Big Data Coming to Baseball

Baseball sabermetrics has really always been about fairly large datasets.  A single batter's hitting line often reaches 700 plate appearances in the course of a season.  There are over 1000 players who have donned a cap in major league baseball this year, with thousands more in the minor leagues.

One of this week's topics in +Andy Andres ' Sabermetrics 101x course this week is Big Data, with the capital B and D.  There are a number of good points made in his introduction, but one of the most important is that big data does not always mean better data.  Big data can be fraught with bias (lack of controls, systematic bias in data collection), a problem that Colin Wyers has long railed against with our favorite defensive metrics.  Big data can also lead to false conclusions because statistical approaches no longer work well.  When your sample size gets into the thousands, P-values below 0.05 get easy to reach, even when there is no actual meaning to the difference found.

This spring, MLBAM announced a new stream of data with its new player tracking system.  This system, which is already in operation in a few parks this year, will track virtually everything on the baseball field: player position, ball trajectories from the pitchers' hand and from the bat into the field.  It is essentially a replacement of the pitchf/x, hitf/x, and fieldf/x system we've been using (or, in the latter two cases, at least hearing about).  

It's incredibly exciting to hear that this will be used.  The question will be how much of these data will be available to the fan community at large.  We probably don't need to be able to download all 7 TB of data.  But hopefully, we'll get something.  My wish list:

• Everything we currently have with the pitchf/x system for tracking pitches.
• The equivalent of pitchf/x, but for batted balls.  Therefore, we'd get vertical trajectory, vector the ball was hit, velocity, spin, and landing location/hang time.  
• For fielding, ball landing location alone would be a great step forward.  It could be fed directly into our UZR-like algorithms, and immediately improve them by removing systematic bias in our hit location data.  Initial player position could also be really useful, if nothing else than to help distinguish shifts when they happen.

Some of the other measures they show in the videos, like acceleration, reaction time, and maximum speed, could be interesting, and it would be great if they were released along with the stuff above.  But I'm also not sure, for the time being, that I'll be all that interested in them.  With what is described above, we'd have a wealth of useful analytical information at our hands.  From the perspective of valuing players, which is often my main interest, it really doesn't matter if a player gets from point A to point B because they get a great jump, because they have great route efficiency, or because they're fast.  What matters is that they get there.  

Of course, if I'm a team, I care a lot more about the minutia.  It might be that maximum speed can't be taught.  But reaction speed might be able to be taught, and route efficiency almost certainly can be (right?).  But personally, I'm most interested in just evaluating player value.

The concern, of course, is how much of the data will actually be available to us.  I'm frankly a bit scared about this.  Potentially, MLBAM could be really stingy with these data, and we might end up with LESS information than we have now through pitchf/x.  I'm hopeful that this won't happen, however, and I've no doubt that folks like +David Appelman at +FanGraphs will be doing his best to have access to, or perhaps even license, some of the critical info that we as a community want.  

Once the data ARE available, the question will be how well the community makes use of them.  I have no doubt that we'll see a lot of spurious conclusions in the early goings.  Fortunately, the sabermetric community is pretty good at policing itself, and correcting its past mistakes.  We have a lot of bright minds in this community, and hopefully, within a few years, we'll have a good grasp of what these data can and cannot do.

Postscript

I embedded one of the videos that MLBAM released above.  What follows (below the break) are the some of the other interesting ones.  It's really exciting stuff!

Kudos to Aaron Harang

Just a small post to recognize the wonderful start of Aaron Harang's 2014 season.  I didn't get to profile him in my Braves/Reds preview because he's not throwing in this series.  But right now, Harang is leading all NL starters in ERA at 0.85.  I thought I'd do a quick retrospective.

When I started this blog in 2006, Harang was the Reds' #1 stud pitcher.  He was joined that season by a surprisingly good Bronson Arroyo, and they anchored the Reds' staff for 2006 and 2007.  We all know what happened in 2008.  But I prefer to remember his run from 2005-2007, when he averaged just under 5 WAR per season.  

When Harang was at his best, he was always a fly ball pitcher.  But he had a good strikeout rate, and an extremely low walk rate, which helped keep the home runs he allowed from hurting him.  While his 3.7ish ERAs don't look particularly impressive now, one has to remember that the scoring environment in 2006 was much different than it is now.  In 2006, the NL average ERA was 4.49.  By 2013, on the other hand, it fell all the way to 3.73.  So, in today's run environment, he would be something much closer to a 3.00 ERA pitcher at his peak.

What's different about Harang's approach this year?  Ben Lindbergh doesn't think he's doing anything differently than he did for Cleveland last year from a scouting basis.  Certainly, his peripherals don't suggest that his 0.85 ERA is sustainable.  But, in his first 5 starts, Harang has seen his strikeout rate, and his swinging strike rate, soar to levels better than even his best years with Cincinnati.  He has also seen his walk rate jump up, and his ground ball rate drop, of course.  Given all of that, it's hard to expect that his end of season numbers will be any better than league average.  

But I'm going to hope he can keep being effective all season.  Harang was a terrific Red, and I will always root for him.

A late look at the new CBA

Image via WikipediaJust some thoughts on the new CBA, going item by item.

Astros move to the AL West, Interleague all season long.

I'm sort of "meh" on this.  I think Interleague is tremendously overrated by MLB.  Aside from the big ticket match-ups (Mets/Yankees, Cubs/ChiSox, ...  dare I say Cleveland/Cincinnati?), most of the series are no better attended than any other.  As a pure fan, though, I generally enjoy interleague.  I like seeing new teams and players throughout the year.  I do not like facing the Astros 18 times per season.  I haven't checked it, but my hope is for a more balanced schedule within the league with the addition of interleague.  Something tells me that's not what they have in mind, unfortunately, as that would mean less Yankees/Red Sox games.

From an analyst's point of view, one thing I always like about interleague is that it serves as a good way to weight the leagues against one another.  The AL has been dominant since roughly 2005 (before that, the leagues were close to even), and despite the constant protests to the contrary by the excuse-makers, I think it serves as the best measure of the league disparity that we have.

Also, I'm sorry for the Astros fans.  They're already a bad team.  This is not going to go well for them.

Expanded Wild Card system, both teams play in a 1-game playoff for access to the postseason.

I love this.  Love love love it.  I like the extra team in the playoff, though that's not the main point.  What I love about it is that this puts a huge premium on winning your division.  Previously, the wild card was essentially equivalent to winning a division.  Sure, you don't get home team advantage, but you aren't assured that anyway if you win your division.  Now, if you don't win the division, and you capture a wild card, you are subjected to an crazy, single-game playoff that determines whether you get to play in the real postseason.  This is far more of a wild card, and a far worse option than it previously had been.

Furthermore, despite the massively increased uncertainty that goes with a wild card, it also puts the wild card teams at a considerable disadvantage.  Most of the time, you would expect that the teams will have to throw their best starting pitcher in that single-game playoff.  This means that a wild card team, like, say, the Cardinals, won't be able to throw their ace as often in the next series.  In a five-game division series, we might be talking only one appearance.  That's a major disadvantage.

No more type A and type B free agent.  Now, compensation requires teams to offer a salary equal to the average of the top 125 players in the prior season.  

This is great.  First, the type-A/B system used on an absolutely horrible ranking system.  It was based, for example, on things like RBI's, stolen bases, etc, and varied by position.  That it goes away is a great thing.

Second, it removes a substantial penalty for signing mid-tier free agents.  A team like the Reds was unlikely to sign a free agent that would result in only a small upgrade but required compensation, as you might lose more out of your lost pick than you gain by signing a player.  This is good for signing teams, and is also good for the players themselves.  Some legitimately good players (Orlando Hudson a few years ago comes to mind) haven't been able to get a decent contract because part of signing them requires a team to give up draft picks.

Third, it still offers protection to small market teams who lose big money players.  When Joey Votto walks a few years from now (assuming they don't trade him), the Reds can still get compensation for his departure if they offer him a halfway decent contract.  This is the main point of free agent compensation, and my feeling is that this new system will work far better.  And there's still protection for top-10 picks, which means that teams in the bottom third the prior year have a slight advantage when it comes to signing top-tier free agents, in that they won't lose their first round pick.  That's great.

Teams that go over slot in the amateur draft will encounter a taxation system, based on the degree to which they go over slot.  And no major league contracts will be offered.

This is stupid.  Stupid stupid stupid.  The amateur draft, even after the rise in singing bonuses we've seen the lats few years, is still a drop in the bucket compared to other expenses of a major league team.  What this new adjustment fixes is that it reduces the degree to which the draft can be leveraged by small market teams as a means of procuring talent.  Going over slot, offering major league contracts, etc, are all ways that small market teams can add extra talent in an economically advantageous manner.

What this basically means is that, for the same amount spent in the draft last year, amateur players will get less money and poorer contracts next year.  That means that more players will opt to go to college, or pursue another sport, because the incentives aren't as good for baseball as they once were.  I see the main effect of this change a reduction in the talent coming into MLB.  I really don't understand how that can be good for the game.

MLB minimum salary goes from 411k to 500k.

It's a small thing, but it's almost certainly the case that 500k is still too little.  From the player's standpoint, this is a gross underpayment for services rendered.  But as the fan of a small market team, the fact that minimum salaries are still "only" a half of a million dollars means that the Reds have a shot at a contention against richer teams.  I'm reluctant to see that change for selfish reasons.

Instant Replay includes home run calls (as before), fair/foul calls, and trapped ball calls

Yay.  I'd like to see it expanded further, but these are baby steps.  We just need a tv/tablet viewable on the field, though...this running into the dugout thing is awkward.  Some of those umpires really should not run on camera.

Blood testing for hGH

As far as I can tell, this is a complete waste of time, money, and privacy (can privacy be wasted?  It makes sense to me).  I've yet to see a credible scientific study demonstrating that hGH has meaningful performance enhancing effects.  In fact, in acromegaly, which is caused by pathologically elevated hGH levels in adults, we actually see a weakening of muscles.  The data just aren't there--it seems to essentially be an extremely expensive placebo.  And the more expensive a placebo is, the more "effective" it will be.  If you want to ban it, fine--if nothing else, taking it is probably unhealthy.  But until there's scientific evidence that it actually has performance-enhancing effects, I don't see a reason to test for it.

My Answers for Reds Blogger Roundtable

This week, Blog Red Machine has organized a roundtable discussion of sorts that included many of the major Reds blogs.


Anyway, the nature of the beast dictated that many of the answers that I submitted were never published.  If they published everything that everyone submitted, the post would be too long.  So, since the original pieces have already run, I decided to publish all of my answers for those who may be interested.


The National League Central Division


1. I am of the opinion that the Pittsburgh Pirates are the division’s biggest surprise and that our Reds are the biggest disappointment. Would you agree or disagee? If you disagree with either, please state why.

The Pirates are hands-down the biggest surprise.  Unfortunately, I think they're unlikely to continue their success. But it's nice to see Pirates fans having fun in July for once.


The Reds have been disappointing, no question. But I think they are tied for disappointing-ness with the Cubs. I think most had the Cubs around .500 this year--not in the hunt, but respectable. Instead, they've been nothing short of dreadful.

2. What one player that plays for an NL Central team do you feel is the most underrated player within the NL Central and why?

Can I cheat and go local with Drew Stubbs? All I hear about when I watch broadcasts is his strikeouts. And they're frequent, no question--the Reds knew that they would be when they drafted him.  I see people talking about sending him down to AAA. But even so, his hitting line of .252/.323/.391 is better than league average (wRC+ of 102) in the lower offensive environment we're experiencing, and he plays outstanding defense in center field. FanGraphs has him as a 2 WAR player while B-Ref has him as a 1.6 WAR player already this season.

I think Drew Stubbs certainly has some room to grow on his performance this season, especially in the power department. But even as he has been this year (and last!), the guy has put up above-average major league numbers. I don't think that's the current perception.

If you want non-Reds, you can pick one of the other NL Central Center Fielders: Andrew McCutcheon, Colby Rasmus, or maybe even Marlon Byrd!

3. For the first half of the season, if you had to select an MVP just from the teams in the NL Central, who would that be?

There are about 7 guys that I think have a good case, but I'll go with Andrew McCutcheon.  He leads the division in both fWAR (fangraphs) and rWAR (at baseball-reference).  I'm a little skeptical of the WAR data, as I don't know if the fielding numbers that you see at FanGraphs or B-Ref are real.  In past seasons, his UZR, TZ, etc, have not been good.  But even so, he's leading the division in oWAR (WAR, ignoring the fielding), plays a premium defensive position, and is arguably the biggest reason that the Pirates are where they are.

4. And what about a Cy Young type candidate from those in the NL Central...Is there one?

There are a bunch of guys in the mix as tops in the division, but none of them holds a candle to Roy Halladay or his ilk. I'll go with Shaun Marcum in the Central, though. While Greinke got more press when he arrived, Marcum has been outstanding this year. He gets more than his share of strikeouts, doesn't walk anyone, and even hit a home run in his start last weekend. And he does all of it despite only having an upper-80's fastball. Also in the running: Johnny Cueto, Jaime Garcia, Chris Carpenter and maybe even Kyle Lohse?

5. Finally, who is the division favorite for the remainder of the season and why...

There is no clear favorite.  BPro's playoff standings have the Cardinals odds to win the division at 46%, with the Brewers at 40%.  The Reds, meanwhile, have division-winning odds of 11%.  Interestingly, the Reds have the highest 3rd-order winning percentage in the NL Central (barely).  But their current win deficit, coupled with the fact that they're so close in quality level with those other teams, makes it pretty tough to compete.

I think the Brewers' rotation is being underrated in those rankings.  Greinke, in particular, is a major puzzle, and I think is likely to have a monster second half.  I don't see any team competing with that rotation, their offense is top drawer, and they even seem to be catching the ball better this year than in years past.  That's the team to beat, as I see it.  

That said, I also think the Reds have a run left in them.  Four games back is just not that big of a deficit, and Walt has done some interesting things at the trade deadline.  

Baseball In General

1.  No question that year after year we see at least one ASG starter voted in due to his popularity with the fans. And this year more than ever, we see many players pulling themselves out of the midsummer classic due to various reasons, ala the NFL Pro Bowl. Now, players that have no business playing are now participating. Is the ASG now migrating into a joke like the Pro Bowl?

It sort of depends on what you mean by "no business playing." Depending on the data you use, some guys can go from ridiculous selections to perfectly legitimate. Are you talking about performance in 2011? Most seem to think this should be the measure by which we judge worthiness, but I think it's ridiculous to put that much weight on 2-3 months of data.  Using only first half data means you're guaranteed to have guys riding lucky streaks into the All Star Game, which makes any complaints about worthiness seem silly to me. I'd prefer to at least use past calendar year stats (at least then we're not ignoring half of the players' work!), if not projections to pick the best players in baseball to represent in the game. But I realize I'm in a strong minority on that one.

As for guys bowing out, I really don't mind in the slightest. It's an exhibition game, and a player's season has to take precedence. No one should really care about who actually plays in the game or what actually happens anyway. I'd be concerned if one of my team's players played in the game out of some feeling of obligation when he really just needed to rest. The guys bowing out are doing their job.

2. Not many people like the fact that the ASG determines home field advantage for the World Series. What’s you best way to award home field for the WS?

I'd just give it to whichever team had the better regular season record, like they did from 1998-2002.  That makes it something that teams have to earn.  I also liked that wild card teams could never have home team advantage in the playoffs.  The current system is beyond stupid.

3. A big discussion was waged over that of interleague play. For or against and why...

I may be in the minority among blogger-types, but I've always enjoyed interleague. I like seeing new teams and players from the other league. And I enjoy playing the Indians every year. And it's fun to see teams have to adjust to a new set of rules for a handful of games, be it AL teams having to bench Travis Hafner and David Ortiz, or NL teams looking at their bench of defensive replacements and trying to find a DH in there.

Also, I love the wild card, and support adding a second one to the playoffs--especially if it screws up the wild card team's pitching rotation for the Division Series. And I think realignment is always fun to think about, although I didn't like the most recent proposal. But the DH sucks (see? I have some traditionalism in me!).

4. So it looks like the Los Angeles Dodgers could be yours in the not too distant future. Would you have a problem if Mark Cuban were to buy the team because apparently...a) he might be interested and b) baseball has not wanted him in the past. Mark Cuban as an MLB team owner. Thoughts...

People seem to love Mark Cuban because he lives a more public life than most owners and is apparently friendly to sabermetrics (at least, based on his basketball ownership this seemed to be the case).  I think most of the excitement about him is just hype.  He's clearly very good at manipulating public opinion.

I really don't have a strong opinion on him either way.  My goal for major league baseball would be to make sure that whoever is allowed to buy the team would buy it and run it as a viable business rather than breaking it into pieces and extracting all of its value like the McCourt's did with the Dodgers.  I'm not in a position to know whether Cuban would be good in that respect or not.  Probably.  But is he the best option for the Dodgers?  I have no clue.

5. You’re a GM of a team. You can make a deal for one player for your team, no restrictions. Who ya got and why?

Is this completely open ended, i.e. pick the player you'd most want in baseball and don't worry about who you have to send away to get him?  If so, I'll go with Evan Longoria.  He's probably the best (offense + fielding) third baseman in baseball moving forward, and his position has become unbelievably shallow.  And he has the most team-friendly contract in the history of major league baseball.  The guy is in his fourth year, and is making $2 million.  Next year, he's making $4.5 M.  He'll make a total of $12.5 million over the next three years, and then the Rays have three MORE years of very reasonable club options at $7.5 million, $11 million, and $11.5 million.  If you're going to build a team around a guy, get the guy who is both awesome at baseball and is locked into a deal that is waaaay below market value.

The Reds

1. We’ve all thought (or most of us anyway) that the Reds are just a move or two away from repeating as NL Central champs. What are those moves?

I think we're all going to have the same answer: starting pitching and shortstop. I don't know if Zack Cozart is going to hit--hope so, but he's not elite-level talent. I do wish we would have had a few months to figure that out. But shortstop has been a huge black hole in the line-up, and upgrading to someone like Jose Reyes could mean a 3 win boost to the Reds in just a half-season. That's an enormous upgrade.

The Reds could also use a top of the rotation starter. I think the Reds' rotation is better than it has shown thus far, so if it's going to be upgraded I think you really need a top-flight talent. A #3 starter isn't really a meaningful upgrade as I see it, because the Reds may well get that kind of performance in the second half from the guys they already have. Erik Bedard should be available, but he's hurt (though with a leg injury, not arm), expensive, and will only have a start or two before the trade deadline to show he's healthy.  

Rumors surfaced yesterday that the Reds are interested in Ubaldo Jimenez. I don't think the Rockies would be interested in trading him, as he's done well in Denver, has a good contract, and the Rockies should be good next year. But if he can be had, I'd give pretty much anything in the Reds' farm except for Devin Mesoraco for him. Would Leake/Grandal/Alonso/Frazier do it? I'm not sure, but I'd pay that for a guy like Jimenez. 8+ career k/9, 50% career GB%, and an acceptable walk rate? Yes please.

2. Most frustrating aspect of the 2011 Reds is...

The fact that their rotation has just not come together like it should have. This is a talented group of pitchers, and they have as a group underperformed both their projections and their peripheral statistics. I expected the Reds' starting depth to overcome their lack of an ace, but instead of a bunch of #3's we've had a bunch of #4-5's. That's the biggest difference between what they've done and what I expected to see this year. Maybe we'll get something else in the second half...hope so.

3. Assign a letter grade for the following (and yes, you can explain your grades):

a. Dusty Baker: B. I really don't hate the guy like so many seem to. I think he's a terrific clubhouse presence, and has done a good job of defending and supporting his young players throughout his time with the Reds. While I disagree with some of his moves, I haven't been consistently frustrated with something this year like in years past. I'd like to see Heisey get a lot more PA's than Gomes over the rest of the season, and I'd like Cordero to not pitch when the Reds have more than a three run lead (warmed up or not). But those are really my only complaints.

b. Walt Jocketty: B-. I don't like making moves for the sake of making moves, and Jocketty isn't someone who does that. I think they've done well in using a patient approach to handle both Volquez's struggles and Chapman's. Willis has looked like a good find, and I think the Reds' outfield situation has been a best-of-a-bad-situation thus far. He hasn't made any desperation trades (yet). I think you have to give him credit for building what he has, and sticking with it. But I doubt I'll ever understand what the heck took so long with Cozart. That move came at least a month late. I don't know if it really will matter, but it's frustrating. I also would hope we'll see him make a move soon to bring in some talent...this grade is based on the assumption that he will do something.

c. Bob Castellini: A. For the most part, he's cheered the team on and kept his bottom out of the front office's business. That's what I want from an owner.

d. the Reds fans: C. I'll just say despite it all, we're four games back people!

4. You’re Walt. Brandon Phillips: extension or no? Why?

I love Brandon Phillips. Good bat, good fielding, important position. So yes, I'd be interested in extending him. But the question is the price tag, right?

Without doing anything rigorous, I'd project Phillips to be something like a 3 WAR player next year in his age-31 season. If we do a five year contract extension, and apply a "standard" 0.5 WAR/yr aging curve, and include inflation, Phillips would be worth somewhere in the vicinity of five years/$60 M. That's a big chunk of change, but he's probably worth that price. Assuming there's room in the Reds' budget for it...

That's pretty similar to Dan Uggla's 5 yr/$62 M deal, of course. I'm sure that deal will be front and center in Phillips' negotiations, and so far, that's not going very well for the Braves (which might be good for the Reds). But Phillips is a very different kind of player than Uggla: Phillips has some power, but is also very athletic and gets a lot of value from his fielding. Hopefully he won't drop off a cliff like Uggla apparently has.

Even if not extended, I pick up Phillips' option. If he's a 3 WAR player next year, he's probably worth $16 M or so, and the Reds would "only" have to pay $12 M. It's a pretty obvious move, I think.

5. It’s highly unlikely Francisco Cordero will be back next season (and I suppose a decent portion of the fanbase is glad for that), who’s the closer for 2012 and going forward?

Who's available as a free agent? I could survive with Nick Masset or Logan Ondrusek as a closer, but ideally you'd like someone who has a bigger arm. Maybe the Reds could acquire someone like Matt Thornton...?

I'm sure the Reds will probably look to Aroldis Chapman to be the closer next year provided he doesn't stink in the second half. They shouldn't. See below.

6. Should Aroldis Chapman be in the minors working on becoming a starter (provided you did NOT say Chapman was the answer for #5)? Is the starter’s role where he would benefit the organization the most?

Yes.  If a pitcher has a shot to be a successful starter he should be used as such until such time that he proves he's incapable of doing that.  Even average starters are as valuable as most of the better closers.  Closers pitch during important times in ballgames, but the number of innings they throw is so small compared to the number of innings starters throw that it's really no contest.  If a closer throws 70 innings and a starter throws 210, the closer's innings would need to be THREE TIMES as important as the starter's.  We can measure that with Leverage Index, and, in general, closers' innings are roughly twice as important.  

Chapman should be starting.  That would give the Reds a guy who is capable of being a top of the rotation pitcher if he can find a change-up.  Don't the Reds have to see if that can happen?

7. A silly question, really. Who wins a Reds only HR Derby?

Bruce!  Votto's got tremendous power, but I think Bruce has him beat in terms of raw muscle.  I'd love to see this.  The Reds should do this sometime...

Who is an All-Star?

Tuesday, I had a piece run at Hardball Times that used projections to pick the participants in the All-Star Game.  The premise is to get away from just using 2-3 month samples to select our players, and instead use projections to pick the players who have the most talent.

One alternative to doing this, which came up in a round-table discussion between myself and Dave Cameron at FanGraphs on this subject, is to instead choose players based on past calendar-year performance.  The advantage here is that it's a bit more straightforward: assuming we can properly measure everything that happens on the field (that's a big if!), all a player needs to do is produce better than the other players in the league at his position to get the nod.  We don't have to worry about aging curves and regression and luck.  Just produce, and you'll be rewarded.

Here's a comparison for how the three approaches would differ (and be similar), comparing my projection-based roster to one according to FanGraphs' WAR Calendar-Year splits, as well as Vince Caramela's first-half All Star selections.

American League

Position	True Talent All Star	Past-162 G All Star	First-Half All Star
C	Joe Mauer	Victor Martinez	Alex Avila
1B	Miguel Cabrera	Adrian Gonzalez	Adrian Gonzalez
2B	Ben Zobrist	Ian Kinsler	Howard Kendrick
3B	Evan Longoria	T-Adrian Beltre & Evan Longoria	Alex Rodriguez
SS	Yunel Escobar	Alexei Ramirez	J.J. Hardy
LF	Josh Hamilton	T-Brett Gardner & Josh Hamilton	Josh Hamilton
CF	Curtis Granderson	Curtis Granderson	Curtis Granderson
RF	Jose Bautista	Jose Bautista	Jose Bautista
SP	Felix Hernandez	Justin Verlander	Justin Verlander
RP	Mariano Rivera	Jonathan Papelbon	Jonathan Papelbon

While you see different names in many of the positions, if you go back and look at my THT article, every single one of the FanGraphs' WAR picks are within the top tier of the projections.  The same is true if you look at the FanGraphs' WAR leaderboards.  Most of the time, even when there's disagreement, the true talent selection is very close.  Heck, at 3B and LF, they were in a tie with someone else!

On the other hand, we do see some putative small-sample selections in the first-half All Star picks.  Alex Avila's an interesting player and has had a great first half, but he ranks 7th by projections and is in a tie for second by the past-162 splits.  Howard Kendricks' even more extreme, ranking 11th by projections (behind even guys like Nishioka and Jason Kipnis) and 4th by past-162 game WAR.

National League

Position	True Talent All Star	Past-162 G All Star	First Hall All Star
C	Brian McCann	Brian McCann	Brian McCann
1B	Albert Pujols	Joey Votto	Prince Fielder
2B	Chase Utley	Rickie Weeks	Rickie Weeks
3B	Ryan Zimmerman	Chase Headley	Ty Wigginton
SS	Troy Tulowitzki	Troy Tulowitzki	Jose Reyes
LF	Matt Holliday	Matt Holliday	Matt Holliday
CF	Andres Torres	Michael Bourn	Matt Kemp
RF	Jayson Werth	Mike Stanton	Lance Berkman
SP	Josh Johnson	Roy Halladay	Roy Halladay
RP	Mike Adams	Carlos Marmol	Craig Kimbrel

There were two fairly major disagreements here between the projections and the past calendar year picks.  First, Chase Headley at 3B by FanGraphs' WAR, who ranked 10th in the projeciton rankings (Zimmerman ranked 2nd by FanGraphs' WAR).  The second was Michael Bourn in CF, who ranked 12th in the projection rankings (Torres ranked 6th in the WAR rankings).  Beyond that, again, there was good agreement pretty much across the board.

Among the first-half All Stars, the biggest outlier pick is Ty Wigginton, but Chase Headley would be my selection based on the extra playing time this season if nothing else.  You can also complain about Lance Berkman, although he was effectively in a tie with Jason Werth for the projection's choice...and a lot of his merit comes down to how good (or bad) you think his fielding is.

In the end, I have no argument with people who want to use past-162 game splits.  In most cases where they disagree with projections, they do so within the margin of error in either measurement or projection (or both).  Where they disagree, however, I side with the projections.  They seem more obviously to be selecting players with the best talent, and those are the players that I prefer to see make it to the All Star Game.

But insisting on using first-half statistics to justify an all-star game roster?  There, I just don't get it.

Regarding Mudgate: I will read from a prepared statement

I did it. It's all my fault. Yes, the attendant rubbed the balls correctly and thoroughly. But I cleaned them. With soap. And a Snuggie.

I deeply regret my actions and apologise to my family, friends, and the baseball community. I will accept any consequences that MLB/Dave Duncan/Tony La Russa deem appropriate.

This concludes my statement. On the advice of my lawyer, I will not be taking questions.

Thank you.
Justin

More on judging deals at the time they are made

Fantastic interview with Chris Antonetti that talks a lot about how the Indians evaluated risk prior to inking Travis Hafner. It's very relevant to my last post about problems with post-hoc analyses. Here's an excerpt:

So then, you must feel perfectly good about that whole process.

Yeah. There are obviously different degrees of outcomes in any decision you make, and I think what we try to look through and look back on the process, and how complete and how effective the process was, irrespective of the outcome. You could have a very good process and still bad outcomes. I'm not saying that's necessarily the case in Travis' case - again, our expectation and our belief is that he's gonna come back and with his physical issues behind him will go back to being a very, very productive major league hitter.

Well ... come on, I think you could say that it has not worked out at least on the medical side the way you hoped it would.

Oh, certainly, so far, that's correct, yeah.

So then, would it be fair to say that, with the management team currently in place, if faced with the exact same set of circumstances again, you go ahead and sign that same contract?

With the information we had at that point? Yeah. Yep. I think that's fair to say.

So from the day you signed the contract until now, a season and a half later, realistically, the contract that that player could command on the market has gone down precipitously, and yet you would look back at that and say, "There's nothing about that experience that would make us change our process at all."

With the information we had at point, no. Yes, I would say that. I would say that's correct. We are comfortable with the process we had to arrive at that decision.

I also thought that this last bit was an interesting comment.

But you've concluded that the process was as good as you thought it was?

And as good as we could have done with the information we had at that point. I think you can get into bigger questions about team building, about committing significant dollars in a market our size to a player at that end of the defensive spectrum. That's a different strategic question than what we thought of the process arriving at the decision and the risk associated with signing Travis to a long-term deal.

Interesting point about market size & player type. I guess you can make the argument that a guy at the far right of the defensive spectrum (1B or DH) is riskier simply because all of his eggs are in one basket. A shortstop might decline offensively, but still provide defensive value. If a 1B/DH doesn't hit, they have no value. That said, offensive projections are more reliable than defensive projections, so that might counter things a bit. Neat idea, anyway.

FWIW, I still think the Reds need to buy out Votto ASAP while he can still be signed for a massively below-market contract. :)

Hat tip to Tango.

My All-Star Ballot

Albert Pujols is better than everyone else. Image via Wikipedia

Sky recently did a piece at BtB using the past 2.5 years to select the All-Star team. My personal predilection is to instead use the last calendar year for these selections, as this lets you better recognize new talents while still allowing second half-performances to matter. Unfortunately, FanGraphs currently can't give a calendar-year split for value, but it can do the past two seasons. So, here is my ballot based on those numbers (exception: relievers, who I'm ranking based on WPA), with weight given to more recent performances when it is close.

Position Players

American League

C: Joe Mauer (9.3 WAR) - With fielding included for catchers, he'd rank even higher. Joe is one of the premier talents in baseball, and as long as he's healthy, he's an MVP candidate. Alternate: Mike Napoli (4.2 WAR).

1B: Mark Teixiera (9.3 WAR) - Probably will become a legend now that he's in New York. Alternate: Kevin Youkilis (8.3 WAR)

2B: Dustin Pedroia (8.4 WAR) - Best little guy of his generation, right? Alternate: Brian Roberts (6.9 WAR), barely, over Ian Kinsler.

3B: Evan Longoria (8.8 WAR) - I have a modest man-crush on this guy. Alternate: Alex Rodriguez (6.9 WAR).

SS: Derek Jeter (6.2 WAR) - I've spent a long time dismissing him. Lately, I've started to appreciate him. A little. Alternate: Marco Scutaro (5.5 WAR!).

OF: Matt Holliday (7.8 WAR) - Hasn't done much yet in the AL, but he'll come around. He's too good not to, Coors Field be damned.

OF: Alex Rios (6.8 WAR) - A surprise, but he ranks where he does mostly due to his fielding. I love having great fielders in the corners, and his bat seems to be heating up.

OF: Ichiro Suzuki (6.5 WAR) - A perennial favorite, he's off to an amazing start. He'd play CF on my AL squad.

OF Alternates: Grady Sizemore (6.6 WAR) and Nick Markakis (6.5 WAR). Both are off to a slower start this year than last, which is why Rios & Suzuki get the nod for the starting spots, despite essentially a 4-way tie for second.

National League

C: Brian McCann (7.9 WAR) - I think he's underrated, and I don't know how that's possible. Alternate: Geovanny Soto (5.0), barely over Russell Martin, mostly for fielding.

1B: Albert Pujols (12 WAR) - Simply the best. Alternate: Lance Berkman (8.0 WAR).

2B: Chase Utley (11.4 WAR) - Mr. Underrated is so good, he's only a half-win behind Pujols. Alternate: Brandon Phillips (5.0), barely over Dan Uggla, due to his 2009 numbers.

3B: David Wright (10.3 WAR) - He'd not overlooked, but might be underrated. Alternate: Chipper Jones (9.2 WAR).

SS: Hanley Ramirez (10.4 WAR) - His defense has apparently improved at least to league average, which makes him god-like. Alternate: Jose Reyes (6.6 WAR).

OF: Carlos Beltran (9.3 WAR) - Is it just Steve Phillips, or do the Mets actually not know what they have with this guy?

OF: Manny Ramirez (7.8 WAR) - Sorry, Bud, I still think he deserves it

OF: Jason Werth (6.9 WAR) - This one was a surprise here, but largely on the back of his excellent fielding, Werth ranks 3rd among all NL outfielders in WAR. Why do I still think of him as a platoon player?

OF Alternates: Mike Cameron (6.4 WAR) and Ryan Braun (6.3 WAR), barely over Matt Kemp (6.3 WAR). Cameron gets the nod because he's done it in less playing time, but Braun vs. Kemp was a toss-up. I went for Braun mostly because hitting stats are more reliable than fielding stats, and Braun's the better hitter.

Pitchers

American League

Starters

Roy Halladay (11.6 WAR) - second-highest WAR in baseball over past two seasons.

Cliff Lee (10.3 WAR) - Picked up where he left off last year

C.C. Sabathia (9.7 WAR) - The Yankees have a new horse.

Zack Grienke (9.2 WAR) - Joe Posnanski has a terrific article on Grienke in this month's Baseball Digest. Sounds like a strange guy, but obviously he's exceptionally talented.

Relievers

Frank Francisco (3.76 WPA) - Yes, I know he just went on the DL. But Frank leads everyone in WPA over the last calendar year.

Mariano Rivera (3.25 WPA) - there are whispers that he's starting to decline, but Mariano's still ranked second in credit for wins than any other reliever over the past calendar year.

Brad Ziegler (3.15 WPA) - the submariner has put up outstanding numbers since his arrival in Oakland, prompting some to ask whether other pitchers might have similar success by dropping their arm angle.

National League

Starters

Tim Lincecum (11.0 WAR) - The Freak. I wonder how many young pitchers are trying to model their delivery after his these days. Have I mentioned that the Reds took Drew Stubbs instead of Lincecum in 2006? I have? What was that about a dead horse?

Dan Haren (9.0 WAR) - With Webb on the DL this year, Haren's become the unquestioned ace of the Diamondbacks staff. I remember thinking he might regress after the trade, but he's just pushed his k-rate higher.

Javier Vazquez (7.8 WAR) - The man who always seems to underperform his FIP is still a really good pitcher. This year, he's got his k-rate over 11 k/9. Wow.

Chad Billingsley (7.0 WAR) - Chad has gotten better every year in the big leagues, and is now the staff ace of the best team in the National League

Relievers

Jonathan Broxton (2.90 WPA) - Broxton is just a beast.

Francisco Rodriguez (2.86 WPA) - K-Rod has an 0.57 ERA and has allowed 17 hits in 31 1/3 innings. I don't care what his BABIP is, that's ridiculous.

Ryan Madson (2.62 WPA) - Big, strong, and this year is averaging 95 mph on his fastball

Red Reporter interviews Reds Asst Director of Baseball Ops Nick Krall

Really exciting interview today by Slyde of Nick Krall of the Reds' front office. Nick reportedly was an extremely nice guy, and the two yammered on for quite a while. Some version of that conversation is what appears at Red Reporter today.

Slyde was nice enough to solicit some questions ahead of time from various people. Here's one of mine and Nick's response:

RR: The Reds have made a few player personnel decisions over the last 6 months that appear to have been in favor of defense over offensive players. What has been the organization's reasoning behind these moves?

NK: Let's use Jay Bruce as an example. Bruce played centerfield last year. Jay Bruce statistically wasn't a great centerfielder, but he was good in right. So he goes from playing out of position to playing his normal position and Taveras comes in to play centerfield. If an average fielder catches 88% of the balls in centerfield and you get around 300-400 balls in centerfield, Willy Taveras is at 90% - he catches 2% more of the balls. So say he catches 8 balls a year more than the average, assuming Jay Bruce is an average centerfielder. If Willy is catching 8 extra balls, that's 8 extra at bats that we'll see. Two or three of the following hitters will get on base because of those at bats, so that's 11 total extra at bats that Willy has probably prevented. Now you add in the other guys that are out there (Bruce, Dickerson, Hairston, etc.) and compare them to the players that were previously out there and you could have upwards of 75 at bats saved in the outfield. We've got a flyball pitching staff - that's no secret. You may get 50 extra balls caught, but it's not just those 50 extra balls that matter. It's the .330 on base percentage after that plus the .330 on base percentage after that. It's not just those outs, but the effects of those outs on limiting the number of overall at bats.

If you are eliminating 75 at bats during the season, it means that fewer pitches are thrown, which means the starter can go longer in the game. If you look at the team in 2006, we had relievers who threw a lot of innings before the All Star break and then by the second half they were worn out or hurt. You are putting a lot of pressure on the relievers. So, improving the defense actually improves your pitching staff because you make more outs and you don't have throw as many pitches. You don't have to get a reliever up and sit him down as many times because the pitcher on the mound has things under control and so the reliever gets a real day of rest. That's where our defense has really improved in that way.

Ok, don't try to follow the math too carefully--this is a transcript of a verbal conversation, not a written argument. But I think the big picture point is right-on, and is something we've been talking about a lot: better fielding can have massive, pervasive effects on innings. I think we all know this intuitively when watching games, but many folk often seem to forget this when evaluating players. Now that we finally have some decent tools to assess defense, we can better understand how important it is.

Also, it's worth mentioning that the runs saved defensive estimates used here, on FanGraphs, etc, do take into account the "inning killer" aspect of better fielding that Krall talks about above in addition to just preventing a play. Tango has a nice discussion of this here.

Basically, the way it works is that the average single is worth ~0.5 runs. That's the additional runs scored you'd expect to get in an average inning that has a single vs. that same inning before the single. So, if you have a guy with excellent range in the outfield who prevents that single, you prevent that 0.5 runs. But in addition to preventing the single, you also caused an out! That extra out reduces the number of runs scored in an inning by, on average, 0.3 runs, because it prevents other players from coming up to hit as Krall describes above (Tango's "inning killer" component). So the total effect of the great play that prevents a single is 0.5 + 0.3 = 0.8 runs. This is where the shorthand 0.8 runs/play conversion that you'll sometimes see used comes from, though the one used in bUZR is specifically honed to the specific position, and I think also the zone, how hard the ball was hit, etc.

FWIW, the Reds' outfield currently sports a cumulative bUZR of +15.6 runs, which is on pace for +68 runs saved at the end of the year. Last year, outfield bUZR was -33 runs. In 2007, it was -31 runs. We're talking close to a 100 run difference (if they can keep on pace, of course), which equals ~10 wins. Don't get me wrong, I miss Dunn's bat...but I don't miss his glove, and I especially don't miss Junior's glove.

Congrats again to Slyde for scoring this interview! It's something he and I (and others) have talked about for a while, but to see it actually happen is really exciting! Hopefully it's just the start of a long, positive relationship between the Reds front office and the blogging community.

American League Update - Through 5/16/09

Tonight, I'm going to take a look at the junior circuit.

Division Roundups

Quick notes on the stats below: All runs data are park adjusted using Patriot's park factors. PythWins is from Hardball Times, and so is PythagenPat. XtrapWins is the wins that the teams' current winning percentage would provide at the season's end. Wif500 is what the team's record should be if they win half of their remaining games. %for90W is the winning percentage the teams will need to get to 90 wins at the end of the season.

East

Team	W	L	PCT	GB	RS*	RS/G*	RA	RA/G*	PythW	XtrapW	Wif500	%for90W
TOR	25	14	0.641	0	219	5.6	167	4.3	24	104	87	0.528
BOS	22	15	0.595	2	198	5.4	180	4.9	20	96	85	0.544
NYA	19	17	0.528	4.5	195	5.4	210	5.8	17	86	82	0.563
TB	18	20	0.474	6.5	203	5.3	189	5.0	20	77	80	0.581
BAL	16	21	0.432	8	186	5.0	216	5.8	16	70	79	0.592

As much as hate to admit it, this is probably the most interesting division in baseball right now. The Red Sox and Yankees should have very good teams, and the Rays--despite their record--have played every bit as tough as the rich teams. But the Blue Jays have been the class of the division thus far, sporting both the best offensive and the best defensive (pitching + fielding) performances. You have to expect some of Toronto's hitter's to fall back to earth--none of Barajas (0.820), Hill (0.927), and Scutaro (0.869) have OPS'd over 0.800 before this season. And they've gotten unreal pitching performances from Romero and Cecil. It will be interesting to see how long they can hold off their division rivals.

Central

Team	W	L	PCT	GB	RS*	RS/G*	RA	RA/G*	PythW	XtrapW	Wif500	%for90W
DET	19	16	0.543	0	190	5.4	163	4.7	20	88	83	0.559
KC	19	18	0.514	1	166	4.5	151	4.1	20	83	82	0.568
MIN	18	19	0.486	2	181	4.9	197	5.3	17	79	81	0.576
CHA	15	20	0.429	4	135	3.8	162	4.6	15	69	79	0.591
CLE	14	24	0.368	6.5	198	5.2	218	5.7	17	60	76	0.613

Another rather evenly matched division, though perhaps without the quality of the east. The Tigers have had the best offense, but the Royals have had the best defense--thanks in no small part to the spectacular first month and a half for Zach Greinke. Nice also to see Brian Bannister having a nice bit of success in the early goings. The Tigers have their own surprise--Brandon Inge is OPSing 0.954--but they have a lot of guys who aren't really hitting yet (Granderson, Guillen, Ordonez) and thus might actually improve even as Inge cools.

West

Team	W	L	PCT	GB	RS*	RS/G*	RA	RA/G*	PythW	XtrapW	Wif500	%for90W
TEX	22	14	0.611	0	200	5.6	174	4.8	20	99	85	0.540
LAA	18	17	0.514	3.5	177	5.1	179	5.1	17	83	82	0.567
SEA	17	20	0.459	5.5	152	4.1	179	4.8	16	74	80	0.584
OAK	13	20	0.394	7.5	145	4.4	162	4.9	15	64	78	0.597

The Angels have had as much bad luck and tragedy as any team I can remember, but still are right at 0.500 thanks their great depth. The Rangers, though, have taken advantage of the weak division, riding hot starts by Ian Kinsler, Michael Young, and Kevin Millwood to a nice little division lead. Not as impressive is the offense of Ken Griffey Jr.'s Seattle Mariners. Thank goodness they at least have Russell Branyon.

Team performance breakdown

Below is a kind of power ranking of teams based on their component statistics. I estimate team runs scored using linear weights (actually, FanGraphs does, though I park-adjust it), pitching performance based on FIP (I do park-adjust the HR's), and fielding based on UZR. I use those numbers to estimate team runs scored and runs allowed, and then use PythagenPat to estimate expected winning percentage. It's not perfect, but this should give us another look at team performance that gets beyond actual wins, losses, and runs scored or allowed.

			Offense				Pitching						Fielding			Overall
Rank	Prev	Team	OBP	SLG	wOBA*	wRC*	ERA	FIP*	K/9	BB/9	HR*/9	FIPRuns*	bUZR	THT+/-	DER	ExptW%
1	-	Blue Jays	0.359	0.460	0.356	225	3.97	4.23	6.8	3.1	0.9	181	6.2	24.8	0.720	0.619
2	-	Rays	0.346	0.452	0.357	214	4.83	4.83	6.5	3.8	1.2	191	16.3	0	0.689	0.595
3	-	Royals	0.335	0.421	0.334	174	3.63	3.76	7.5	3.4	0.7	149	-3.8	-4.8	0.685	0.560
4	-	Rangers	0.335	0.496	0.357	201	4.66	5.02	5.3	3.4	1.2	192	15.2	19.2	0.712	0.560
5	-	Tigers	0.340	0.426	0.337	170	4.31	4.29	7.3	3.6	1.0	160	4.7	8.8	0.704	0.544
6	-	Red Sox	0.364	0.451	0.351	204	4.82	4.67	7.4	4.0	1.1	185	-12	-7.2	0.677	0.517
7	-	Yankees	0.350	0.467	0.358	211	5.41	5.28	7.2	4.2	1.4	203	-4.1	4	0.691	0.509
8	-	Indians	0.355	0.420	0.345	204	5.62	5.03	6.6	4.0	1.3	199	-2.8	-12.8	0.673	0.505
9	-	Angels	0.344	0.414	0.339	174	4.70	4.61	6.0	3.4	1.0	171	-4.1	-10.4	0.690	0.497
10	-	Twins	0.347	0.417	0.338	183	5.20	4.91	6.2	3.0	1.4	192	-0.1	-0.8	0.696	0.476
11	-	Orioles	0.342	0.437	0.340	182	5.51	5.11	6.6	3.3	1.5	196	-9	-20	0.666	0.444
12	-	Mariners	0.305	0.378	0.305	138	4.25	4.46	6.8	3.7	1.0	178	10.6	-4.8	0.689	0.411
13	-	White Sox	0.317	0.385	0.306	132	4.71	4.05	7.0	3.8	0.7	150	-14.4	-19.2	0.667	0.398
14	-	Athletics	0.309	0.338	0.294	116	4.12	4.57	6.2	3.6	1.0	167	-0.2	-11.2	0.689	0.333

The only major disparity I see is with the Indians, who have apparently hit and pitched well enough to expect a 0.500 record, and yet are somehow 10 games below 0.500. Their true pythagorean record is 17 wins, but indications here are that they've given up quite a few more runs than expected. Are they just unlucky, or are the methods missing in this case? Part of the answer might be with the fielding estimates: bUZR rates them as only slightly below average, but THT's batted-ball based plus/minus stat (as well as straight-up DER) pegs them at second-worst in the league. Perhaps bUZR is overrating their defense?

What's up with the Athletics offense? They add Giambi, Holliday, and Nomah, and yet have a wOBA under 0.300? Beane must be blowing a gasket.

• Top hitting teams (wOBA*): Yankees, Rays, Rangers, Blue Jays, Red Sox (holy AL East!)
  
• Top pitching teams (FIP*):Royals, White Sox, Blue Jays, Tigers, Mariners
  
• Top fielding teams (bUZR): Rays, Rangers, Mariners (maybe), Blue Jays, Tigers
• "Expected" leaders: Blue Jays, Royals, Rangers
• "Expected" Wild card: Rays

Blue Jays rank in top five in hitting, pitching, and fielding. What will they have to do to keep their lead the rest of the way?