THE BOOK--Playing The Percentages In Baseball

Tuesday, May 15, 2012

Andre The Hawk Dawson speaks

A nice little interview with Hawk.

Euro 2012 Preview

Big Lead Sports gives us a preview, starting with host Poland.

How to beat the shift

David Ortiz shows that he’s man enough to bunt.  Surprisingly, this is an extremely rare occurrence, as Ortiz, in his career, has only five bunt hits on ten bunt attempts entering today, and now 6 for 11 for a 55% success rate. 

The question is how often can a hitter bunt to make it more effective for him to produce runs, than to work the count and/or swing away.  We know that Ortiz has a career wOBA of close to .400 when he doesn’t bunt, which is where you will find great hitters. 

With the bases empty, the wOBA equation gives a weight of almost 0.9 for a single with 0 outs and under 0.8 with 2 outs.  So, in order to get a .400 wOBA (and breakeven for a great hitter), a batter would need to successfully lay one down about 45% of the time with 0 outs, and 50% of the time with 2 outs.

Ortiz as I said, is a career 55% success rate.  That is of course based on only a sample of 11 attempts, so we really don’t know how good a bunter he is.  Any hitter who can lay one down over 50% of the time against the shift should simply keep bunting.  As Jeff notes, it’s the batter’s equivalent of the IBB.

The Mariners, like most baseball teams or every baseball team, shift the infield for David Ortiz. In the fifth inning today, Ortiz dropped a perfect bunt down the third-base line for a single. Fans eat it up when players do this, even though it takes the bat out of their hands. When players get intentionally walked, they get the bat taken out of their hands, and fans can’t stand it. Fans are so weird.

Anyway, a player once told me that he could lay one down 100% of the time if the pitch were over the plate, and 50% of the time if the pitch was off the plate.  That would mean that roughly 75% of the time, a great hitter should be able to lay one down.  Even if this player is exaggerating, let’s say it’s 80% of the time he can lay one down if it’s over the plate and 40% if it’s not.  That still sets the success rate at 60%, and that’s if the hitter bunts equally if the pitch is a strike or not. 

What if we’re trying to be realistic and more complex?  How about if it’s a strike, he can lay one down 70% of the time.  If it’s off the plate, he can lay it down 35% of the time.  And let’s say that he’ll attempt the bunt on 90% of the strikes and 30% of the balls.  And let’s say pitchers throw an equal number of balls and strikes. That gives us a success rate of .9*.7*.5 + .3*.35*.5 all divided by .9*.5 + .3*.5 equals 61%.

So, we should be setting our expectation that a great hitter would lay one down and be successful 60% of the time, which would give them a wOBA of .500 to .550, and turn them into Barry Bonds.

That we don’t see this happening is a huge inefficiency among great hitters who are shifted.  These batters, when shifted with no runners on, should bunt, bunt, and then bunt some more. 

Among average to poor hitters, the breakeven point is that much lower.  Whereas the breakeven point for a great hitter is 45% to 50% success rate on bunts, for an average hitter, it’s all the way down to close to 40%, and for a bad hitter, it’s around 35%.  And, we’d expect average hitters to be able to bunt better than great hitters (because of experience), and similarly, the bad hitters may be the best bunters (because they need to learn whatever to survive as hitters).

So, to shift against an average or worse hitter is about the worst defensive alignment you can imagine, and the average or worse batter needs to bunt any chance he gets, when the bases are empty.

Will Pujols end the season with at least 30 HR and .500 SLG?

Just fantastic work from Poz:

This is a very good example of how regression toward the mean works.  A better one would be to add pro-rate his current 2012 stats for one more week, and then repeat what Poz did starting May 22 to end of year for each season.

And the best way is to do the pro-rating for one week, then May 22 to end of month give him some random player’s stats for May 22-31 (any random player), and then take Pujols’ June 1-onward like Poz did.

As you can see, just going with the Poz approach gets you most of the way there.

Really, a very good way that Poz is doing it.

Kershaw v Strasburg, part 2

This is an update from last year, comparing Kershaw since July 26, 2011, to Strasburg’s career:

IP: 140, 140
K: 172, 128 <--
BB: 31, 28
HR: 8, 10
H: 108, 100
R: 43, 31 <--

Strangely enough, all those extra Ks are being counteracted by outs on ball in play for Kershaw.  They are matching on walks, hits, and HR.  Strasburg’s career BABIP is close to league average, which must mean that Kershaw’s BABIP since July 26 is among the league lows.  (Now that I think about it, I should have used Pinto’s Day-by-Daytabase to do this.)

On top of which, with the much lower runs scored, despite matching component numbers, means that Kershaw has favorable splits with men on base and/or Strasburg has unfavorable ones (relatively speaking to their own greatnesses).

It’ll be interesting to see how long Kershaw can continue to keep pace, since Kershaw is winning in the two things that sabermetrics would argue is filled with random variation (BABIP and performance with men on base).

And since we can presume that Strasburg’s “raw stuff” is superior to Kershaw’s, that must mean Kershaw beats him on location and/or sequencing and/or having more good luck go his way in order for him to match him overall.  Compared to Strasburg, Kershaw is the “crafty lefty”!

Did Manny Pacquaio actually quote Leviticus?

The article reads like Manny is talking about basically being god-fearing.  But, in the paragraph in question in the article it shows:

Pacquiao’s directive for Obama calls societies to fear God and not to promote sin, inclusive of same-sex marriage and cohabitation, notwithstanding what Leviticus 20:13 has been pointing all along: “If a man lies with a man as one lies with a woman, both of them have done what is detestable. They must be put to death; their blood will be on their own heads.”

I bolded that part.  When I read it, it seems like it’s the author, not Manny, that is quoting Leviticus.  However, the other media are quoting the article as if Manny repeated Leviticus.  “Notwithstanding” is an odd word to use in this case.  The author is saying that, in spite of what Leviticus is saying, Manny is saying to fear god.  There’s no spite there.  Indeed, it’s the exact opposite of spite: in accordance with.

So, using notwithstanding is out of place there, and, the author seems to quote Leviticus, rather than attributing the quote to Manny.

Mets fielding storylines

For those who can’t get enough, Mark has you covered.

Rolodex: Psst… wanna work for some major league team?

NOTE: Thread originally posted Nov 17, 2008, but it’s always relevant.

I get asked every now and then “if you know some guy...” who wants to work for some major league team (baseball or otherwise).  I figure I should create a rolodex, so that when the time comes, I can be a good matchmaker.  So, send me an email (tom~tangotiger~net), type at least the word Rolodex in the subject line, and with numbered answers, to the following questions:
1. your preferred sport(s)
2. where you live
3. where you’d consider relocation (or answer “no")
4. general skillset

This particular criticism of UZR is unfounded

This blogger didn’t read the fine print!

As MGL pointed out yesterday, the shift plays that Lawrie (or anyone) makes are excluded from the calculations.  I’d respond directly to the blogger, but AN is blocked at the office.

Other than the over-shifting, then, yes, UZR is a measure of range and positioning.  Indeed, EVERY fielding metric is a measure of range and positioning.  Why is that?  Because other than the over-shifting, we aren’t being told where the fielders are being positioned.  So, while you as the observer can tell where the fielders are roughly positioned, no one is actually recording this information for us to use!  You can’t then fault a metric for not using data that it’s not being given.  Indeed, if we had the data, WE WOULD USE THE DATA.

WAR-Powered SI Rankings, or, Should we really treat BABIP differently for offense and defense?

Any time there’s a new concept introduced, there’s bound to have hiccups.  Sean recently rolled out the new WAR, and the two big problems were quickly identified by the community, and corrected by Sean.  Fangraphs has actually rolled out a ton of features over the years, and I think the biggest glaring issue they had was when they went with my “quick” park factors over the more standard ones (like Patriot has).  David corrected that fairly quickly. 

I should also point out that I love everything David does with Fangraphs, a true steward of sabermetrics.

Anyway, now we come back to the power rankings they are posting over at SI.  While I ignore all such power rankings, and wouldn’t bother commenting on them if they were just squirreled away in some corner of the internet, they get feedback by the saber-followers.  And anytime saber work gets a black eye, I like to whisper something, and hope it has a little impact.

The latest example is the Blue Jays, who are all the way down at 28th.  The rankings are based only on performance in 2012, which while not a great idea, we’re going to at least evaluate the rankings based on the constraint they impose on themselves.

The Blue Jays have scored 16 more runs than they’ve allowed.  But if we break it up by component, we see that the offense and defense has been an almost perfect match.  They’ve hit 44 HR and allowed 45.  The hitters have 14 fewer 2B+3B and 23 fewer BB+HB, but 24 more singles and 6 more reached on errors.  They’ve stolen 5 more and grounded into 8 fewer bases.  When you add it all up, the wOBA for the offense and the defense are identical.  So, a power ranking should at worst call them .500, and at best use their runs totals and make them above-average.

The key component is that the offense has a BABIP of only .264, while the defense has a BABIP of an equally low .259.  In fWAR-speak, the offense BABIP “counts”, while the defense BABIP is ignored.  In its place is only UZR on the fielding, but not on the hitting.  The end result is that any luck associated with the hitting BABIP is included in WAR, but any luck associated with pitching BABIP is excluded in WAR.  We have a mis-match, as I talked about last week, in that offense is treated under one set of assumptions while defense is treated under another set of assumptions.

(There’s also a typo that shows 16 actual wins and a .543 win%.  The Jays were 19-16, so they showed losses, not wins.)

Anyway, going back to the issue at hand, and that’s basically BABIP for offense and defense.  If someone wants to run an even/odd correlation of BABIP for offense and one for defense, I’d like to see if there’s any meaningful difference.  My guess is that the correlation is going to mimic that of the other components (BB, HR, SO).  That is, we’ll get a slightly higher correlation for offense BABIP than for defense BABIP, and they’ll be close enough that they should be treated in the same manner.

Someone can do the work to prove me right, or prove me wrong.  I don’t know the answer, but the above is the bet I’m making.  And if I need to be more specific, I think the difference in r is going to be about .05, and I’d be shocked if the difference would be more than .10.

Adam Jones said “UZR” and “wOBA”

I get a little kick when a player says anything sabermetric related, and in a good way.

There are stats now like defensive UZR, and all that, but they don’t show the placement of the player. Everything nowadays is so statistical. Like wOBA. We have a wOBA this year –weighted on-base average. It’s pretty cool to look at that, even though we joke about it.

Monday, May 14, 2012

When to buy Facebook?

There’s fundamentals, and there’s technicals.  This is only about the technicals.

Google opened at 100 on Aug 19, 2004 and could still be had for that price on Sep 7, 2004.  That’s about three weeks.  In between, the price went as high as 113, and as low as 99.  After that, the run started.

That’s just one data point.  Make of it what you will.

Sale the starter v Sale the reliever: why can’t he supposedly start?

Kyle speaks up.

UZR leaders, 2002-2012

Min 3000 IP, runs shown per 150 games:

SS: +15 Everett, rest: Br Ryan, Hardy, Vizquel, McDonald, Counsell, Izturis

CF: +21 Guti, rest: Andruw J, Ca Gomez, Patterson, Taveras, Bourn, Rowand, Chavez

Does this pass the 80/20 rule of Bill James, that 80% of the time it gives no surprises, and 20% of the time, you get a surprise?

Compounding effects of the home-site advantage

Phil looks at one component, to show how the relative rate of points scored by the home team is dependent on how “easy” it is to score per possession.  The easier it is to score, the less the relative rate of points will be earned by the home team.  The harder it is to score (the more things have to compound in order for a goal or point to score), then the larger the relative rate of points earned by the home team.

In basketball, the current rules have it that the home team ends up getting 52% of the points.  But, Phil changes the rules so that more compounding actions have to happen in order to get points, and he can change that to 54% of the points goes to the home team.  Or, he reduces the number of compounding actions so that the home team gets barely more than 50% of the points.

It’s still part of the overall theme of “confrontations”.  For example, Nadal’s clay-court advantage is different if we just look at total points earned, as opposed to matches-won.  If you count as a “win” each point earned, maybe Nadal wins 150 points on court, while losing 100 or something.  That is, he gets 60% of all points scored.  But, if you count as a win each game, then he might win say 70% of all games.  If you count as a win each set, he might win say 80% of the sets.  And if you count as a win each match, he might win say 90% of all matches.

So, you can change the rules, like Phil is describing, to control the home-site advantage.  And the compounding effects of the confrontations is how you do it.

When two runners converge toward the same base

I did not see the play, and I have a hard time believing the account here is accurate.  While I can believe that in the heat of the moment, a fielder is so befuddled that he doesn’t know to first tag the runner and then tag the base, I have to believe that as time goes by, it becomes obvious to the announcers of the game, especially if it’s Keith Hernandez.  Is there more to this story than the account suggests?

I looked up the rule, just to make sure there hasn’t been some recent change like a “base is yours until you give it up, no matter what” or something, like when you are playing tag at school.  Rule 7.01 says in part:

“He is then entitled to it until he is put out, or forced to vacate it for another runner legally entitled to that base.”

So, even if you want to stay on the base, you no longer are allowed to that base.  Again, it’s such a basic rule that I can’t believe the account in the blog is accurate, that something else must be going on.

Can someone confirm the account?

Has Brett Lawrie ALREADY saved 14 runs so far this year?

If you save 14 runs over 150 games, you’ve had a very good to great fielding season.  Dewan’s DRS has Lawrie with having saved 14 runs so far this year.

The way fielding systems work is conceptually simple and obvious: how many plays did the player actually make, and how many plays would a league average player for that position PROBABLY make.  That’s it, in a nutshell.

With batting stats, the league average player faces the same kind of conditions in virtually every park (with some slight exceptions like Coors and Petco).  He faces the same kind of pitchers, all hittable pitches are thrown in a 2x2 foot box (or equivalent ellipse), and so on.  Not identical, but similar enough that there’s not that much difference in the hitting conditions for hitters.

Fielders are not like that at all.  A fielder is completely at the mercy of whatever batted ball distribution there happens to be.  Indeed, he can see the exact same hitter-pitcher combination for 300 innings (imagine a never-ending batting practice), and he STILL might face a different distribution of batted ball, compared to another player at the same position facing the same hitter-pitcher combination.

So, for a fielder, he can end up looking really good if he: (a) actually gets to more balls than the average fielder did, and/or (b) makes it look like the average fielder PROBABLY would have gotten to fewer balls, IF WE KNOW what the batted ball distribution he faced.

Now, let’s go back to DRS.  It not only shows Lawrie with having saved 14 runs already after only 309 innings this year, but he ALSO saved 14 runs last year playing only 380 innings.  That’s 28 runs saved on 689 innings, or a rate of 59 runs saved per 162 games.  (UZR is at 21 runs saved per 162 games for his career.)

Now the Blue Jays 3B are #2 in the league in assist.  So, we know that Lawrie is making more plays than just about anyone else.  The question is if he’s making those plays because he’s getting harder opportunities (lower baseline makes him look really good), and/or, because he’s actually making plays that no one else would make.  That is, is the way DRS describing those opportunities somehow biased?  Is Lawrie actually getting really easy opportunities, but DRS is making it look like they are much harder than they are?

As I said, UZR doesn’t have this extreme viewpoint, even though they are both looking at the same dataset.  Is it something about how plays are tracked in Toronto or with Toronto TV feeds?

This is where I think it would be lovely if MGL/Dewan can give us a description of Lawrie’s career so far, and why MGL thinks that Lawrie’s performance is great Gold/Glove caliber, but Dewan thinks it’s otherworldly.  Unless of course they are precluded from discussing this for whatever reason.

Shutdowns and Meltdowns: describing the bullpen story

This is a great use of the Shutdown and Meltdown metric.

There’s two ways to approach using metrics: do you want to use the metric to illuminate something that is (possibly) transient, or do you want to use the metric to illuminate something something that is (possibly) persistent?

If the Marlins fans have this feeling that the bullpen has not been doing its job, or the writer of the story wants to explain to the fans that the bullpen has not been a good story to-date, then using Shutdowns and Meltdowns is a great way to go.  The whole discussion is based on the past, something that is probably transient, and so, we’re not really looking for answers.  We’re just looking to crystallize a story that has unfolded.

If we really care about the short-term future, where players will simply play to their own talent levels (plus whatever random variation that comes with it), we need metrics that better describe that persistent set of traits.  Shutdowns and Meltdowns is not necessarily the best tool to use there.

Sometimes you need a Phillips, sometimes you need a flathead, and sometimes you need a hammer.  Know what tool to use for the job you need.

Tiger then Albert: is Pujols a changed hitter?

Poz says the swing change of Pujols is obvious.

Even if it is, does it really mean anything?  Some players used to be famous for constantly changing their swings.  Obviously, no one actually believes that Pujols talent is hanging on a batting stance thread, that would turn him from arguably the best hitter in the league to the worst in the league.

Anyone want to chime in with anecdotes, or data?

Sunday, May 13, 2012

Irrational managers (so what else is new?)…

This happens 10 times a day…

Hosmer, one of the best hitters in the KC lineup against a RHSP bats 6th because he is “cold” so far this year.

Hector Santiago is the White Sox closer and Matt Thornton is in middle relief? WTF?? Someone explain that to me…