May 23, 2013

THT Essentials:
About Us
Contact Us
Our RSS Feeds
Our Stats Glossary
Reference Articles
THT Reports
 
Fangraphs Player Search:
 

Rich Barbieri
John Barten
Kyle Boddy
Brian Borawski
James Gentile
Matt Hunter
Frank Jackson
Chris Jaffe
Brad Johnson
Jason Linden
Dan Lependorf
Bruce Markusen
Jeff Moore
Greg Simons
Scott Spratt
Dave Studeman
Shane Tourtellotte
Steve Treder

And here's the full roster.
 

Now available


You can now purchase the Hardball Times Baseball Annual 2013, with 300 pages of great content. It's also available on Amazon and Kindle. Read more about it here.
 



Or you can search by:
Category
Month
Author
 

THT E-book


Third Base: The Crossroads is THT's e-book, available for $3.99 from the Kindle store. The good news is that anyone can read a Kindle book, even on a PC. So enjoy the best from THT in a new format.
 



Get your very own THT merchandise from our CafePress store. We've got baseball caps, t-shirts, coffee mugs and even wall clocks with the classy THT logo prominently displayed. Also, check out the THT Bookstore. Please support your favorite baseball site by purchasing something today.
 


Creative Commons License
All content on this site (including text, graphs, and any other original works), unless otherwise noted, is licensed under a Creative Commons License.

On old school and new school values

by Chris Lund
October 31, 2012

The 2012 season has left us all in an interesting position with respect to what the baseball community terms as valuable. The AL MVP debate, a two-horse race between Miguel Cabrera and Mike Trout, has dug up a debate that appeared to have been partially buried as we trudge forward into the future. It may be completely ludicrous to many, but alas, here we are once again. The never-ending "old school vs. new school" tussle.

The discussion surrounding Cabrera as AL MVP was minimal until the Triple Crown became within striking distance. When he ultimately clinched it—the first player to do so in 45 years—the old school unanimously clamored for the simultaneous anointment of Cabrera as the MVP.

Not so fast.

On the other end of the spectrum was Angels wunderkind Mike Trout. Quite simply, Trout put together one of the greatest rookie seasons in baseball history, and one of the greatest single seasons in the modern era. Unfortunately, much of this isn't readily apparent and we must dig deeper and use advanced metrics to truly capture what a phenomenal year Trout had.

Wrapping up the season with these two stars pitted against one another in the spotlight set an interesting tone of analytical condescension throughout the playoffs. As someone who digested as much playoff baseball as I could muster, I can safely say that the opportunities for one side to take a jab at the other were not wasted.

The silliness of a commentator using the example of a seeing-eye ground ball beating a shift as a platform for demeaning the concept of statistics is almost impossible to express. Yet, unfortunately, this is the climate for discussion once again.

Quite simply, the climate for reasonable debate has become untenable.

One doesn't need to be well-versed in Zen concepts to understand that eventually the two parties will need to meet somewhere in the middle and from there on out they can live in harmony. Baseball has actually created an interesting platform for this false dichotomy as it is the first sport to truly have a pronounced and influential statistics-based community.

Basketball, football and hockey are all seeing those communities grow in North America while soccer's metrics are gaining more acceptance overseas, but insofar as math being used to aid and evaluate on-field products, baseball still reigns supreme.

As such, there is no proven method on how to weld these parties together and baseball will have to take the lead on the issue. Eventually sports and stats will live in harmony; they just need someone to show the world how.

While I don't know where or when this debate will resolve itself, I have an idea on how to get started and it has to do with altering rigid ideas of value.

From the old school perspective, it's time to acknowledge that stats aren't invented by pencil pushers. These are metrics extracted from watching games play out and breaking them down to ensure they properly represent what transpired. The notion that advanced statistics are fictional beings hellbent on fraying sport at its very fabric is a truly puzzling phenomenon.

In the case of baseball, these are events which take place in a baseball game. The fact that they are extracted and analyzed differently than convention has historically dictated changes nothing from game play to evaluation. The ability to get greater depth when evaluating a player should be welcomed as a good thing.

Statistics isn't a catch-all word for evil. Batting average, home runs and runs batted in are all statistics as well. They just don't paint as clear a picture of a player's ability as others do, though they do play a role. It's possible to give credence to old school stats while acknowledging new school interpretations are worthwhile.

For the new school, it's time to acknowledge that old school concepts do matter. They may not lend value to the purposes of advanced metrics, but they do affect the game itself and painting with a wide brush is a dangerous precedent, as we all know.

Take the concept of pitcher wins, for example. We know that, for the purposes of evaluating a pitcher's quality, wins are irrelevant criteria. However, to say that they simply do not matter disregards the weight they hold in baseball circles. Managers will try to get pitchers in line for wins, and pitchers regularly admit to altering their approaches in order to earn a win on their record. Ergo, pitcher wins matter because they affect games. That,, in turn, affects data.

The same goes for other traditional concepts like RBIs. While having more RBIs doesn't make one batter better than another, the quest to create RBI situations for particular hitters influences lineups and in-game strategy which, in turn, affect other, more pertinent offensive outputs. RBIs influence in-game decisions, which means they matter, despite the fact that they don't anoint one batter as more valuable than another as well as weighted runs created, for example.

As baseball trudges forward, its community of analysts and fans need to reach a resolution of some sorts, and this will require give and take between the two schools of thought. It's time to alter our definitions of what "matters" and accept that another half of the circle brings value on some level, even if it does not directly correspond to the information we are seeking. It's possible to name Mike Trout the MVP while acknowledging that the Triple Crown is a compelling accomplishment, and vice versa.

Flippant dismissals of one party by the other don't lend themselves to progress. Acknowledging that the other lends value, though not necessarily in the same ways, is a start and ought to be a focus going forward.

The sooner we can bring these two closer together, the better.

What do our readers think can be done to constructively weld the links between the old and new schools of thought?

Chris is a writer-at-large and encourages you to talk baseball.
For further baseball discussion, you can follow him on twitter under @thechrislund or send him an e-mail at chris (dot) lund89 AT gmail (dot) com

Comments

TomH said...

WAR is too complicated for a stattha most will be able to get a handle on. We ought to start but getting old-schoolies to acknowledge that OBP is far batter than AVG, and that scoring is as important as driving them in. That the ‘triple crown’ (TC) wasn’t drawn up by Moses (pre-Ruth it was irrelevant), and a better measure of offesnive breadth (what the TC tries ot capture) would be OBP, R and RBI, or OBP, SLG, and R+RBI, or OBP, HR, and R+RBI.

Old and new schoolers both already know that defense is impt.

Posted 10/31  at  07:34 AM
Beau said...

Tom—While the equation for WAR may be “complicated” (though is it, really?) I don’t think the concept of WAR is something that most baseball fans can’t wrap their minds around.  Granted you are right that things like OBP and OPS or even K/9 or WHIP might be better first steps I do think that WAR is ultimately something that the entire community could accept.

Posted 10/31  at  08:37 AM
obsessivegiantscompulsive said...

Frankly, I think many new schoolers have gone too far to the other side with the pendulum. 

Yes, OBP is very important, thus making the ability to take walks very important, but I think that becomes a short-hand that people state matter-of-factly, without remembering that it is very important to get hits too.  People forget that while OBP is better, and that taking walks is good and something that was overlooked before, it is not like BA and getting hits is not important too, as that affects both OBP and SLG positively. Getting hits is still very important.

I find too many people focusing too much on the ability to walk without also considering getting hits important too.  People see “OBP > BA” and you can tell by the way they argue that they dismiss BA totally, when it should be considered part of the toolset that any good sabermetrician should be using to evaluate a hitter.  Just because it is not as good does not mean that it has no value in comparing hitters, nor that it has no value in scoring runs.  It is not “either/or”, it’s “good and best”, used in appropriate doses.

I think the data here shows my point well:  http://www.eg.bucknell.edu/~bvollmay/baseball/runs1.html

This dataset shows the traditional analysis to show that OBP (and SLG) is a better measure of a hitter’s value than BA.  Only 0.672 correlation for BA, but 0.835 correlation for OBP, OBP trumps BA as a measurement of runs creation.  We all know this drill.

However, this analysis also captures on the bottom something that is never mentioned when discussing this:  only 0.404 correlation for walks.  Much, much worse than BA (and therefore hits), even worse in relationship to BA as BA to OBP.  Yet, any time I get into a discussion of any prospect, I get a lot of discussion about how good or bad he is at taking walks, while his ability to get hits is dismissed because the “OBP is way better than BA” mantra.

Getting hits is very important.  It is only that getting on base is very, very important.  Yet people treat getting hits as nothing to be impressed by, because “OBP > BA”, and touts a guys ability to walk while ignoring his ability to get hits.

And really, while getting walks is important, what is more important is a batters “walk/strikeout” ratio, as that correlates strongly with being a good hitter.  Getting a lot of walks is good, but it is much better if you walk as much as you strikeout, than if you strikeout 2 or more times your walks.

These people also forget a very elemental fact of baseball:  you can drive in a lot more runs with a hit than a walk, in fact, driving in runs with a walk is extremely rare, it is hits that rule the day in driving in runs.  They forget that while being on base is important, but so is driving in runs, and walks do not show up at all in SLG.  Driving in runs is a very important component of run creation - a statement anyone can acknowledge - yet when I get into a discussion about a player, his ability to get hits and thus drive in runs is totally dismissed.  And that is a large chunk of run creation.

Posted 10/31  at  12:28 PM
Marc Schneider said...

I agree with the previous comment.  Some sabermetric analysis act as if getting a walk is better than getting a hit.  If you hit, say .350, that’s important.  Similarly, I think sabermetricians downplay the value of psychological factors, such as leadership, that baseball people value.  While I agree that the notion is certainly overdone, it’s hard for me to dismiss the idea that confidence, leadership, morale, etc. play some role in performance. I think sabermetricians unduly dismiss the perspective of people who do, after all, play the game.

At the same time, baseball, especially, is saddled with a tradition of anti-intellectualism that leads players, coaches, and many broadcasters, at least-not so much GMs anymore-to dismiss any kind of analysis that seems counter to their observations.  The fact is, baseball players and coaches are probably the least educated of any sport and this probably has some impact on their willingness to accept advanced statistical analysis.

Posted 10/31  at  04:11 PM
Bob Rittner said...

I have 1 minor suggestion and 1 that I doubt will ever be implemented.

1. Stop the snark-on both “sides” (as if there really are two sides). It simply suggests arrogance and rather than encouraging people to consider an opposing view causes them to dig in their heels.

2. In all discussion/debate, recognize and dignify opposing views and arguments that have merit. Rather than trying to “win the argument”, or show off one’s superior grasp of the issue, consider it an effort to find some consensus-or some common ground from which to augment everyone’s understanding.

True, on all sides there are some people whose lack of reason or information make their comments difficult to digest. But even there, as in just about all cases, there is an element of truth (however small) in all error and an element of error in all truth. Pick it out and use it rather than simply dismissing the entire argument.

Even in cases where someone is simply rejecting all your arguments out of hand, it is possible to discern what underlies the irrational hostility and recognize its roots which may be, if not legitimate, at least understandable. And you can often defang the hostility without being condescending.

But even more critical, most arguments are multifaceted and have some legitimacy. It is useful to applaud such points while indicating why you think another argument is stronger (not necessarily perfectly right) or why, despite its legitimacy, that argument has some weaknesses.

By dignifying each others reasoning and points of view, we can discuss rather than quarrel and listen rather than reject. There is no winning for one, but there is potential growth for all.

Posted 10/31  at  04:44 PM
Marc Schneider said...

Bob,

I wish your view would prevail in our politics as well.  I think the anonymity of the Internet encourages the problems you describe.  Most people would behave a lot differently if they were talking to someone in person.

Posted 10/31  at  05:03 PM
David said...

TomH—

I found your comment inspiring, so I spent some time today figuring out how many times each of those things has been done.  I did it by hand, so I might have made a mistake here or there, but I kind of doubt it.  Anyway, it’s A LOT more that any of those “Triple Crowns” have happened.  The fewest was 19, of the OBP/R/RBI variety.  Then 45 of the OBP/SLG/R+RBI, and 32 of the OBP/HR/R+RBI variety.  So there you go.

Posted 10/31  at  11:30 PM
Jason W. said...

“For the new school, it’s time to acknowledge that old school concepts do matter. They may not lend value to the purposes of advanced metrics, but they do affect the game itself and painting with a wide brush is a dangerous precedent, as we all know.”

Do statheads not acknowledge these things? And what’s the value in so acknowledging even if they don’t? Or, more particularly, what’s the value vis a vis “the other side”? Statheads must acknowledge reality, yes, and reality includes managers making decisions because of outdated statistics that do not actually add to the odds of winning the game. But how does our acknowledging those things bring us closer to the middle ground or the synthesis or whatever it is that you’re proposing?

And for that matter, why is the middle ground so great? Pitcher wins are flat-out wrong. What’s the point of the stathead community moving even an inch in that direction?

Posted 11/01  at  04:31 AM
Bob Rittner said...

“And for that matter, why is the middle ground so great? Pitcher wins are flat-out wrong. What’s the point of the stathead community moving even an inch in that direction?”

I don’t think it is a matter of accepting the middle ground. Rather it is important to consider all views-seriously consider- in determining the best understanding.

For example, is there no value in considering pitcher wins? Don’t they suggest, at least for starting pitchers, the ability to go at least 5 innings per start? Doesn’t it indicate something about a manager’s confidence in a pitcher-at least in some cases? Perhaps not critical points, but why dismiss them without consideration when they might lead to other insights? After all, perhaps a manager’s confidence in a pitcher might be one of those intangibles that leaves tangible traces that can be studied.

Acknowledge does not mean “agree with” or “accept”. In discussion, it is a means of encouraging open minds rather than ideological purity.

What is useful is moderating the tone of the discussion, not the passion for seeking understanding. And rather than simply dismissing those views or arguments you consider wrong-headed or superstitious, to posit alternatives that may even include whatever kernel of validity exists in the opposing view-when that is possible.

And I think there are very few cases where it is impossible. It seems to me Chris addresses your concern about acknowledging “outdated” ideas in the paragraph that ends “Ergo, pitcher wins matter because they affect games. That, in turn, affects data.”
and in the following paragraph about RBIs. Since using such ideas affects games, they are part of the reality we are trying to understand.

John Stuart Mill defended free speech not simply because it permitted us to exchange error for truth but because it allowed us the “livelier impression of truth from its collision with error”. We can’t have that collision by talking (yelling?) at each other, only by mutual respect that true conversation requires-and only by the modesty to recognize that some of that error may be ours no matter how certain we are.

Posted 11/01  at  06:23 AM
hopbitters said...

I think the biggest problem is that stats on both sides are misunderstood and misused.

Average is a great statistic. It’s simple to calculate. It has an obvious and direct relation to what happens on the field and it tells you exactly what its name implies. And like it or not, it is an indicator of offensive performance. It’s not a perfect indicator, but that’s why it isn’t called “absolute offensive value rating”. It is what it is and if you ascribe your own biases to it, either way, then that’s your own doing and not a reflection on the stat itself.

Pitching wins, which I admittedly have come to loathe, are very clearly defined, but somewhat less directly related to a single event or set of actions by the pitcher. Nonetheless, they describe a very specific situation that has occurred on the field. The problem comes when people use it as an indicator of value or performance, which it is not. Again, the stat is accurately describing what it claims to describe, but the near-universal (mis)usage of that information causes the acrimony.

There are similar examples on the sabermetric end. WPA is one that I think is routinely abused by certain media. WAR seems to be becoming more mainstream and will undoubtedly be warped beyond recognition before long.

Numerous other examples exist in both realms. I don’t think either side should blindly accept deceptive use (intentional or not) of stats and they should question the value of them, even in the correct context. Perhaps, a little less bitterly.

Posted 11/01  at  11:31 AM
Andrew in Toronto said...

I think the description of what the new school needs to acknowledge does not go far enough.  It sounds like you are saying that the new school only needs to recognize that, despite the wrongness of the old school, they still influence things in the real world and so should not be forgotten.  See Jason W’s comment, for example.  That’s more patronizing than compromising.

Rather, a middle ground is more likely if the new school recognizes the value that actually does exist in the old school way of looking at things.  Because there is some value.  You just have to change your point of view.

Take RBI for example.  We all know the shortcomings of this statistic.  Essentially, RBI says as much about the availability of opportunities as it says about talent.  The old school loved this statistic but the new school realized that it didn’t say much about how good a player was nor was it very useful in predicting future performance.

Note that the new school view here is focused on analysis from a forward-looking point of view.  How useful is this stat in helping me decide who I’d want to put on the field tomorrow, or next season?  Who is “better”?

But there is another point of view, which is not forward-looking but backward looking.  It is about telling a story about what has actually happened.  It answers the question of what events took place that mattered, regardless of why they happened?  An RBI is a run batted in and that’s an event that actually mattered.  The score changed and the fans stood and cheered.  It was important and you can’t blame people for wanting to count how often it happened.  Where they make a mistake is in what they infer from that RBI tally.

Posted 11/01  at  01:28 PM
Bob Rittner said...

I think you are exactly right Andrew although I will quibble about the phrase “middle ground” with a similar objection to the one Jason W. made. Perhaps this is getting into semantic quibbles, but I do agree that the idea is not to find some middle ground as if the greater truth lies midway between positions but rather to incorporate what is correct in all positions-and even to use that with which we may disagree to widen our vision and sharpen our convictions.

Posted 11/01  at  03:26 PM
rubesandbabes said...

For me, the problem is not the New School which has been around for some time now - Bill James, Ron Shandler, and Nate Silver all getting real jobs coming from the New School, naming a few.

It’s the combination of the fact that there are suddenly like half-a-dozen new methods of measuring the components, and the crappy ideas of people that have always been there on both sides who throw around the stats without really grasping them. Two typical examples:

The use of WAR to explain away guys like Jack Morris and Miguel Tejada, for different reasons, but also for lack of any real argument otherwise. Younger fans simply and naturally don’t have the same grasp of the stars of a previous era as they do for the stars of the present era. Tejada is one of the top SS’s of all time, and it is common for him to be dismissed because his defensive stats work against him somehow, even though he played many years and was paid $6mil as recently as 2011 to play SS. Dismissing Tejada on defensive grounds using the stats as a convenience, but also ignoring the context of MLB during this time..Okay, Tejada is a sly example, but hey check that all-time HRs by shortstop leaderboard, and quit ignoring those 150rbi years, please.

The basic fact that every time a new school guy tries to explain a “sunk cost” in baseball salary terms, they just fail. People, yes you read Moneyball, so please get it right: Just because Hanley has a huge sunk cost going forward, doesn’t mean the Marlins shouldn’t have tried to make lemonade out of a lemon by trading him and his forward salary obligations. Probably the Marlins will keep going along beneath the surface, so it’s hard to call anything they do a success, but this Boston bloodletting probably saved Ben Cherington’s job, and likely will impact the AL east greatly before too long.

==

All the different stats old and new grind the components - the stats are a help - but being familiar with the components is the most helpful.

Oh, and hoping WAR loses prominence - it is a circular stat that passes out the points for just taking the field at some positions, but not others, and pretty much not all that useful when talking about all-star and HOF players, who hopefully can field their positions.

Posted 11/03  at  12:43 PM
rubesandbabes said...

Yes, the other comment I would like to make concerns the idea of the “Replacement Level Player.”

Watching the game, it is just very clear to me that the imaginary guy in the WAR stat is just a very different animal than the true replacement level player on the field.

It is possible for a replacement level player to have success, and we see really a wide range of levels of contributions from these so-called players from the fatter part of the bell curve.

Take a good look at the rosters of the more budget conscious teams, where replacement level players abound. But WAR says if they play enough they start to accumulate value over themselves - it just doesn’t make sense.

Posted 11/03  at  12:52 PM
staiton said...

There is one area that the old school has always been superior to more modern analysis: the point of baseball is to win, not acquire the most cumulative team WAR. A lot of people seem to forget that these statistics are an attempt to predict a player or a teams future value, be it for a season or for a day, not a way of judging past performances. A team or a player should not be discredited as “lucky” for playing over his head, or winning with a poor pythagorian record. The outcome didn’t fail the stats; It is the stats that failed to predict the outcome.

I’m not in the Bruce Jenkins crazed anti-stat camp by any means, but do think its important to remember that baseball, as in life, does not always follow along with the most statistically likely scenario. This might be due to “intangibles”, but I think that is a cop out. Nothing is intangible, just immeasurable using current methods. There is a lot going on on and off the field that current statistics don’t take into account, just like stats can give new perspective that scouts of the past may have missed. The RBI nor pitcher wins was not handed down to Moses by God, but neither is OPS+.

When advanced stats are seen as the bottom line of success or failure in baseball, and not as a tool to improve future team performance, sabermetrics loses its scientific basis, and suddenly stops being a new school ideology. It becomes a kind of religious zealotry.

Posted 11/03  at  01:35 PM
Cooldrive said...

Stalton, doesn’t it seem to you that some stat-heads actually get upset when the World Series winner may not measure up statiscally?

Posted 11/04  at  10:07 AM
Bob Rittner said...

I don’t think the question Chris Lund asks is what is wrong with a stats approach (or with a scouting approach). It seems to me he is asking for suggestions as to how best to weld the two approaches so that we can have profitable discussion rather than rants.

I suppose proper modesty about the usefulness and dangers of progressive statistical analysis is one positive suggestion, but so too would be that those dependent on scouting or “old school” analysis also accept the limitations of their approach and welcome the use of stats to confirm or challenge their conclusions.

In fact, isn’t that the point. Rather than generalized denigrating or pointing to the fundamental flaws of each other’s perspective, the more productive approach would be to apply critiques to specific questions and not generalize from those specifics to wider ranging criticisms of methodology or viewpoints. Sort of like in a marriage, it is useful to express dissatisfaction or anger even because of a specific incident but fatal to generalize from that to demean the others character-you know, “why do you always….?” or “you never….”

I always like Lorelei Lee’s comment to her prospective father-in-law in “Gentlemen Prefer Blondes”:

“Don’t you know that a man being rich is like a girl being pretty? You wouldn’t marry a girl just because she’s pretty, but my goodness, doesn’t it help?”

So too with baseball. You wouldn’t make a judgment based purely on stats (or scouting), but my goodness, doesn’t it help?”

Posted 11/04  at  01:09 PM
RA Rowe said...

trying to get stupid people to listen to smart people is a steep uphill battle. and just like in the rest of life, there are a lot more stupid people in charge of things, and so they decide a bunch of stupid stuff should happen. Miguel Cabrera will be the MVP.

If you want to see a stupid person’s head start smoking, do this: Ask them “So if Miguel Cabrera and Mike Trout had the exact same numbers at the end of the season, but Mark Reynolds had ended up hitting .203 but hit 50 homers, would Cabrera still be your MVP.”

Posted 11/05  at  01:51 AM
Marc Schneider said...

It’s arrogance and smugness like this that guarantees no one will listen to you.

Posted 11/05  at  10:16 AM
hopbitters said...

To be fair, proponents of both schools cherrypick their stats when making the case for their favorite players. I know I do.

Posted 11/05  at  11:07 AM
Page 1 of 1
Commenting is not available in this weblog entry.



     Next Article:  Worst of the worst, Part II>> <<Previous Article:  BOB: World Series ratings