1 out for each missing PA to get him to 502 as long as he doesn’t miss games due to suspension.  That’s crystal clear, but Sean is saying “no”.  Which is odd.

If Sean uses the current 3.1 x games played even in the years when this rule did NOT exist, then, sure, ok, I’ll go along.  But, if he sticks to the contemporaneous rules in those cases, he should stick to it in this case.

You want to hear something funny?  If the Giants are rained out and they don’t make up the game, then 161 games x 3.1 PA = 499 is the threshold for Melky, and even under the new revised rule, he WILL qualify!  Ah-hahahaha.

  If god can’t solve the middle east crisis, he can at least interject in this case.  Even the Holy Writers would turn against god.

UPDATE: Correction: According to qualified sources, it’s based on 162 games, regardless of rainouts, or tiebreaker games.

Friday, September 21, 2012

“Oh, No! Not Another Boring Interview With Steve Carlton”

The story behind the story.

For you young kids: Carlton rarely spoke to the media.  So rare, that you can probably say “never”, and you might be right.

Different year, same situation

Well, this is a fantastic little find from Dave.  Trout 2012 is to Josh Hamilton 2010 and Cabrera 2012 is to Cabrera 2010.  And Hamilton finished ahead of Cabrera in MVP.

Expos 1981

“...it was the season of Light, it was the season of Darkness”.

Wednesday, September 19, 2012

How is Mike Trout NOT the obvious and sole MVP contender?

What WAR does is provide a framework for discussion, a framework that can be expanded to include timely performance if one so wishes (be it at the game-level or the date-level).  You then just need to create your own personalized implementation of that framework.

Mike Trout is so massively ahead of the pretenders that I cannot believe a discussion is even possible.  But if someone wants to try, go ahead.  But, please, don’t be myopic or political.  That means not to just look at those things that support you, while dismissing or diminishing those things that work against you.

I expect to have zero comments in this thread, so if you are brave enough to make the first comment, make it a good one.