Baseball ProGUESTus

Why does the sabermatic love AAAA sluggers so much? Year after year I hear about those guys being touted as undervalued. Re-reading old BP articles shows an unbelievable arrogance on this subject.

Mr. Tango....what are your thoughts on modern bullpen alignment and usage? Should "closers" be used more liberally in key spots in the middle innings if indeed they're the best guy in the bullpen?

If there's an equivalent chance the starter or middle reliever gets the next out. Who should pitch to the batter? What if the starter or reliever has a slightly better chance?

I guess basically, Do the numbers show an effective usage pattern for season long maximization of your pitching staff?

What are the most harmful mistakes that teams make by ignoring sabermetric advice?

Also, how does sabermetrics make itself more relevant - it seems to me that sometimes more and more good work is done, only to be ignored by much of the media and baseball as a whole.

As an add-on to this - what proportion of save situations ought to be pitched by the best reliever? I suppose I was thinking about closers by committee, and that, while there is the odd situation where someone else ought to close (three lefties due up, or three run lead facing the bottom of the order or whatever), most of the time, you would just want the best reliever who hasn't pitched yet out there. Is the real problem then that closers should be used much more frequently in key situations earlier in the game, in the same way that the Red Sox appear to be using Daniel Bard this year?

Sorry, that was supposed to be a follow-up to lemppi's post further up. BP commenting is hard to do correctly.

In DIPs theory, does the variance of the rates at which pitchers give up doubles and triples look more like the variance of the rates of singles allowed or HRs allowed?

DIPs states that pitchers have no control over what happens after the baseball leaves their hand, yet statistically some pitchers are definitely "ground ball pitchers" and some are definitely "fly ball pitchers". Statistics also prove that ground balls and fly balls turn into outs MUCH more often than line drives (and ground balls more so than fly balls), so how can DIPs be true?

If a manager's ability to lose a game is far greater than his ability to win a game and if a manager with superior tactics/strategy will win only a very small number of games more than a manager with mediocre tactics/strategy, then does it makes sense to hire a manager who is a great teacher and motivator and who commands the players respect and for whom the players will always give 100%?

You could argue that the players ought to be self-motivating and should always give 100%, regardless of who the manager is, but I think that would ignore the fact that the players are human beings, not robots. It would seem like any managerial tactical or strategical deficiencies can be corrected or compensated by providing the manager with a set of rules regarding lineup, pitcher usage, when to sacrifice bunt, when to steal, etc (or would such guidance insult the manager?).

Ultimately, what the question boils down to is if you had to choose between the two, would rather have a manager with superior dugout skills (tactics/strategy) and mediocre clubhouse skills (player relations) or a manager with superior clubhouse skills with mediocre dugout skills?

I loved "The Book" ... will there be a sequel?

Statistical footprints of PED use -- is there any way to catch this great white whale? Steven Levitt was able to find statistical evidence of cheating in sumo wrestling and standardized test taking. Is there any method that might have some promise?

I’d like you to tell us the one thing you learned about women. Then I’d know one thing too!

You work for a couple of teams ... have any teams made substantial sabermetric discoveries that would be of high interest to sabermetricians in general?

That is, suppose you have a scale of sabermetric discoveries something like this (your rankings may vary, feel free to recalibrate):

Runs Created: 10
Voros and DIPS: 8
Players' aging curves: 6
Strikeout pitchers have longer careers: 4
Clutch hitting doesn't exist much: 2

And supposed you ranked teams' top five discoveries that the general sabermetric public doesn't know about. How would they rank on that scale?

Studies I have seen in the past state that "catcher's ERA" does not show any statistical significance.

Nonetheless, we constantly hear about catchers' role in calling a game and helping the pitcher.

Have there been any developments on this issue?

From a pitchers point of view, it seems that unpredictable randomness or "luck" factors into his results and "numbers" much more than previously thought. Especially since the pitcher is so reliant on things totally out of his control---run support, inherited runners stranded by the reliever, closer effectiveness, "fluke" hits, hard smashes right at fielders vs. a 5 bouncer that finds a hole, etc. Thoughts?

Riffing off of Behemoth's question: is it important that MLB and its fans and reporters recognize the value and findings of sabermetrics, or is it OK to be content to observe and learn?

Looking back historically, if I want to say something about a player's "true" woba on a particular date, using their woba performance for that date is a little noisy, so its not so great, even thought it is literally correct. I have been thinking it makes a little more sense to look at performances on *either* side of the date, before and after, and use some sort of model (probably time weighted) to make a "true" woba estimate for the player on that day. It seems like most people only use subsequent data, or haven't really addressed the question as I have phrased it. If this philosophical shift seems valid, it seems like this would warrant a shift in how we evaluate projection models too...What do you think?

Twenty years ago or so, Bill James said that the only predictor of future pitching performance he could find was K/BB ratio --- and that it was not particularly reliable. Has ensuing research unearthed anything more useful?

Questions about your optimal lineup construction from "The Book":
why bat your 5th best hitter 3rd instead of your 3rd or 4th best hitter? I am assuming it's to prevent the lineup from being too top heavy?
The other is why in the AL you bat your worst hitter 9th and in the NL you bat the pithcer (your worst hitter) 8th? Is this because of just how bad pitchers are at hitting that it becomes optimal to bat them 9th, and at what point would the worst hitter on an AL team have to be for it to be optimal to bat them 8th and not 9th?

What is your take on the recent shufflings of writers towards "the mainstream" (ESPN, CNNSI, MLB.com) and away from "the mainstream" to places like TheSweetSpot, THT, BP, etc.

It seems like owners, general managers and managers are constantly making personnel and strategic decisions that seem foolish on the surface. The reactions amongst fans range along a spectrum to "that person is an idiot!" to "they obviously know more than we do, there has to be a reason!"

My question is: From your experience, what are some of the "things they obviously know" that we, as typical fans, do not? In most cases, are even foolish-looking decisions truly rational when all factors (including those unbeknownst to the general public) are considered? Or is there truly just a lot of bad decisionmaking taking place on major league diamonds and in front offices?

Recently I watched Travis Snider go head-first into first base and thought 'Wow, that's really dumb, and dangerous.' And when the usual jammed thumbs and broken fingers result on head-firsts into second and third base I always think 'who started this stuff? Rickey Henderson?'
Why did foot first sliding disappear? I recall two types of foot-first slides. The hook slide away from the tag and the straight ahead foot-first of which I have fond memories of Lou Brock's classic form going into the bag, and popping up ready to take another base. Has head first sliding been proven in some way more efficient?

I'm very happy to see the ProGuestus series bringing BP's subscribers the benefit of so many respected points of view under it's one roof. (Best idea since BP-Idol, Ben!) Thanks very much for fielding our questions, Tom. I'm looking forward to your answers.

I appreciate your take on three questions, if you can do so without reavealing anything that your MLB employers would frown upon:

1) What sabermetric advancement do you think is the least-appreciated by a majority of franchises?

2) I like to flatter myself that I'm an 'early adopter' to the sabermetric perspective on the game, even though it's been so many years since its introduction and uptake by those like yourself. Is sabermetrics already 'mainstream' in your mind, or how long do you think it will be til it is?

3) What was / will be the tipping point to #2?

Thanks in advance.

Here's an exasperated thought: there is no "a lot" v. "alot" debate. "Alot" isn't a word unless it's this: http://hyperboleandahalf.blogspot.com/2010/04/alot-is-better-than-you-at-everything.html

Tom, thanks for creating new conversations. Outside of areas requiring enhanced technology (continued evolution of Pitch/fX, Hit/fX, Field/fX) is there an area of the game, particularly on-field events, that have not yet been studied for their effects/relevance?

1.) Why has Miller Park's runs park factor (from ESPN) fluctuated so much since 2007? How does ESPN compute park factors and do you think they are reliable?

Bonus Question: What are your thoughts on playing slot machines and if you played what would your strategy be?

Hi Tom- Not really a sabremetric question, but rather something that has always bugged me.

The Phillies are currently 27-17, which is called 10 games over .500
The Mets are currently 22-22, which is .500

Yet the Mets are considered only 5 games behind the Phillies.

Why do we interpret team's relative standings two different ways?

Most of the sabermetric work I've seen focuses on personnel decisions (i.e. the GM level) or in-game tactical decisions and assignment of playing time (i.e. the field manager level.) I'm always encouraged to hear about players such as Brian Bannister who have tried to incorporate sabermetric analysis into their play. Do you think there are opportunities for players to use analysis to improve their games, and if so, are there current major league players who are doing so already? Is it something limited to pitchers and catchers (pitch selection and such) or can hitters and fielders also take advantage of this work?

Ever since Voros McCracken made his stunning breakthrough observation, one bizarre and seemingly irreconcilable implication of DIPS has nagged at me. That is, that pitchers are "responsible" for a home run, but a ball drilled to the warning track and caught is thrown into the same Magic Soup of "ordinary batted balls" as the dribbler to the mound and the infield popup. A red flag of "discontinuity" goes up for me there.

Put more starkly, in one park a deep drive would be the pitcher's "fault" because the fence is a few feet closer - while in another stadium, DIPS says that there but for Lady Luck that same deep drive could as easily have been a harmless swinging bunt, since it stayed in the park. Now I realize that HR/FB normalizations are done in the more sophisticated defense-independent pitching stats (like xFIP). But how can an identical batted ball be purely the pitcher's "doing" in one park, but essentially the batter's "doing" in the other? I cannot get my head around this seemingly artificial construct.

I've noticed that sometimes my kids make the most random but insightful observations, on things that I completely missed. Have you ever had a baseball or a sabermetric insight based on something that your son told you that he noticed, that you did not notice?

The other day my wife asked me who gets "credit" for bases on balls -- that is, are walks generally the result of pitchers making mistakes, or hitters being patient? I told her that a generation or two ago bases on balls were almost solely blamed on pitchers. But we now know that walk rates for both pitchers AND hitters are fairly stable, suggesting that both parties are contributing to the outcome. Is there a way to determine HOW MUCH each side is contributing? For example, can we say something as definitive as: walks are 55% due to pitchers, 45% due to hitters? Further, can we give percentages to other factors, such as umpires, park effects (visibility, I suppose), even catchers? Or are those things pretty negligible?