PDA

View Full Version : Interesting post from Sons of Sam Horn...



Wheelhouse
04-22-2006, 01:12 AM
Guess they have the same conflicts we have...

Sorry statheads, "Clutch" does indeed exist
Sille Skrub
Apr 13 2006, 08:22 AM
Post #1


NSFW


Posts: 3,095
From: America's greatest city




Hey statheads, I’m looking at you!

I can appreciate the statistical look at the game off baseball. I’ll be the first to admit I have learned a heck of a lot from you saber guys. To me, sabermetrics is a fascinating new (to me) way to look at the game. Before I started reading SoSH, I looked at the game in a totally different manner, the manner in which most fans do. But along with some of your posts, the books “Moneyball” and “Mind Game” and a few ganders over at B-P, I have come to grasp an elementary understanding of the saber point of view. To be honest, I still don’t fully understand terms like VORP, EqA, ERA+, AEqRA and WARP. One thing I do like Sabermetrics is how it especially valuable in predicting how younger players will project in the majors.

With all that being said, it is always downright comical to me when I hear some stathead cry, “There’s no such thing as clutch!” IMHO, statheads fear (yes fear) the term “clutch.” The reasoning behind it is that there is no way for them to put their arms around the term. To quote Chris Farley in “Tommy Boy,” you can’t touch it, feel it, pet it or even massage it. There is no formula, no postulate, no book, no website where they can go to effectively quantify how “clutch” plays into the game of baseball or sports in general for that matter.

There is a human element to sports that simply can’t be quantified by any numbers. Above all this is a game; a game played by human beings who put their pants on one leg at a time. They aren’t computers or robots. They have good days and bad. Go through family problems, weddings, divorces, births, deaths and sickness. Clutch exists. Think: David Ortiz over the past three years, Nomar in the 1999 playoffs, the Basketball Jesus (thanks sptguy33!), Tedy Bruschi grabbing an interception when you think all is lost. Now, if clutch exists then anti-clutch has to exist. Think: A-Rod when the game is on the line, Clemens against Dave Stewart, Skrub when it comes to women, etc.

Clutch comes and goes. Clutch players aren’t always going to succeed in a clutch situation. Anti-clutch players will also sometimes succeed when the game is on the line because luck will sometime play a factor. All I’m saying is there is a factor in sports called “clutch” where certain players will exceed more often then not when the game is on the line. I will concede it is vague statement; but again, there will never be a valid away to quantify it.

I really don’t know why I typed this all out. Heck, I’ll probably get flamed for it on this board. I just wanted to throw out an anti-stathead point-of-view. Stats are good, but they don’t tell the whole story. IMHO, stats are only good at looking at the offensive side of the game. Call me crazy, but all of the defensive metrics I have read about seem flawed.

So, statheads, sorry to break it to you:

Clutch exists. Deal with it.

Fire away.

Caveat Emperor
04-22-2006, 01:32 AM
He could've just summarized by saying: "Clutch exists because I say so. Take that you evil-doing statheads!" That would've saved me the valuable 5 minutes I spent reading this posting.

If you want to prove "Clutch," come with something better than a few ESPN highlights of late inning heroics. Give me something to wrap my fingers around as to how it impacts the game or should make a discernable impact on how I (and others) look and evaluate teams and players.

There's nothing of the sort here. It's just opinionated drivel stated with conviction but no real backing at all.

harangatang
04-22-2006, 01:38 AM
There needs to be a balance between both a general knowledge of the game and advanced statistics. Too much of either and not enough of the other can cause problems and frequently does.

FWIW, the same problem exists in the weather field too. I've seen more than a couple of guys read all the weather models number for number and 95% they're wrong. Now if they couple the general trends from the weather models and throw in some theoretical meteorology their forecasts all of sudden becomes much better. There is nothing perfect to forecasts and everybody is wrong sometime no matter what. Baseball is the same way, there is no perfect way to predict a player's performance.

ochre
04-22-2006, 01:41 AM
Seriously, if you think that is compelling in any evidentiary way then you win. That's nothing more than a pointless rant with no substance. The fact that it occured on SoSH does not give it any weight beyond what the statements made possess of themselves. So clutch is an instance. It comes and goes as it pleases and is totally unmeasurable. Whew. I sure am glad that's been settled.

savafan
04-22-2006, 01:50 AM
This post comes from the forum that banned me simply because I don't live in the Boston area.

I will not even dignify it by reading it.

westofyou
04-22-2006, 10:11 AM
That's nothing more than a pointless rant with no substance.

Guess they have the same conflicts we have...

IslandRed
04-22-2006, 10:24 AM
Clutch plays exist. But do clutch players?


certain players will exceed more often then not when the game is on the line. I will concede it is vague statement; but again, there will never be a valid away to quantify it

That's the core contradiction of the guy's argument. It's not that hard to figure out "when the game is on the line" and then quantify how the player did in those situations and compare it to his normal level of performance and what other players do in those situations. If clutch is a true ability and they really do "succeed more often than not," such an analysis would show it. But are the likes of Ortiz and Jeter raising their games in those situations or are they just the same great hitters they always are? Guys who can't handle pressure don't get to the top of the mountain.

And yet...

The number of true game-on-the-line situations, as opposed to more generic clutch situations that we have shorthands for like RISP, close-and-late, etc., is sufficiently small that the sample-size argument can be invoked against drawing a conclusion one way or the other. I think that's what Bill James argued in a followup to his earlier work -- he said we may never be able to prove clutch, but we can't necessarily disprove it either.

Reds Nd2
04-22-2006, 10:28 AM
So, statheads, sorry to break it to you:

Clutch exists. Deal with it.

Fire away.

Leprechauns (http://www.youtube.com/watch?v=nda_OSWeyn8&search=leprechaun) exist too. Deal with it.

redsfanmia
04-22-2006, 09:37 PM
Someone please explain to me how clutch doesnt exist. I am not a stathead guy, never will be a stathead guy. I have watched and played tons of baseball and have seen guys who consistantly performed in big game breaking parts of the game and guys who consistantly dont perform in those same situations. This is not an always or never sitiuation but more often than not situation. It has more to do with the mental make-up of a person more so than physical attributes. This is my opinion now throw stats at me to prove me wrong.

cincinnati chili
04-22-2006, 09:47 PM
Someone please explain to me how clutch doesnt exist.



One of the big reasons I think "clutch" is overrated, or perhaps doesn't even exist at the major league level, is because most baseball players who lack the ability to perform in "clutch" situations are weeded out long before they get to the major leagues.

Sure, you see a Rick Ankiel every so often, where some latent issue comes to the forefront. But for the most part, guys who are adveresely affected by adreneline, butterflies, or off-the-field issues are spit out in high school, college, or the low minors.

I used to believe that clutch mattered at the major league level (or at least the "non-choke" factor - opposite-of-clutch factor mattered), but the statheads have pretty much brainwashed me the other way.

MWM
04-22-2006, 10:10 PM
I am a firm believer in clutch in every aspect of every sport, including baseball, with the exception of hitting. It's the only thing I believe clutch might not exist. I'm not quite convinced completely yet as I tend to believe in clutch hiting bein the absence of choke. And my belief isn't really based on stats, it's more based on the nature of hitting. Of all athletic actions, it's the one that's more instinctive than anything else. There's no time to think about what you're doing, it's all reaction and reflex. A batter must make so several decisions and act with a great level of precision within a fraction of a second. Plus, it's an action that even the best of the best fail at more than they're sucessful. I just have a hard time believing people can change all the factors that go into hitting in certain situations. They still don't know what pitch is coming or where it's going to be, and they still have to react in less than a half of a second. You've either got it or you don't. Now I do believe that players come up to the plate in certain situations and fill their heads full of all kinds of nonsense based on that situation. Those players are likely to fail more often when they're thinking too much about the situation. But as chili mentioned, most of those guys are not going to have what it takes to make it to the pros. heck, if you'e trying to make it to the majors, all ABs could be considered clutch.

I believe in clutch pitchers more than hitters. But it's a completely different action altogether. It's all premeditated.

remdog
04-22-2006, 10:29 PM
How do I give this guy 'rep points'?

Rem

forfreelin04
04-22-2006, 10:39 PM
I think both schools can learn something from each other. Both the old school and the sabres Billy Beane Types. However, the problems arises that stats and such can be learned by a former player. But a stathead cannot go back and be a Major Leaguer. There are certain things that "fans" will never understand because they never played competive baseball. All these stats really go from a logical standpoint is predict the liklihood a player or team will do something at a given time. It uses what already took place and then applies it to probably will occur in the future.

Zimmazamma
04-22-2006, 10:43 PM
"Clutch comes and goes. Clutch players aren’t always going to succeed in a clutch situation. Anti-clutch players will also sometimes succeed when the game is on the line because luck will sometime play a factor."

Which side was this guy arguing for??? He makes the point for the anti-clutch crowd. If clutch comes and goes, then it's simply chance or luck. He basically admits it. Clutch doesn't exist.

remdog
04-22-2006, 10:51 PM
".....If clutch comes and goes, then it's simply chance or luck. He basically admits it. Clutch doesn't exist.

You've come and gone for the last year. How do I know that you exist? :)

Rem

RFS62
04-22-2006, 11:07 PM
I am a firm believer in clutch in every aspect of every sport, including baseball, with the exception of hitting. It's the only thing I believe clutch might not exist. I'm not quite convinced completely yet as I tend to believe in clutch hiting bein the absence of choke. And my belief isn't really based on stats, it's more based on the nature of hitting. Of all athletic actions, it's the one that's more instinctive than anything else. There's no time to think about what you're doing, it's all reaction and reflex. A batter must make so several decisions and act with a great level of precision within a fraction of a second. Plus, it's an action that even the best of the best fail at more than they're sucessful. I just have a hard time believing people can change all the factors that go into hitting in certain situations. They still don't know what pitch is coming or where it's going to be, and they still have to react in less than a half of a second. You've either got it or you don't. Now I do believe that players come up to the plate in certain situations and fill their heads full of all kinds of nonsense based on that situation. Those players are likely to fail more often when they're thinking too much about the situation. But as chili mentioned, most of those guys are not going to have what it takes to make it to the pros. heck, if you'e trying to make it to the majors, all ABs could be considered clutch.

I believe in clutch pitchers more than hitters. But it's a completely different action altogether. It's all premeditated.



Nice post.

And well said, that "clutch" is too broad a term to use across the board for every motor skill involved in baseball.

Consider that the pitcher is the only player on the field who actually initiates the action. The play doesn't start until he does. Everyone else is in reactive mode. He's the initiator.

The mindset behind pitching is very similar to golf. You imagine the action you're about to initiate, and you start the play. If you're prone to mental distractions or negative thinking/visualization, you're in big trouble.

Hitting and fielding is reactive in nature. Similar to return of serve in tennis. You are waiting for someone else to start the play, and reacting rather than initiating. Good fielders go over their responsibilities before the play starts, then shift to visual thinking as the play starts.

Hitters do the same. Good hitters, to varying degrees, think about what the pitcher might throw, then shift to visual thinking, totally reactive and non-verbal thinking. If they're engaged in internal dialogue when the pitch starts, they're doomed.

Choke is too broad a term to encompass all these motor skills and the thought process involved in each.

The ability to quickly shift between linear thinking and internal dialogue analysis to non-linear awareness is a key skill that's not talked about much. In fact, many great athletes do this and could never engage in a discussion about it... they just do it and don't realize the dichotomy between the two.

top6
04-22-2006, 11:34 PM
One of the big reasons I think "clutch" is overrated, or perhaps doesn't even exist at the major league level, is because most baseball players who lack the ability to perform in "clutch" situations are weeded out long before they get to the major leagues.Exactly. These guys are used to performing under pressure. We assume that the biggest "clutch" at bat for these guys is when they're up in a big game in the bottom of the 9th with 2 outs. But by that point they're already making hundreds of thousands or millions of dollars playing baseball. The at bats where there was real pressure were the 50 or so ABs they had in spring training when they were on the bubble to make the team, the 4-8 ABs in high school the day a big league scout came to town, the ABs in low A ball where all they wanted to do was get called up, etc., etc.

IslandRed
04-22-2006, 11:53 PM
Exactly. These guys are used to performing under pressure. We assume that the biggest "clutch" at bat for these guys is when they're up in a big game in the bottom of the 9th with 2 outs. But by that point they're already making hundreds of thousands or millions of dollars playing baseball. The at bats where there was real pressure were the 50 or so ABs they had in spring training when they were on the bubble to make the team, the 4-8 ABs in high school the day a big league scout came to town, the ABs in low A ball where all they wanted to do was get called up, etc., etc.

There's an oft-paraphrased quote from Lee Trevino about that, something about there's nothing hard about putting for $100,000 when you already have that much in the bank. Pressure is playing a $20 Nassau with $5 in your pocket.

RedsManRick
04-23-2006, 01:01 AM
I hate this argument so much. Clutch exists in so far as some players have performed well in "clutch" situations. However, clutch does not exist such the certain people predicitably perform better in clutch situations over time.

Yes, some people, even of the course of an entire career perform better in "clutch" situations than normally. However, this does not happen enough or to any particular subset of people such that it can't be explained by random variation. Given 1000's of players, we would expect numerous players to be "clutch" over their entire careers.

The real point is -- using the term to describe a PERSON rather than a PERFORMANCE implies some level of predictabilty. Namely "given situation X, player A is more likely to exceed his standard level of performance than player B is". If it is meant to imply that -- it's wrong. If it isn't meant to imply that, then it's a empty, vacuous claim which serves only to make the announcer seem as if he/she has particular insight which they really don't have.

David Ortiz was clutch in 2004! -- Absolutely
David Ortiz has been clutch -- Possibly, I haven't seen the numbers
David Ortiz is clutch -- Prove it.

BCubb2003
04-23-2006, 01:12 AM
Some good arguments here, although clutch might be the ability to maintain under stress that skill to shift from thinking to acting, while another player is taking too much time thinking about how the game is on the line when he should be shifting into action mode.

As far as filtering out the non-clutch player, that may be, although the majors haven't filtered out the .220 hitter, the brain cramps, the check swings and the slumps.

But there simply aren't any numbers. And it seems pretty clear that nobody gets noticeably better in clutch situations. Do some good hitters get noticeably worse? That's the challenge in proving that clutch exists.

RFS62
04-23-2006, 08:48 AM
But there simply aren't any numbers. And it seems pretty clear that nobody gets noticeably better in clutch situations. Do some good hitters get noticeably worse? That's the challenge in proving that clutch exists.


I'd have to agree that there aren't any numbers to prove that clutch exists.

From a players point of view, I believe it's a mental thing entirely. How one handles pressure, or the elevated pressure of an important game situation.

It's the difference between the attitude of "please don't hit it to me" vs. "I want it to be hit to me" on defense. And the players know. It's the psychological side of sports that doesn't show up in numbers. It's "makeup" that's often laughed at because it's not statistically measurable.

But I can tell you that on every team in every sport I've ever played in or coached, the players knew and the coaches knew who they wanted when the game was on the line. Admittedly, the higher level of play, the more likely the mentally weak player had been sifted out.

When Rod Carew filmed his landmark hitting session for Sybervision, he talked about the mental side of hitting. One of the interesting things he mentioned was the fact that stress has a measurable effect on eyesight. When you hear players talk about the "zone", I believe it's a mental state where you've eliminated your internal dialogue completely, just like the state of mind a distance runner can achieve, and you're in a state of heightened awareness in which "the ball slows down" and "the ball seemed much bigger", comments you often hear hitters make when they try to describe the experience.

Trying to think yourself into "the zone" can't be done. It's the exact opposite of giving yourself mental commands. It's the absence of such self talk. Some players have a natural ability to do this, others have a harder time letting go of the internal dialogue.

Imagine running down a mountain. If you try to pick every step, you stiffen up and you'll fall. But if you abandon yourself and let go, you somehow can run with effortless ease, and your subconscious mind picks the steps far better than your ego mind, your internal dialogue, ever could.

That's the zone. That's where you want to be when you play. Stress interferes with achieving that state of mind. The heightened stress of a "clutch" situation affects players in different ways, and the players who can maintain their composure under such stress are the players I would call "clutch". The evidence is all anecdotal, probably known only by the participants.

It's pretty clear after all the discussions we've had that it's not statistically measurable. But don't ever tell a player that it doesn't exist.

redsfanmia
04-23-2006, 08:59 AM
I'd have to agree that there aren't any numbers to prove that clutch exists.

From a players point of view, I believe it's a mental thing entirely. How one handles pressure, or the elevated pressure of an important game situation.

It's the difference between the attitude of "please don't hit it to me" vs. "I want it to be hit to me" on defense. And the players know. It's the psychological side of sports that doesn't show up in numbers. It's "makeup" that's often laughed at because it's not statistically measurable.

But I can tell you that on every team in every sport I've ever played in or coached, the players knew and the coaches knew who they wanted when the game was on the line. Admittedly, the higher level of play, the more likely the mentally weak player had been sifted out.

When Rod Carew filmed his landmark hitting session for Sybervision, he talked about the mental side of hitting. One of the interesting things he mentioned was the fact that stress has a measurable effect on eyesight. When you hear players talk about the "zone", I believe it's a mental state where you've eliminated your internal dialogue completely, just like the state of mind a distance runner can achieve, and you're in a state of heightened awareness in which "the ball slows down" and "the ball seemed much bigger", comments you often hear hitters make when they try to describe the experience.

Trying to think yourself into "the zone" can't be done. It's the exact opposite of giving yourself mental commands. It's the absence of such self talk. Some players have a natural ability to do this, others have a harder time letting go of the internal dialogue.

Imagine running down a mountain. If you try to pick every step, you stiffen up and you'll fall. But if you abandon yourself and let go, you somehow can run with effortless ease, and your subconscious mind picks the steps far better than your ego mind, your internal dialogue, ever could.

That's the zone. That's where you want to be when you play. Stress interferes with achieving that state of mind. The heightened stress of a "clutch" situation affects players in different ways, and the players who can maintain their composure under such stress are the players I would call "clutch". The evidence is all anecdotal, probably known only by the participants.

It's pretty clear after all the discussions we've had that it's not statistically measurable. But don't ever tell a player that it doesn't exist.
Excellent post.

MWM
04-23-2006, 09:02 AM
Honestly, if it exists it exists. If it doesn't it doesn't. Just because players do or don't believe it doesn't change that. One thing we've learned over the years is that what players and managers believe and actual reality aren't always one in the same.

And everything you're describing RFS is the ability not to choke.

redsfanmia
04-23-2006, 09:08 AM
What is the difference between clutch performance and not choking? I think its one in the same.

RFS62
04-23-2006, 09:13 AM
Honestly, if it exists it exists. If it doesn't it doesn't. Just because players do or don't believe it doesn't change that. One thing we've learned over the years is that what players and managers believe and actual reality aren't always one in the same.

And everything you're describing RFS is the ability not to choke.


Yeah, I agree. It's the ability not to choke.

But just because something can't be expressed mathmatically doesn't mean it doesn't exist.

I've always wondered why many of the outstanding mathmeticians studying the game and making such strides in understanding what's going on find it so easy to completely ignore the psychological side of the game.

It's not enough to know what happened. That's what stats tell me. And they have evolved into pretty good predictors of what will happen.

I want to know more than that. I want to know WHY it happened. And the myriad of psychological, physical, and HUMAN interactions that go on before and during play are infinitely more interesting to me than counting numbers after the fact.

I'm not knocking statistical analysis. But trying to use statistics to define "clutch" is like using a hammer to change a tire. It's the wrong tool for the job.

The problem, it seems to me, it that the statisticians framed the argument when the subject first started being discussed. They defined the parameters within which it would be considered. I heard over and over that there's no measurable evidence of increased OPS or Batting Average with RISP. And that's clearly true, as has been shown over and over.

But that's not what a player thinks of when he thinks of "clutch". The definition of terms is much more broad to the participants. So, we're really arguing about more things than the statisticians can address, and the statisticians insist on saying the argument is over once the range of what they can comment on is done.

I don't understand how so many smart people can completely ignore the psychological and behavioral side of sports just because it can't be mathmatically expressed.

Newman4
04-23-2006, 09:17 AM
My fundamental question with those who claim to be able to "raise their game" in certain situations is this : If you are always performing at you maximum efficiency then how can you raise your game? To me, the there is only room for "lazy" people to perform up to their actual potential, thus making them appear to be "clutch" performers or people who perform at their highest level and because of external factors fail to perform at this level in key situations, thus appearing to be "chokers".

Certain players have more success against certain teams for some reason, like Roy Oswalt against the Reds. Is he clutch? I don't think so. He's just pretty good anyways and by coincident has a good run against one team.

MWM
04-23-2006, 10:00 AM
RFS, I'm not sure how this became a stat vs. psychological discussion. My beliefs on the matter have nothing to do with math and everything to do with psychology and the physics of hitting. And yes, I agree, just because it can't be proven with numbers doesn't mean it doesn't exist. But that doesn't mean it does either. And just because players and managers say it exists doesn't make it so either.

And just because someone doubts the existence of "clutch" hitting, doesn't mean they're ignoring the psychological in the name of math, which seems to be your implication. It seems like every time someone who's seen as a "stat guy" says something questioning conventional wisdom the argument always comes back about how they're ignoring the mental aspect of the game, when it's not even close to the truth. Believe me, I do understand the mental aspect of the game and I do give it its proper due. I think you can make a serious case for the absence of clutch without using a single number, which I did, although at a very basic level, above.

RFS62
04-23-2006, 10:11 AM
RFS, I'm not sure how this became a stat vs. psychological discussion.


I'm not knocking your beliefs. You and I have had this discussion plenty of times, and I know exactly how you feel, and I agree with you.

What I've observed in all these threads though is statistical evidence only, with little if any regard for what I consider the bigger picture. The psychological side of performance is a big interest of mine, and what I enjoy talking about much more that the statistical side. And I think it goes largly ignored in many of our discussions.

IslandRed
04-23-2006, 11:28 AM
I'd bought "Baseball Between the Numbers" but hadn't read it cover-to-cover. Nate Silver spends about 20 pages on "clutch," so I read that just now. Interesting conclusions. For example, he pointed out the obvious fact that all situations are not the same; macro-view metrics like RC are not exact enough in a discussion about clutch because the value of different outcomes changes with the situation. Also, it's not good enough to draw a line and proclaim at-bats "clutch" and "not clutch." For example, a guy who leads off the game with a home run increases his team's chance of winning by 9% -- a much higher impact than many supposedly clutch situations that occur late in the game. Clutch should be defined along a spectrum, not as an either-or. The best tool for that is Win Expectancy -- based on the situation when you came to plate, how much did the result of your at-bat change the likelihood of your team winning?

He attacked the problem combining the player's normal hitting ability and leverage (which measures the situational importance of the player's at-bats) to predict a Win Expectancy, and compared that to actual Win Expectancy results. He found -- surprise -- that David Ortiz was mondo clutch in 2005, adding nearly four wins to the Red Sox simply from his ability to produce extra in key situations.

And then to the equally unsurprising news: Ortiz' 2005, like most big clutch seasons, was an aberration. Take last year out of the equation and his career clutch ratings hover around normal, maybe 1.5 wins to the good. The Red Sox should not be banking on Ortiz continuing to be money... at least, not beyond whatever his normal level of performance is.

After running the same numbers on players with a minimum of 5000 PA since 1972, Silver concluded that there's enough clutch out there to support the notion of it existing, but the effect is negligible when looking at the overall performance of a hitter. Maybe 2% of a hitter's total ability to produce wins at the plate relates to his ability to exceed (or not) his norms in high-leverage situations.

gonelong
04-24-2006, 12:39 AM
If it cannot be reliably identified, does it even matter if it exists or not?

If I was interested in looking for it, WPA would be the first place I'd begin looking

In my personal opinion, "lack of clutch" or choke does exist at the MLB level. Clutch itself, not so much, IMO. You have to have some level of talent/drive to make a MLB roster, and IMO, pretty much negates "clutch" from one guy to another.

I can't bring myself to put much stock into what players/GMs/announcers have to say ***FOR PUBLIC COMSUMPTION*** ... so when they talk about clutch, it doesn't hold any sway for me.

GL

RedsManRick
04-24-2006, 12:54 AM
Yeah, I agree. It's the ability not to choke.

But just because something can't be expressed mathmatically doesn't mean it doesn't exist.

I've always wondered why many of the outstanding mathmeticians studying the game and making such strides in understanding what's going on find it so easy to completely ignore the psychological side of the game.

It's not enough to know what happened. That's what stats tell me. And they have evolved into pretty good predictors of what will happen.

I want to know more than that. I want to know WHY it happened. And the myriad of psychological, physical, and HUMAN interactions that go on before and during play are infinitely more interesting to me than counting numbers after the fact.

I'm not knocking statistical analysis. But trying to use statistics to define "clutch" is like using a hammer to change a tire. It's the wrong tool for the job.

The problem, it seems to me, it that the statisticians framed the argument when the subject first started being discussed. They defined the parameters within which it would be considered. I heard over and over that there's no measurable evidence of increased OPS or Batting Average with RISP. And that's clearly true, as has been shown over and over.

But that's not what a player thinks of when he thinks of "clutch". The definition of terms is much more broad to the participants. So, we're really arguing about more things than the statisticians can address, and the statisticians insist on saying the argument is over once the range of what they can comment on is done.

I don't understand how so many smart people can completely ignore the psychological and behavioral side of sports just because it can't be mathmatically expressed.

But if that's the case, that "clutch" exists but it cannot be measured, then what is it actually doing and why should we care about it all. If it's not effecting the outcome of the contest, I'm not sure why we should care if a performance, or a player, is clutch. If it is effecting the outcome, then we should be able to measure it, no?

Outshined_One
04-24-2006, 01:29 AM
I think one of the critical aspects that has to be addressed in this situation is a player's overall performance compared to that player's clutch performance.

Let's take David Ortiz, for example. People who believe that clutch hitting exists will point to him as a terrific clutch hitter who has had a number of game-changing (if not walkoff) home runs in his time with the Red Sox. Most managers would want some one like him at the plate in a critical situation.

However, there is a point worth making regarding this line of thinking. David Ortiz is one of the best hitters in baseball, period. Whether it's the first inning or the ninth inning, any manager would want him at the plate. I would expect his career numbers would reflect his abilities in nearly any hitting situation, especially in clutch situations. Given a large enough sample size for clutch hitting, I would expect the two sets of numbers to look really similar to one another. It's not like Terry Francona would prefer to have JT Snow at the plate over David Ortiz, you know?

There's also another thing worth bringing up here. How can we define clutch? I think almost everyone would agree that bottom of the ninth with two outs in a tie game would be a clutch situation. However, what about a man on second base with two outs in the seventh inning in a pitcher's duel with the opposing starter momentarily struggling? When is a game really on the line? What about RISP, which counts whether a team is down by 12 or up by 12? Are there enough clutch situations out there to effectively measure clutch hitting in a given player?

Michael Jordan was probably one of the greatest examples of a clutch performer in professional sports. He had a sense of The Moment. Through effort and sheer force of will, he could squeeze a win out of his team no matter the circumstances more often than not. If the game was on the line, I'd want the ball in his hands. However, he was one of the greatest basketball players ever. There is something to be said for the fact that he was just so incredibly good that it carried over into clutch situations.

SteelSD
04-24-2006, 01:38 AM
But just because something can't be expressed mathmatically doesn't mean it doesn't exist.

And yet, for something to exist it must be able to be identified. If we can't find evidence that it exists, then we can't assume it exists.


I've always wondered why many of the outstanding mathmeticians studying the game and making such strides in understanding what's going on find it so easy to completely ignore the psychological side of the game.

Hmn. I guess my take on that is 180-degrees opposite. From where I sit, the only folks who've actually been out looking for "Clutch" are the outstanding mathemeticians to whom you refer.


It's not enough to know what happened. That's what stats tell me. And they have evolved into pretty good predictors of what will happen.

I want to know more than that. I want to know WHY it happened. And the myriad of psychological, physical, and HUMAN interactions that go on before and during play are infinitely more interesting to me than counting numbers after the fact.

I'm not knocking statistical analysis. But trying to use statistics to define "clutch" is like using a hammer to change a tire. It's the wrong tool for the job.

The problem, it seems to me, it that the statisticians framed the argument when the subject first started being discussed. They defined the parameters within which it would be considered. I heard over and over that there's no measurable evidence of increased OPS or Batting Average with RISP. And that's clearly true, as has been shown over and over.

The parameters have moved far past simple RISP scenarios to those involving high-leverage situational Win Probability value tracking. The search has evolved to include studies involving situations in which any reasonable player could be expected to feel additional stress and/or see a manifestation of anxiety above and beyond that which we could expect him to feel otherwise.

The search currently involves a level of sophistication you may not have considered.

And yet, whenever something is found it's transient and/or very low-level performance above what a hitter would do normally.


But that's not what a player thinks of when he thinks of "clutch". The definition of terms is much more broad to the participants. So, we're really arguing about more things than the statisticians can address, and the statisticians insist on saying the argument is over once the range of what they can comment on is done.

For "clutch" to exist as an active skill set, it has to translate to performance somewhere and every little crumb of performance is tracked. If it can't translate to performance then it either doesn't exist or it doesn't matter if it does exist. Those are your current selections.

BTW, I agree with you on the existence of "the zone". Problem is that "the zone" doesn't appear to be anything but a transient mental state. Players find themselves in "the zone" and then they're out of the zone. They try to get back in the zone but can't. If the zone itself were capable of being entered at will, we'd find athletes who could enter it at will and/or stay there all the time. But I've yet to see a single athlete build a house in the zone.

Well, except for Michael Jordan. He had plans for a housing development there but right now he's just the association board president who's waiting on his first real live tenant to show up. People keep popping up to browse the lots, but they pop back out before Mike can get so much as a list of the association covenants in their hands.

So MJ sits patiently and re-reads notes left by past prospective zone tenants. The one he just put down reads, "Orel Hershiser was here."


I don't understand how so many smart people can completely ignore the psychological and behavioral side of sports just because it can't be mathmatically expressed.

If those "smart people" were "completely ignoring the psychological and behavioral side of sports", they wouldn't be the ones looking behind trees and under rocks for a skill set that no one before them has been able to find.

I've looked for it myself as have thousands of others.

You don't go out looking for Bigfoot if you ain't interested.

Cedric
04-24-2006, 02:36 AM
And yet, for something to exist it must be able to be identified. If we can't find evidence that it exists, then we can't assume it exists.





But what are you trying to identify? Before anything can be looked for there has to be a clearly defined definition of what IT is. From this thread it seems that is the issue as much as anything.

SteelSD
04-24-2006, 04:03 AM
But what are you trying to identify? Before anything can be looked for there has to be a clearly defined definition of what IT is. From this thread it seems that is the issue as much as anything.

The issue is that, regardless of what "definition" is used, either nothing can be found or that which is found isn't all that meaningful. Those are the search results.

Not speaking to you, Ced, but one of the biggest problems is that every time someone returns from a search, there's a "clutch" proponent there telling them that they're don't know what they're supposed to be looking for, or aren't looking for the right thing, or didn't do the search right, or weren't looking in the right place. Or worse- that "clutch" can't be found but it must exist because they swear they've seen it.

After a while it just gets kind of tiring to hear about search issues when the real problem is that nothing can be found.

Wheelhouse
04-24-2006, 10:58 AM
Glad I started the thread--in spite of some initial responses from those who felt it was invalid discussion, it seems like some lively and thoughtful discussion on the subject has emerged...

blumj
04-24-2006, 11:15 AM
Glad I started the thread--in spite of some initial responses from those who felt it was invalid discussion, it seems like some lively and thoughtful discussion on the subject has emerged...
Did you read the whole thread (or most of it) that came out of the original post on sosh? It's almost like an experiment in cross-pollination or something.

Johnny Footstool
04-24-2006, 11:29 AM
Seems like we have this discussion two or three times a year, and it always unfolds the same way.

While I agree there are clutch situations, and sometimes players come through when the game is on the line, I think it's more a function of luck than any kind of skill.

Francisco Cabrera came through in the clutch in 1991. Luis Gonzalez came through in the clutch in 2001. Neither one did so via any special skill outside the realm of their normal abilities.

Jr's Boy
04-24-2006, 11:29 AM
Why so much adulation over Sam Horn with these guys?

TRF
04-24-2006, 11:29 AM
Clutch exists. We are discussing it right now.

However, The definition of what clutch is... that is the sticky point. Also in baseball clutch is almost entirely attributed to hitting. It is clearly true that it exists in pitching, especially when discussing closers. Rivera was clutch. Graves... not so much. You could see how Rivera thrives under pressure. He is beyond good. In fact he is good beyond his incredible talent.

I also believe clutch is mental in that ballplayers believe in it. And belief can affect the outcome of a game. Pitchers know who they are facing, and a player that has the rep of being a clutch player (Joe Randa) can get in a pitchers head. And every "clutch" performance adds to that rep, deserved or not.

Johnny Footstool
04-24-2006, 11:32 AM
Pitchers know who they are facing, and a player that has the rep of being a clutch player (Joe Randa) can get in a pitchers head. And every "clutch" performance adds to that rep, deserved or not.

That reputation for succeeding in the clutch also causes people to overlook all the times that particular player fails in the clutch.

Clutch is a situation, not a skill.

TRF
04-24-2006, 11:37 AM
Sure it is, but the definition of what is a clutch player is hard to determine.

Is it close and late only? Is it getting a hit with 2 outs and nobody on? runners on? bases loaded? Is it determined by BA? OBP? SLG? OPS?

See that is the biggest problem. determining what a clutch situation is, then applying a metric to it.

M2
04-24-2006, 11:43 AM
My guess is that if clutch exists, it exists in "big game" situations not in specific at-bats.

Check how a player does in playoffs and late pennant race situations and that might net you something. For instance, if a player can improve or maintain his overall game (like Derek Jeter) against the best of the best, then maybe you've stumbled into a "clutch" player.

It's also possible that "clutch" isn't something that you can own. Bob Gibson was clutch in the 1967 World Series, or perhaps dominant is a better term. Jack Morris was clutch/dominant in the 1991 World Series. Yet Gibson got beat in Game 7 of the 1968 series and Morris fell flat in the 1992 playoffs.

You couldn't get Ken Griffey Jr. out in a key situation in the 1995 playoffs, but in 1997 he laid an egg. Anyway, building on what's already been said in this thread, maybe "clutch" is real just being "in the zone." We all know that happens. Players get hot. Yet maybe what the stats have been telling us is that the pixie dust of right time/right place wears off.

RedsManRick
04-24-2006, 12:21 PM
Again, we get back to the purpose of discussing clutch. I think everybody is in agreement that there are situations which significantly tax a player's abilities -- where the outcome of said situation has a proportionally greater impact on the team's success than nearly any other situation. Furthermore, I think we all agree that performing well in these situations is something to be lauded. However, the disagreement seems to lie in the "purpose" of using the term clutch.

First, my definition of clutch: A sitution is "clutch" when:
The value of the outcome of the situation is significantly greater than the value of the same specific event outcome in it's typical occurance. That is, there is a disproportionally large aspect of "control" over the larger outcome. Ex. A pitcher striking out a batter clutch when it not only ends the current at bat, but ends the game or better yet, the season.

In the personal context, there is an assumption that human nature makes it more difficult for a person to perform a given task when the significance of the outcome is greater. A "clutch" player is able to overcome this natural tendency and thus has greater value. I think there is an assumption that the "normal" effect of clutch situations is a decrease in the expected level of performance.

When we try to claim that a given individual is clutch, it seems we are making a statement that they are more likely to perform up to their full ability in such situations than a "non-clutch" person. If that's the case, numbers would bear out that ability. Every study I've seen suggests that the ability to do so either doesn't exist, or the effect is so minimal as to be irrelevent.

Clutch performances should be lauded as historical occurances. However, one should not base current managerial decisions based on past clutch performances. The basic point is that in any "clutch" situation, you want your best player at the plate, taking the shot etc. I don't care how clutch Robert Horry has been, if I'm down a point with 10 seconds to go, our chances of winning are best if the ball is in Kobe's hands. If the bases are loaded with 2 outs and we're down 3, I want Dunn or Junior at the plate over Rich Aurilia or Brandon Phillips. Would the Yankees want Aaron Boone at the plate instead of ARod if they needed 10th inning heroics again?

Let's celebrate those occasions where fate smiled on somebody and they succeeded when it mattered most. Let's call it a clutch performance and move on. But assigning the clutch label to people is a waste for the reasons we've already discussed.

All that said, I am curious however regarding the effect of sample size. Perhaps part of the problem with these analyses is that no one player has enough opportunity to bear out a "clutch" skill beyond random variation. Furthermore, if it is a very rare skill, the fact that it is a non-normal distribution with a very pronounced middle peak would cloud the analysis. A few outliers can be written off as outliers.

Perhaps aggregating player performance by some set of characteristics, be it game-relevent skill sets, personailty traits, etc., it would be easier to see statistically significant variation in "clutch" sitations. The problem of course is find the set of characterstics common to "clutch" individuals. Seeing as how these studies have been done by people smarter than me, I'm going to guess that somebody has already gone down this road...

Chip R
04-24-2006, 12:29 PM
Sure it is, but the definition of what is a clutch player is hard to determine.

Is it close and late only? Is it getting a hit with 2 outs and nobody on? runners on? bases loaded? Is it determined by BA? OBP? SLG? OPS?

See that is the biggest problem. determining what a clutch situation is, then applying a metric to it.

And that is very true. If a player hits a home run in the 1st inning of the 7th game of the World Series and his team wins 1-0, was that a clutch HR? If a guy drives in the winning run from 1st base is that not a clutch hit since the winning run wasn't in scoring position? If a guy hits .350 in the clutch, he's failing the other 65% of the time. That's a lot of failure. And it's also dependent on the runners on base. Take the situation in MIL on Friday night. Freel hits a chopper to 3rd and Phillips, who was on 2nd, scores on the play because he was fast and he was running hard from the time Freel hit it. Freel gets a RBI and a hit with a RISP. Put Rich Aurilia on 2nd instead of Phillips and there's no way he scores. Freel gets a hit with a RISP but no RBI. So it can be tricky defining clutch.

IslandRed
04-24-2006, 12:36 PM
Perhaps aggregating player performance by some set of characteristics, be it game-relevent skill sets, personailty traits, etc., it would be easier to see statistically significant variation in "clutch" sitations. The problem of course is find the set of characterstics common to "clutch" individuals. Seeing as how these studies have been done by people smarter than me, I'm going to guess that somebody has already gone down this road...

In the Nate Silver piece I referenced earlier, he had done something like that, and concluded that there's a slight tendency for guys with good bat control and plate discipline to be more clutch than low-walk, high-K hackers. Makes a little sense, as in high-leverage situations a hitter who can control the at-bat and maybe hit a pitcher's pitch is better off than a mistake hitter who is often at the pitcher's mercy. But it's just a slight tendency, and given that we're talking a very small slice of the hitter's total picture anyway, not something a team would bank on.

smith288
04-24-2006, 01:02 PM
Ill do y'all a favor...

Clutch Defined:

http://www.coastlinetrans.com/usa/images/clutches/clutch%20copy.jpg

RedsBaron
04-24-2006, 01:04 PM
It's also possible that "clutch" isn't something that you can own. Bob Gibson was clutch in the 1967 World Series, or perhaps dominant is a better term. Yet Gibson got beat in Game 7 of the 1968 series.


Gibson pitched great in the 1968 Series; he won game one, striking out a record 17 hitters, he won game four, and he lost game seven, 4-1, primarily because Gold Glove centerfielder Curt Flood misplayed a flyball.
Gibson was a great regular season pitcher of course. He went 251-174 for his career, with a 2.91 ERA. In 3884 innings he fanned 3117 and only walked 1336. He had 255 complete games and 56 shutouts.
For all that, Gibson was significantly better in the postseason. He went 7-2 with 8 complete games in nine starts. He had a 1.89 ERA. In 81 innings he fanned 92 and walked 17. He threw two shutouts. Twice he won game seven of the World Series, once on two days rest.
Sandy Koufax was another pitcher who was even better in the postseason than the regular season. His career mark was 165-87, with a 2.76 ERA, 40 shutouts, and 2396 Ks in 2324 innings. In the postseason, Koufax was only 4-3 (he lost a game in the 1959 Series 1-0), but his ERA was 0.95, he had four complete games and two shutouts, he fanned 61 and only walked 11 in 57 innings (he allowed only 36 hits). In 1965, after pitching a shutout in game five, on two days rest he threw another shutout in game seven, as LA won 2-0.

Team Clark
04-24-2006, 01:08 PM
Nice post.

And well said, that "clutch" is too broad a term to use across the board for every motor skill involved in baseball.

Consider that the pitcher is the only player on the field who actually initiates the action. The play doesn't start until he does. Everyone else is in reactive mode. He's the initiator.

The mindset behind pitching is very similar to golf. You imagine the action you're about to initiate, and you start the play. If you're prone to mental distractions or negative thinking/visualization, you're in big trouble.

Hitting and fielding is reactive in nature. Similar to return of serve in tennis. You are waiting for someone else to start the play, and reacting rather than initiating. Good fielders go over their responsibilities before the play starts, then shift to visual thinking as the play starts.

Hitters do the same. Good hitters, to varying degrees, think about what the pitcher might throw, then shift to visual thinking, totally reactive and non-verbal thinking. If they're engaged in internal dialogue when the pitch starts, they're doomed.

Choke is too broad a term to encompass all these motor skills and the thought process involved in each.

The ability to quickly shift between linear thinking and internal dialogue analysis to non-linear awareness is a key skill that's not talked about much. In fact, many great athletes do this and could never engage in a discussion about it... they just do it and don't realize the dichotomy between the two.


Now, how did I know you were going to bring Golf into this? :laugh:

MWM
04-24-2006, 04:15 PM
http://www.sportsecyclopedia.com/nl/atlbraves/Lemkeatl.JPG

smith288
04-24-2006, 04:20 PM
Mark Lemke...

Scrappiness with stylish specs

MWM
04-24-2006, 04:28 PM
One of the funniest signs I've ever seen a fan display at a game was when the Braves were in the World Series when Lemke had a few big hits. Someone in the opposing crowd was holding a sign saying, "What's a Lemke?"

Raisor
04-24-2006, 07:12 PM
One of the funniest signs I've ever seen a fan display at a game was when the Braves were in the World Series when Lemke had a few big hits. Someone in the opposing crowd was holding a sign saying, "What's a Lemke?"


Lemke does the pre and post game radio show for the Braves. He's HORRIBLE.

This is his second year with the job, I'm hoping they find someone else next year.

blah.