THE BOOK--Playing The Percentages In Baseball

Perhaps Rios does have attitude problems that need adjusted every so often...however you can’t prove anything with a 67 pa sample.

I suppose that’s what mgl is railing about.

When I first glanced at it, I thought it was a parody, making fun of the Blue Jays hasty decision to release Frank Thomas.  When I looked back, and saw that it was serious…

To be honest, I don’t see THT as a sabermetrics site.  It’s better analysis than any of the analysts on the Yahoo fantasy baseball site, and not nearly as complex (and anal) as this site… and that’s why it’s my #1 recommendation to baseball fans I know that might be interested in getting into sabermetrics.

I don’t think I’ve ever read more than a couple Brittain pieces, because it’s pretty clear to me he’s more of a diehard fan/writer than an analyst (and this is NOT an insult, it’s just my take on the few pieces I’ve read)… and I just don’t care that much about the Jays

Yes, John is not an analyst.  But we love him anyway.

A sabermetrician I most assuredly am not.

I do realize lineup construction doesn’t have a major impact over the course of the season--but it’s not about that.

It’s not about sample size either.

It’s about a guy who has always needed the occasional kick-start to keep him focused. 

It’s about a guy whose swing has gone south. It’s about a guy who has been very tentative at the plate. It’s about a guy who is letting pitchers dictate at bats by watching very hittable pitches go by in attempts to work the count rather than looking for a pitch to turn on.

It’s about a guy who gets into hitter’s counts and cannot square around on his swing and pulls off the ball.

It’s about a talented guy that keeps making the same mistakes over and over again. Sabermetrics cannot tell you how to take a guy with a million dollar arm and a ten cent head and make him a Cy Young calibre pitcher.

To convert talent to performance in baseball is the same as in any walk of life--they have to be taught how to use that talent optimally. The education process involves both motivating the student and the use of discipline (from the same root word as disciple: i.e. A disciple of Ted William’s Science of Hitting).

Right now, Alex Rios is flawed and underperforming--If he continues in the same pattern his numbers will not return to the mean. It’s not a sample size issue Everyone here knows the power of sample size. The larger the sample size the more reliable your statistics. The problem with any sample is that it is based upon the assumption that for any sample that is drawn from events across time, the thing that is being measured (in this case ability) remained unchanged over the period being considered.

Often, this assumption turns out to be valid wherein lies its value. The difficulty is that there is no way of knowing, on the basis of numbers alone, whether this assumption is valid. Put another way, statistics are not self-grounding ergo we would want to ground those statistics in meaningful first hand observations.

There are many instances where know beyond the a shadow of a doubt that a player’s ability/skill set has changed. For example:

* when a player is young and developing their skills
* when a player is old and losing skills
* when a player is injured
* when a player’s mechanics change either deliberately or due to injury or body change (ie: bulking up, slimming down)

In all of these cases the assumption that underlies statistical analysis is illogical. An alteration of circumstances has occurred and disrupted the sample size that gives statistics its authority. Observation tells us that Rios’ swing of 2008 differs from 2007 and 2006 therefore to assume that he’ll eventually revert to the same results of 2007 and 2006 is unreasonable.

Past performance is not a reliable indicator of Rios’ future production because of a different swing and plate approach than he had in the past. To get it back to where he is productive he needs discipline (education) and motivation to implement the necessary changes. Which is the reason behind the article.

Granted, not sabermetrically sound to be sure however we can agree that winning baseball and sabermetric orthodoxy are not synonymous--there are correlations but they are not absolute due to human variation--such as the variation in a hitter’s swing.

Best Regards

John

[[Yes, John is not an analyst.  But we love him anyway.]]

I’ve been told I’m half way there!!

What’s an yst anyway?

Best Regards

John

I have always enjoyed John’s posts at BBTF, but for some reasons his THT articles have never entertained me (and I am a Canadian who follows the Jays a great deal.)

I stopped reading this piece after the first two sentences.  Teams should ALWAYS be looking to improve, regardless of if they’ve lost two in a row or won 50 in a row.  I can’t even begin to describe how stupid I think the “if it aint broke dont fix it” mentality is.

I can’t even begin to describe how stupid I think the “if it ain’t broke don’t fix it” mentality is.

That is 100% correct (that is a stupid and counterproductive mentality, or strategy, if you will).  If you are doing something wrong and still winning or it seems to be “working” and you continue to do that “something” wrong, then obviously you are reducing your chances of winning from that point forward. That is why they call it “wrong.” Because it is “wrong.”

Success comes from being process and not results oriented.

It is funny and ironic that if we go back in history in look at any point in time at everyone who was having the most horrendous spates of performance you can imagine, and then we look at any time subsequent to that, we just can’t seem to find players who continue that bad performance (not individual players, of course, but “classes” of players).  Yet, somehow, in the present, everyone is “pressing” and their “mechanics are out of whack,” etc.  Somehow everyone is an exception, yet historically we just can’t seem to find that.  Funny how that works.  I just don’t understand it.

I don’t care whether THT puts itself out to be a sabermetric site or not (and I think it does).  Either you speak the truth or you don’t.  Obviously there are gray areas in between.  If you don’t (speak the truth), then prepare to be criticized.

Who above said that I “ranted” about this article?  All I said was, “There is so much wrong with it, that I don’t know where to begin.” That is a rant?

OK, I’ll rant.

A team shouldn’t make changes when the team is winning.

As Ryan said, above, that is really stupid.  Obviously, nothing “causes” winning or losing in the short run.  Or I should say that there are hundreds of things that “cause” winning and losing in the short run, not the last of which is plain old luck (and that “luck” could be the result of thousands of things that are in “someone’s” control - for example, even though we call the result of a dice throw “random,” obviously it is the result of a person shaking and throwing the dice in a certain way, which theoretically could be controlled, modeled, or predicted).

If you are doing something wrong, you should fix it whether you are winning or losing.

...it gives the Jays the opportunity they need to make a change—to light a fire under Alex Rios and drop him down in the batting order. His swing is off and he’s hurting the ball club’s ability to score runs.

Actually, lineup order is so inconsequential, plus a manager would not know an optimal lineup if it stood up and hit him in his face, that I don’t care what a manager does with his lineup.  If he wants to “send a message to a player or two” that is fine by me.

As I said, yeah, almost everyone who is doing poorly (and actually he is NOT doing all that poorly in lwts so far this year, certainly WELL within the range of typical fluctuation after 100 some odd PA’s) has a swing that is “off.” Yet, somehow we just can’t seem to find these players when we look for them “after the fact.” They must just disappear in a puff of smoke.

I had the epiphany last Sunday. It began in the top of the second inning. Kevin Mench opened the inning with a walk and Joe Inglett reached on an infield single. David Eckstein dropped a bunt on the left side advancing the runners and bringing up Marcos Scutaro.

Rios was up—as I detailed after the game I knew what was coming and was pleading to whatever forces might be listening not to trust the at-bat to Rios. As you’ll see in a moment, I had ample reason to foresee what was about to happen. For example:

I don’t know what the true odds are with Rio up in a DP situation, but anytime John is “certain” that Rios is going to hit into a DP, I’ll give him 2-1 odds on his mortgage for a year.

Over his last 67 plate appearances, Rios has hit 29 ground balls, 22 of which were turned into outs and five of those into two outs. He has had 47 base runners and left 41 on base. He has had 23 runners in scoring position and cashed in six while hitting into five double plays. His last 67 plate appearances have resulted in 55 outs.

I’m sure that his performance has been brutal so far, otherwise we wouldn’t be reading this article would we?  Just like the 1,623 players I looked at in the chapter in The Book on hot and cold streaks had brutal performance over 50 or 100 PA also, but somehow, just somehow - it must be some kind of magic - they all (collectively of course) managed
to hit at exactly their recent career averages over the next 5, 10, 20, or 100 PA. How is that possible if even 20% of those players with awful performances were “not swinging the bat well?” It isn’t.  Oh, I know, somehow I was able to choose the cold streaks at exactly the end of the streak.  I must be a genius or have a crystal ball (or I cheated, since everything already occurred when I looked at it).

Ideally, your No. 3 hitter should be the team’s best batter.

No, your team’s best batter should be your #2 or your #4 hitter, but it depends a little bit on the lineup.

Over Rios’ last 67 PA…

In both cases we see that Rios is ill-suited for the premier spot in the lineup.

We don’t make lineup decisions based on 1/3 of a season’s worth of data or 67 PA of data because we don’t project player performance based on that kind of samples of data, because if we did, we would be creating BAD projections which would mean that our lineups would be bad (if lineups mattered all that much which they don’t).

Even more worrisome is that his home run power tails off in the second half.

I think he means that it HAS tailed off in the second half in some past seasons.  The idea that that has ANY predictive value (that we expect his HR power to tail off this or in any subsequent season) is ridiculous, unless we are talking about regression toward the mean, which would require the first half to be above average for his type of player.

In 2006, he hit 15 home runs in the first half and just two in the second half. Rios was coming off a nasty staph infection however in 2007: first half, 17 HR; second half, 7 HR. Thus far in 2008 he has three home runs; he had 10 at this point in 2006 (.305/.355/.547) and 13 last year (.358/.396/.622). He’s batting just .262/.320/.369. Adding to the misery is that he is well below league average in driving home available base runners. Even worse, just one game into June he’s topped his 2006 total of hitting into double plays and matched 2007’s effort and is just five away from setting a career high and the season isn’t yet 60 games old.

Blah, blah, blah.  Yeah, we get that Rios has not hit all that well this year, especially in situational hitting.

No matter how you look at it, this isn’t a three-hole hitter at the moment.

Well, if I could go back in time, I might not bat him in the top 5 in the lineup.  But I can’t (go back in time).

As I stated (above) in previous years, he is the kind of guy you have to build a fire under.

How in the world does he know that?  Does he know this guy personally?

Move him down in the batting order, maybe call up Adam Lind and sit Rios from time to time. Get him motivated to fix his swing, make him earn back the elite spot in the batting order. As distasteful as it sounds, it could be his new contract has brought on a degree of complacency and it’s time for a little tough love on John Gibbons and J.P. Ricciardi’s part.

Whatever.

It’s time to play Rios according to his present production and not his past seasons.

Why?  If that were the case, then we should project players according to their “present” production (whatever “present” means) and ignore their past seasons, right?  Because how we “play” them should be EXACTLY according to how we project them to perform. Does John have a better projection system than Marcel, Pectota, Zips, or Chone?  If he does, I would like to see it.

It’s time for the Blue Jays and Alex Rios to put some serious work into rectifying his swing. It’s time to stop trusting the offense to a man not producing.

It’s time to demote Alex Rios to the bottom of the lineup.

O.K.

Now that may be a “rant”!

MGL, I don’t disagree with anything in your rant, except that I also don’t think of THT as a sabermetric site.  (and my opinion counts!).

We call ourselves a baseball site with insight.  Lots of times that entails sabermetrics, but not always.  I actually work hard to make sure we don’t get too wonkish, which is why I appreciate writers like John and Lisa Gray.

I love Mike’s comment, because that’s exactly how we think of ourselves: as an insightful baseball site that works hard to be accessible to the common baseball fan.  Whether we succeed is certainly arguable, but that’s our intent.

Also, we don’t edit our contributor’s work for analytic thinking; we only edit for clear writing—not always a strength of John’s either. We find writers we like and trust, and give them space to express themselves.

Of course, you should criticize and rant when you disagree with a THT article.  I’d expect nothing less.  And, as I said, I don’t disagree with your points.  But if we only posted articles on THT that I (and our editors) agreed with, it would be a very boring site.

I think THT is a SABR oriented site, with sabermetrics being a part of it. The historical articles, and ‘what if’ articles certainly seem to indicate that it’s that way.

Brattain merits the comment that was made here about Olney, “He knows just enough statistics to be dangerous.”

It’s pretty frustrating to read these articles, because he defends his conclusions to the death (or has in the past) even in the face of overwhelming evidence to the contrary.

I always wonder how someone like Brattain would fare were he to work as a GM for a decade.

MGL - actually, it was RAIL, not RANT
rail against or at to complain bitterly or loudly about [Old French railler to mock]

I do think players can get out of or into whack (alter their hitting mechanics) but these shouldn’t last very long as they are professionals who should know these things and have professional coaches to help them. I’m not a pro, but after ripping a LD double, for the next couple games I’m pulling my head and looping my bat, popping everything up.

I had no info to say if John’s conclusion on Rios was correct, even after reading the article. “It’s not about sample size either” but then he goes on spouting stats to prove why Rios is not the player he used to be.

This article doesn’t even pass the “common sense” test.  Don’t make changes when things are going well?  Really? 

THT should aim higher.

I think the main point to take is that John is correct when he says we do not care about the past.  However, John is incorrect when he says we care about the present.  If we’re making personnel decisions, it should not be made based on the past OR the present:  It should be based on the future!  Our decisions need to be based on what we think will happen, or in other words, “projections.”

If we had a reason to believe that Rios was going to continue hitting the way he has over the last X PA’s, then OF COURSE we should bat him lower.  However, as far as I know, there is no real reason to believe that Rios’s production over the next X PA’s is going to be the same as the past X PA’s so it doesn’t make sense to use him as if they will be. 

As far as I know, the best expectation for Rios’s performance over the next X PA’s is something resembling his career stats, and if that’s the case then he should be batting in the upper half of the lineup.

Let’s say there’s a manager who bases his decision for tomorrow on the results of today.

His star player bats 3rd and goes 0/4.
He moves him to 6th, and he goes 3/4.
He moves him back to third and goes 0/4.
He moves him back to 6th and he goes 3/4.

If this continued for 2 weeks, I have a feeling that Brattain would draw the following conclusion:
The batter should always be hitting 6th in the lineup.

After all, in 10 days he’s hitting 15/40 (.375) overall, but 0/20 in the 3rd spot, and 15/20 in the 6th spot!

I don’t have a great projection for Rios anyway.  It is below that of an average corner outfielder, so it is not like he should be batting in the best slot in any batting order, unless it is for a team like the Giants, Twins or KC.

There are certain batters that are frustrating, especially if you watch them.  The ones who swing at bad pitches. The ones who strike out a lot.  The ones who don’t seem to be great with situational hitting.  It seems to be that Rios was always one of those batters who swings at bad pitches, but when he does make good contact he hits the ball hard.  There are lots of players who overall are above-average hitters, not great, but are very strong.  They typically look horrible at times.  Francouer is one of those.  Wily Mo is another one, although he is not even an average hitter overall.  There are tons of them actually.  Big, strong hitters who if they actually had a good eye and good swings mechanics, they would be ridiculously good hitters overall.  But because they are one-dimensional, i.e., they swing at a lot of bad pitches, and even take some pitches that are good, they tend to look bad quite often.  I can’t stand when I am rooting for the Braves and Francouer gets up to the plate.  Yet I know he is a decent hitter overall.  I also know that he will swing at just about anything and is easy to strike out if you happen to throw him good quality pitches out of the strike zone (although he is not nearly as bad as he used to be at chasing bad pitches).

The there are decent hitters who have decent eyes and good swings, don’t strike out a lot, but just don’t have a lot of power.  And there are decent hitters who have terrible eyes, strike out a lot, but when they do hit the ball, they hit it hard and far.  Rios is one of those hitters, and he is easy not to like.  If you have a good eye and hit the ball hard when you swing at good pitches, you tend to be a great hitter.  There are not that many of those.

That is why I don’t like to make too many judgments about players from watching them play. Or at least I always operate from the assumption that my eyes and my mind will deceive me, or at least will bias me.

And now I feel bad for John and THT.  I like both John and his articles and of course I love the THT site.  The two sites I read daily are BP and THT, and BP is becoming increasingly irrelevant, at least for me.

Rios’s “slump” is almost entirely explained by his lack of patience/discipline/eye/whatever since a flukey start to the year (in that regard.)

Rios started the year by drawing eleven walks in the first eleven games (51 PA.) Since then he has drawn only eleven more in 47 games (211 PA.)

Just another example of random streaks not telling us anything (at the start of the year, many thought Rios had “learned” to be disciplined or gotten better eyesight or something.)

Having a bad “eye” is seen as a character flaw which is why Rios must be punished and Eckstein is seen as scrappy.

True, pitch recognition is something that improves with experience and some may not work as hard at it as others, but I have no doubt that there’s an aptitude there as well. Soriano will never have an eye like Jack Cust. Likewise, Cust will never be able to hit a ball off his shoelaces for a homerun.

First off, don’t read too much into the 67 PA sample size--I just went through game logs in reverse order and when I felt I had enough games to establish that his play has been off for some time I stopped. I opened research with a simple assumption--I’ve watched Rios struggle at the plate for quite some time now. Was this merely an impression or did the numbers confirm what I was observing. After I backtracked through enough game logs I was satisfied that my observations were not a hunch or misperception. 

Also, we don’t edit our contributor’s work for analytic thinking; we only edit for clear writing—not always a strength of John’s either.

Hey, not my fault you don’t speak Canadian eh.

Brattain merits the comment that was made here about Olney, “He knows just enough statistics to be dangerous.”

And don’t you ever forget it!

I always wonder how someone like Brattain would fare were he to work as a GM for a decade.

I’d suck at it obviously. As would everybody else here. Is there anybody here who think they’re well-qualified to be a GM of a major league team?

Anyone? C’mon--don’t be shy.

I had no info to say if John’s conclusion on Rios was correct, even after reading the article. “It’s not about sample size either” but then he goes on spouting stats to prove why Rios is not the player he used to be.

You’re misreading me--I’m saying he has a problem with his swing that needs correcting ... not that’s he’s in decline.

This article doesn’t even pass the “common sense” test.  Don’t make changes when things are going well?  Really?

That point was poor wording on my part, I should’ve stated not to make major changes when things are going well. Again, the human element has to be taken into consideration--many humans are superstitious, athletes have their various rituals that they strictly adhere to because they feel it makes them perform optimally. Granted, it’s a placebo effect but you don’t tell a player to sit on a different spot on the bench during a no-hitter because the numbers show it has no effect on what the pitcher does.

A team on a roll should be allowed to roll and major changes should be put off until the roll stops because you don’t want to disrupt the positive mindset of the team by throwing in a distraction.

It’s pretty frustrating to read these articles, because he defends his conclusions to the death (or has in the past) even in the face of overwhelming evidence to the contrary.

Not really, if you go back years you’ll see I’ve altered my stance on a lot of issues. However, I have yet to see “overwhelming evidence” that the human element within the game is close to irrelevant. Baseball is a game played by individuals with various levels of talent and motivations--however talent has to be motivated, utilized properly and harmoniously for success. 

Numbers don’t tell you how to do that.

Let’s say there’s a manager who bases his decision for tomorrow on the results of today.

After all, in 10 days he’s hitting 15/40 (.375) overall, but 0/20 in the 3rd spot, and 15/20 in the 6th spot!

Again--no. you’re drawing conclusions without trying to understand where I was coming from. Slumps happen for a variety of reasons--in Rios’ case it is a clearly identifiable problem … a bad swing. Do you accept the status quo or do you take steps to nip it in the bud?

There’s some great feedback (and I really wish mgl had gone into more detail at the outset because I do value feedback and a simple “ There are so many problems with this THT article, from a sabermetric perspective…” isn‘t real useful to me--Post No. 8 should‘ve been Post 1)

My biggest issue is with this line “THT should aim higher.” This reads to me as “THT should have writers that agree with my POV.”

Here is where I get anal.

Obviously, I am not an analyst nor am I sabermetrician. I do have holes in my knowledge that I am seeking to fill. My father was a LHP that played in the Tigers minor leagues in the early 1960’s, my brother and I played a ton of baseball, I read Bill James’ abstracts in the 1980’s and have almost 40 years following the game and I do my homework.

My point?

I’ve seen the sport from a number of angles and I’ve learned that there is no single “Holy Grail” in baseball. Parks change, equipment change, rules change and whichever team could adapt best to the circumstances of the time were the ones that enjoy success. There is “no one size fits all” approach to the game. The “Hitless Wonder” White Sox, the Bronx Bombers of Ruth and Gehrig, the pitching and defense Dodgers of the 1960’s, the speedy Cardinals of the mid 1980’s, the Bash Brothers A’s of the late 1980’s-early 1990’s, the Maddux/Glavine/Smoltz Braves etc.

Anybody think the walk/three run bomb approach would’ve worked at old Huntington Avenue Grounds with the dead ball in 1909?

I do my homework and research, I talk to people and get feedback, I check, double-check, and triple check my work and I arrive at a conclusion--just because it doesn’t line up with your POV the subject don’t imply that I am “unworthy” to write about the sport or for THT. Your statement “THT should aim higher” implies that you’re the final authority on what is correct baseball thought and that it B.S. I’ve written some bad articles in the past and what I wrote 10 years ago may not line up with what I write now. However I would never presume to name myself the final authority on any subject within the game because I’m continually learning and adding to my database on *all* subjects and chances are, what I write in 10 years will differ from what I write at this point in time.

THT succeeds because we ‘don’t aim higher’ and mindlessly embrace a single orthodoxy. It succeeds because of its diversity of thought and opinion. THT uses the widest sample size possible in trying to understand the sport. If history has taught us one thing it is this--those who feel they are worthy of establish “what is right and true” are far more dangerous that some nitwit who knows just enough about stats to be dangerous.

It’s also the demarcation point where education stops and indoctrination begins.

To suggest that THT embrace a collective or hive mind of correct sabermetric thought retards--and not enhances learning about the sport. It is the same mentality that causes the mainstream media to shy away from sabermetrics--the MSM feels they know all about baseball and those that embrace the new-fangled stats should ’aim higher’ by getting their noses out of a spreadsheet and watching a game once in awhile.

If you hate mainstream (baseball) thought then don’t embrace their approach otherwise you just became the sabermetric version of Bill Conlin.

Best Regards

John

Is there anybody here who think they’re well-qualified to be a GM of a major league team?

As long as Rejean Houle was once a GM, that would mean that about a million people qualify to be GM of a major league team.  Who’s the MLB-equivalent?  Steve Phillips?

***

The best thing to come out of John’s article is that he relied on observational analysis to tell us that there was something fundamentally wrong with Alex Rios.  Something fundamental and basic, that would take say 5 games to notice, would require say 200-300 controlled PA for us to pickup on the data with the same amount of reliability.

If we could get all fans to flag a player for concern based on purely observational analysis (ignoring the actual resultant PA), we’d be far better off.  We can actually then study whether there is persistence in the concern.

I think we would have been better served if John ignored the numbers (which looks like they were used to support his position), and instead relied on more observational analysis, things we can’t easily see in the numbers.  Like, when he’s down 0-2, are pitchers throwing him curveballs away, and is he chasing them (aka Andres Galaragga syndrome).  And, is this something new or par for the Rios course?  Does he take pitches with guys on 1B more than usual?  What does he do on 3-0 and are pitchers throwing it to him down the middle or still on the edges?  Basically, what kind of decision-making is he involved in that would lead us to think that something fundamental has changed for those of us who don’t see him?

Everything about baseball is about the batter/pitcher matchup.  The HR, the BA, etc, those are just the results, but whose reliability is severely impacted by the sample size and the uneven context. 

Observational analysis is 50% of sabermetrics, and so, John is perfectly poised to fill a huge gap here.

I would also look forward from Pirates fans on Jason Bay, without seeing a single quote about his HR, SLG, or OBP.  That would be the ultimate sabermetrics article.

Rios went 2 for 3 with a walk last night. I guess he must have read the article.

I’m a Blue Jays fan, but I find Brattain’s stuff to be just a variation of the fanboy ranting you see on any team-oriented website.

I’ll just second Tangotiger/#19. And I’ll add that I did not enjoy that article because it focused so heavily on statistics that have very little or no predictive power. I would’ve enjoyed the article quite a bit more if the swing analysis amounted to more than just:

His swing is off and he’s hurting the ball club’s ability to score runs.

I would have also enjoyed the article more if you had showed us a photo/animated gif or two of when he was going well versus now. A good article (in this context), in my opinion, is simply a hypothesis, “Rios should be dropped in the order,” and some support for that hypothesis. You gave us the hypothesis, but not much in the way of support.

Again, the human element has to be taken into consideration--many humans are superstitious, athletes have their various rituals that they strictly adhere to because they feel it makes them perform optimally.

Sabermetricians don’t deny the “human element.” What they don’t like is when someone uses 67 plate appearances to draw conclusions about the human element.

Tangotiger linked to this a few weeks ago, and if you haven’t read it John, you really, really should. I think you’ll enjoy it.

http://ussmariner.com/2008/05/19/evaluating-chemistry/

I admit that my comment, “THT should aim higher” was a cheap shot, but it was motivated by frustration, not elitism.

If I want to read an article where the author presents his mostly unsupported opinion as fact, I can go anywhere.  ESPN, local newspaper, sports radio… whatever.  It’s not hard to find a person with a strong opinion that wants to share it.

I was originally turned on to THT because I felt that they had set the “evidence” bar for presenting an opinion a little higher.  Maybe not “master’s degree theses” high but the authors seemed to understand that, if they wanted to offer up a conclusion, they had better give their readers a good reason to go along with it.  Even then, as all good analysts do, they knew when to sprinkle in the phrases, “it appears”, “the numbers suggest” or “I suspect” to keep them honest.  I keep using past tense but there are many THT authors that are still churning out great stuff all the time.

And I’m not saying that every article needs to be a study.  I’ve been reading Aaron Gleeman for years and very rarely does my “sabermetric siren” go off while I’m reading one of his articles or posts.  My siren broke, reading the Rios article.  Maybe that kind of writing appeals to some but I suspect that it doesn’t appeal to the majority of the people who support THT by buying their excellent books.

I really wasn’t tying to slam John.

I did say he may very well be right on Rios, and I told MGL that I do believe players can get out of whack and underperform for periods of time.

It’s just that I didn’t think that John proved his point, and the explanations wandered around.

As Tango says above, if you think “something fundamental has changed” about Rios, focus on that and make us believe you just might be right.

Tom, as a Pirates fan who has watched Bay, I guess I’ll have to use stats, but I do think his performance was injury related. Bay had off-season knee surgery before 2007, and in the first two months his babip, hr rate, bb% & so% were all in line with his established level for each. He crashed around the beginning of June and stayed that way the rest of the year. His babip, hr/fb, xbh/h, gbh/gb, fielding rating were all down, while his bb% & so% were pretty stable. To me, he wasn’t running as well and wasn’t able to drive the ball when he made contact. This year, everything seems to be back where he was before.

And I’m not saying that every article needs to be a study.  I’ve been reading Aaron Gleeman for years and very rarely does my “sabermetric siren” go off while I’m reading one of his articles or posts.

I hate to break this to you but if your sabermetric siren doesn’t go off when you’re reading one of my columns then you’re not reading it right.

Most here view sabermetrics as the lens through which they examine the sport--I don’t. I appreciate sabermetrics I understand sabermetrics to an extent but I view it as a supplement to what I have learned in almost four decades as a fan and 10+ writing about it.

Here are where we are at odds--mathematics is the ultimate truth. 2+2=4 even if you’re an accountant for a major league team. Baseball is a game with almost endless variation as to what can happen in a given game--to me that leaves a lot of leakage at the edges.

If you’re deeply into sabermetrics I strongly urge you to avoid everything I write because I do not fully embrace the sabermetric model and I never will. I will use stats, I will use my own experience playing the game when I was a lot younger than I am now, I will use what my dad learned from guys who did play in the majors (he had Sal Maglie as a pitching coach at one point), I will use what I have gleaned in talking with players, fans, sabermetrician, managers, coaches--the works … all data counts.

And I will use that most unscientific of tools--intuition. I know what a slump looks like, I know what they feel like and sometimes I see something familiar in what is going on during a game. The Rios double play that ended the ninth inning against the Angels was easy to call--a tentative struggling hitter, a guy hitting the ball on the ground or softly most of the time he’s making contact of late. I saw Rios take a ninth inning bases loaded, two out strike three against Francisco Rodriguez earlier in May. The Angels led by two, Scott Rolen was on deck, and Rios had been struggling before then as well.

He was expecting a pitch away.

Why in God’s name would anybody expect K-Rod to pitch away to a struggling hitter when a walk makes it a one run game and brings up the Jays hottest hitter--the one guy who has been hitting well with RISP?

The numbers do not tell us--experience does. Rios froze. There is no logical, rational reason not to expect a fastball over the plate in that situation. It’s what happens when you’re off--he wasn’t expecting a pitch away, he was hoping for it to be away.

When I’m “analyzing” a situation I will not throw every other data point I have learned in almost 40 years and stick to the sabermetric orthodoxy when reaching conclusions--that’s bad analysis.

“I’m a Blue Jays fan, but I find Brattain’s stuff to be just a variation of the fanboy ranting you see on any team-oriented website.”

Then you haven’t been reading carefully enough. Jerry Howarth thinks enough of my fan boy ranting to have referenced it during his radio casts--I hate to name drop like that but I’m gonna defend the effort I put into my work. Just because I use a different methodology that you approve of doesn’t mean I’m not doing my homework. I do a lot of it and I draw from many sources and not just what is written at Baseball Prospectus.

So please, do us both a favour and save us both aggravation and simply read writers who espouse the same views as yourself. My experience with baseball differs considerably from yours so don’t expect my conclusions to line up with yours.

Best Regards

John

Rios just had a nice cut against Wang--thanks Melky .

Best Regards

John

I (heart) Matt Stairs.

:-D

Best Regards

John

So please, do us both a favour and save us both aggravation and simply read writers who espouse the same views as yourself. My experience with baseball differs considerably from yours so don’t expect my conclusions to line up with yours.

Fair enough, John. If you’re going to publish, not everyone is going to agree with you. And I like to read writers besides those I agree with. I am kinda curious, however, how you presume to know what my experience with baseball is.

Tango: “If we could get all fans to flag a player for concern based on purely observational analysis (ignoring the actual resultant PA), we’d be far better off.  We can actually then study whether there is persistence in the concern.”

There are two problems, however:

1) John doesn’t say anything particularly interesting about Rios’s swing or approach.  It doesn’t help anyone to hear that his swing is “off” or “broken.” What does that mean?  If you don’t want people to call BS you have to be more specific than that.

2) As always, it’s after the fact.  What is that old adage about usually finding something when you look hard enough?  It’s easy to say “Rios’s stats are bad, let’s look at his swing, well look at that...” Of course if you’re looking for something faulty (based on his stats) you’ll probably find it.  I would love it if anyone, if ANY of the swing mechanics or pitching mechanics “gurus” could identifiy these things regardless of the numbers. 

Here’s an experiment I’ve always wanted to try: take 50 animated GIFs of batters, 25 of whom are “slumping” and 25 of whom are “streaking.” Then show all 50 of them to a hitting coach or a hitting guru or expert or what have you, and ask him if he can identify, based only on the swing, without ANY statistics, which 25 are streaking and which are slumping. 

If anyone could do that, then I would be greatly impressed, and would take him far, far more seriously when he tells me someone is slumping because his swing is “off.”

It is also possible that swing mechanics fluctuate randomly from day to day such that a player who did have a slump did have bad swing mechanics but that still there is little to no predictive value to those bad mechanics.  Which leaves us in the same place is if there were no bad mechanics which contributed to or caused the slump.  Plus, one of the ways in which hitter performance randomly fluctuates from PA to PA is the types of pitches that a batter happens to get and swing at.  Let’s say that a certain batter is somewhat of a guess hitter.  And let’s say that he happens to have a spate of PA where he guesses wrong a lot.  It will probably “look” like his swing is off simply because he is guessing wrong a lot.  Ditto if he happens to have gotten a lot of really tough pitches.  Ditto on the opposite side if he happened to have gotten a lot of cookies, or guessed right on a lot of pitches. He would look like he was really “locked in.”

What I’d like to see is if scouts/fans or whomever can look at video of a player slumping or streaking, and from the mechanics tell us whether that streak or slump is going to continue for some time.  People like John ought to be able to do that since they are the ones calling for benching or moving a player a player in a lineup, which means that they believe that they can project performance well over and above a regular projection algorithm, based on short-term recent performance.  We know that we can’t do that (project performance over and above a regular projection algorithm) from looking at stats only (to any significant degree), so the only thing left is to be able to do that from watching a player.

"I know what a slump looks like, I know what they feel like and sometimes I see something familiar in what is going on during a game.”

And I require evidence that you possess such an ability before I take it seriously.  I don’t think that’s unreasonable.