Whitlock: "Stat geeks are ruining sports"

[Baldkin]

Now it may appear that the statheads are pushing away from Wins and RBIs so much that they are writing them off as useless, but it is more because of the need to convince the average fan that they should not the holy stats they used to be.

They aren't useless, they have their context, but for predicting how well someone will do in their next season (or next AB) it is about 45th on the list of things I would look at.

[topper09er]

I'm sure there are certain factors that contribute to streakiness

(health being the greatest I suspect). I hope we can all agree that the

vast majority of time that when a player is deemed hot or cold, they

are simply exhibiting a random bunching of good or bad performance.

 

I am not willing to easily accept this, I think the results (batting average) are more of a bunch of random good and bad outcomes (balls finding holes, or lineouts), but performance is different than results. I think a player can bunch together good performances for reasons other than randomness.

 

I would absolutely love for you or anyone else to prove your ability to identify which players are hot by watching them.

 

This is not easy obviously, but since I believe hot streaks exist for reasons other than randomness then there must be some way to identify it. I cant prove it and maybe I don't know the right thing to look for, but I think there must be a way to do even if no one can do it consistently now. It may come down to simply asking the player how they feel, because I know from experience of just playing in high school that there were times when I knew I was gonna hit it hard 4 times that day and other days when I really hoped to not see first pitch strikes on the corner and fall behind in the count early.

 

Obviously I would think someone who has spent 30 years in pro baseball would have so much more experience than me that they should be able to be better at identifies these things than me. There are definetly things that can be learned from watching the game that stats don't show, at least right now. I would not be surprised in the future to see stats that use cameras on the batter to identify things like bat path and swing speed to correlate with other things, but as of now the only way to gain this information is to watch the game. I can tell from coaching that a good coach should be able to pick a team just based on watching the kids play catch for 10 minutes on the first day of tryouts. You don't need to see anything else to know who the best players are going to be.

[JimH5]

Whitlock is basically complaining because he can no longer write down any garbage and turn it in. Now he has to have evidence, which makes his job harder.

 

Great observation.

 

How many of us screech at Davey Nelson or Rob Dibble or Mike Francesa whenever they tell us things we know are factually incorrect. It's the whole premise of the Stat Boy/Tony Reali segment of Pardon the Interruption. Two loudmouths get to rage on and on about something, and then they get put in their place at the end, when there is data that proves them wrong.

 

Stat-geeks can ruin the fun of sports, the same way that stat-avoiders can. Joe Morgan was a terrible analyst because he refused to acknowledge that newer statistics can tell us things we didn't know before. I get annoyed with some stat geeks when they can't even enjoy it when something happens that the stats say shouldn't.

 

Like with Yuni. . .I really dislike having him on the team, but sometimes when he does make a good play, it's like some people are angered by it--since that's contrary to what their beliefs about him say.

 

In the end, this stuff is supposed to be about fun. Nobody likes an arrogant guy who thinks he's all that.

 

Whitlock isn't always right about everything. But he is a great writer and he often raises interesting points, and offers a unique perspective.

[rluzinski]

I'm sure there are certain factors that contribute to streakiness

(health being the greatest I suspect). I hope we can all agree that the

vast majority of time that when a player is deemed hot or cold, they

are simply exhibiting a random bunching of good or bad performance.

 

I am not willing to easily accept this, I think the results (batting average) are more of a bunch of random good and bad outcomes (balls finding holes, or lineouts), but performance is different than results.

Of course they are but at the end of the day, the batter still has some (unknown) percent chance of getting a hit in a particular AB. It is stochastic in nature as a result. Because, we HAVE to see some amount of streakiness in performance. We can even make some estimates on what minimum level of streakiness we would expect from luck alone.

"I think a player can bunch together good performances for reasons other than randomness."

Of course they can. Who is arguing otherwise? That mostly gets drowned out by the "coin flip" streakiness of the results. It's fairly easy to prove this. The streakiness we observe in the results is simply a combination of luck and variation of day-to-day skill (from health, "seeing the ball well" or whatever else you want to attribute it to). We can estimate the variation we expect from the coin flip luck and see that it makes up a large portion of the the overall observed variation.

I can't say this enough. I am not arguing that there isn't non-luck based variation. Perhaps some are so resistant to sabermetric principles because they don't really even know what they are in the first place! If that's even in part because of self-professed "stat geeks" misusing the science, that is unfortunate.

[mothershipconnection]

Briggs just toppled the biggest, baddest strawman of them all: the fundamentalist stat cult member who never played baseball, doesn't watch games, and makes definitive statements that no legit SABR expert would ever agree with.

[Funketown]

Briggs just toppled the biggest, baddest strawman of them all: the fundamentalist stat cult member who never played baseball, doesn't watch games, and makes definitive statements that no legit SABR expert would ever agree with.

No he didn't, he didn't say anything about living in a basement and failing to communicate with the opposite sex.

 

I think it was Dave Cameron who wrote a good article on this topic on Fangraphs. Pretty much about how he found it frustrating the way Moneyball portrayed the SABR crowd. At this point, I think a lot of the top guys have a huge amount of respect for the value of scouting, etc.

[jerichoholicninja]

Sports writers who base their opinions off of intangible things like grit and clutch ruin sports for me.

[gregmag]

On this board, the most statistically sophisticated posters that I can think of happen to be among the most thoughtful and reasonable people who regularly post. I'm thinking especially of Russ. We also have some equally smart, nuanced, and thoughtful stat skeptics. I'm thinking especially of danzig. Most of the best posters on the site are somewhere in the middle methodologically -- people who have a facility with stats but most commonly make posts that don't reference advanced stats heavily. I'm thinking especially of Toby. I learn from all three of those gentlemen, and from lots of others who fit each of those descriptions.

 

Zealotry, as far as I can understand it, inheres in believing that you are always right about matters that most would agree are contestable; treating reasoned disagreement as an occasion to cry out in martyrdom; and stereotyping most or all people who disagree with you as sharing negative character traits. In my view, very few people who post on this board -- statheads, stat-skeptics, or just plain folks of any stripe -- reflect those characteristics. I imagine the person who started this thread thought that, by doing so, he would expose the zealots. I wish he hadn't tried, because the effort only poisons our discourse. But he sure did succeed.

[Bombers]

Jason Whitlock is a nob and I dismiss everything I hear from him as there is basically nothing I agree with him on. He is ruining objective journalism with his claims that come off as fact and not just his opinion. Or maybe he should try and understand statistics than to just write them off.

Jason - do some homework, understand your topic, and create an informed opinion. To say stataticians are ruining the game is completely ridiculous. As others have said it's just another tool in the managers tool belt to try and give his team the best chance at winning.

[Funketown]

I just wasted two minutes of my life reading the article. He ends the article with "STFU". Clearly, Whitlock is one of the most brilliant men of our generation.

[Raw Biz]

I would absolutely love for you or anyone else to prove your ability to identify which players are hot by watching them.

 

This is not easy obviously, but since I believe hot streaks exist for reasons other than randomness then there must be some way to identify it. I cant prove it and maybe I don't know the right thing to look for, but I think there must be a way to do even if no one can do it consistently now. It may come down to simply asking the player how they feel, because I know from experience of just playing in high school that there were times when I knew I was gonna hit it hard 4 times that day and other days when I really hoped to not see first pitch strikes on the corner and fall behind in the count early.

I don't think this works. I'm pretty sure McGehee said a few days ago that he feels great at the plate and that he feels like he found his swing (or something to that effect). Roenicke has also said this at other points of the season. Obviously McGehee's "great feeling" hasn't resulted in much success.

[topper09er]

"I think a player can bunch together good performances for reasons other than randomness."

 

Of course they can. Who is arguing otherwise?

________________________________________________________________________

Um what does this mean then?

 

"I hope we can all agree that the vast majority of time that when a player is deemed hot or cold, they are simply exhibiting a random bunching of good or bad performance."

[homer]

"The stat geeks won't let you argue." That is a terrible blanket statement. Most "stat-geeks" are simply looking for ways to better understand the game and would like as many tools at their disposal to do so as is possible. If someone wants to just trust their eyes or just trust the spreadsheet while more or less ruling out the other that is obviously their prerogative but it will lead to a very limited view of the game.

Take if from someone who has questioned the stat geeks. Whitlock is right on the money. Check out the threads that discuss Mark Kotsay for instance. Because some stat geek came up with "Wins above Replacement" as a measure of value, and Kotsay has been marginally negative on this for the last several years. the argument (and if you refute it you are an idiot) is that Kotsay is completely worthless and doesn't belong in the game. The true stat geeks are zealots. Sabermetrics is their religion.

You believe you learned everything there is to know about baseball about 30 years ago. I haven't. You are always certain that your opinion is correct. I don't. Who's the zealot?

Well said, Russ.

[TooLiveBrew]

It's just annoying that some people feel the need to expound upon how wronged they are by "stat geeks". You know what ruins this kind of discussion or "sports" for me? Self-righteous moaning.

[Flood Pants]

"Stat geeks" are the reason why I came to love baseball. It sure as heck wasn't because I grew up playing the game, because I didn't. I could never hit. I could never throw. I could never field. I would stand in right field and zone out while balls were hit over my head. Sports made absolutely no sense to me because my mind simply isn't wired for coordination and reflexes. As a result, any attempt to play sports usually resulted in embarrassment and teasing. Thus, I was alienated from the world of sports. But I was always good at academics. Abstract things that other people had trouble visualizing always came easy to me. Your prototypical "geek" if you will.

 

Once I discovered that baseball had a subculture of geeks who crunched numbers, I was sold. I had no attachment to the traditions of baseball because they were never a part of my life. I began to view sports not simply as these embarrassing activities that I could never succeed at, but as a rich world of cause and effect ripe for analysis. As an intellectual pursuit. As something progressive, not static. As something that, like a great novel or film, became more complex and meaningful through dissection and discussion. I know I don't speak for everyone, but stat heads certainly haven't ruined baseball for me, they've saved it.

[chuckiehacks]

Keith Law has been pretty interesting on this subject lately. In all the Moneyball movie hype, it's been discussed that Law was cocky and VERY anti-scout in his interviews with Michael Lewis when Lewis was writing the book. Law addressed this the other day, talking about how is views have evolved over the years - that he now sees the balance between SABR-based evaluations of players and traditional scouting.

 

I think this discussion is similar. There have certainly been advances in stats over the years, yet most players will tell you that there are times when they are "seeing the ball well." The challenge for both camps in this situation is trying to predict when the hitter will stop seeing the ball well. Fundamentally you can't, no matter what stats, data points or eyeball evaluation you use.

[TooLiveBrew]

Stats describe what has happened. Scouting explains why it happened. I can't think of one person on this board that celebrates being anti-scouting, but there sure seem like a lot that celebrate being anti-stats.

[reillymcshane]

A walk is better than a hit!?

 

Yes, that's exactly what stat geeks so.

 

Or maybe it's just a line people like to attribute to others to denigrate them and their ideas.

[H4P]

I find the discussion on streaks particularly interesting. Yes, players go through phases where they swing the bat particularly well and hence produce many hits. At the same time, I think we've all seen periods of time where someone was swinging the bat well and in a SLUMP. And vice versa. I do have some major league experience as I was called up recently in MLB the Show so I thought I'd share that direct knowledge.

 

As far as "stat geeks" go, there are definately those that go over the top and muddle a good conversation. But I've been reading this board for years (though I don't post much) and I've learned more about the game of baseball reading some of these spirited discussions than I could ever explain. Some threads become unreadable to me because of the (in my opinion) overkill on stats. So...I don't read them.

 

Other conversations on this board seemingly meander meaninglessly until they can be framed with some numbers that add base and substence to the argument.

 

The reason that I don't post much is that I don't feel I often have much to add to good discussions as I'm one of the people who follow every game either on TV or the radio. I have a general feeling of who's doing what based on what I see and hear every day. I dig into the numbers from time to time but I'm the kind of fan who likes going to games, complaining about the manager and overreacting to winning streaks. So that's fine for me.

 

But to say stat guys are pretentious or ruining sports is just ignorant. EVERYONE thinks they're opinion is right...I hope. The dichotomy in baseball is why I love it. I started taking my significant other this year who never watched baseball in her life. By July, when a certain SS would come up, she would mutter "He's not very good, is he?" And when Braun came up..."He takes too long to hit but he's pretty good."

[pebadger]

I love advanced statistical analysis as a tool for constructing a roster with an eye toward the 162 games you have to play in the course of a regular season, and maybe more importantly with an eye toward what the roster will look like two seasons from now. I like them a lot in terms of constructing default tactical approaches to situational baseball.

 

That said, it's still a game played by men and you need a manager who can manage those players as both players and people. You need a manager and players who can work together to create an environment that is conducive to optimum performance. Statistical insight can greatly inform those processes, but in the end all the statistical information is still just another input into the decision-making tasks confronting the front office and field management. Figuring how to weight that input in the context of the various decisions to which the information is applied is the real trick.

 

When statistical techniques are applied to any competitive endeavor (business or otherwise), the early adopters usually get a huge initial boost. In essence, they identify and take advantage of the stupidity in their marketplace that usually exists due to reliance on untested assumptions. Once the cat is out of the bag, it doesn't take long for the "new" fundamental truths to become well known throughout the market. It takes a bit longer for people to really adopt and understand the methodologies that went into establishing those new insights. At that point, once all the low-hanging fruit has been picked and the "easy" insights are harder to come by, the real work begins.

 

In a way, this brings us to the kind of ironic part of the "Moneyball" phenomenon, as least as it is generally (mis)understood. People think that using stats makes it easier for the little guys to compete with the big guys. That WAS (more) true, when the little guys were using it to create a competitive advantage when other guys weren't. Now however, everyone is using it and there aren't really huge incremental gains to be made. In fact, to the extent that there are greater insights to be had, the big boys will probably tend to dominate in this field of organizational evaluation and development as well. Finding and hiring folks that can both do the math and then make use of it is expensive. Really good new ideas will likely not spread to all teams quickly; they will be well-guarded proprietary secrets.

[FVBrewerFan]

The problem I have with advanced stats is that some are not true math. Uzr and WAR are examples. Sure some amount of these stats involve math to make them look and feel objective, but they're not. Worse still is when those type of metrics are presented as absolute proof player x is underperforming, lucky, etc.

 

I've learned a lot here, including when to tune out anyone who uses WAR as the sole argument, or even main argument for someone's position.

[LouisEly]

The problem I have with advanced stats is that some are not true math. Uzr and WAR are examples. Sure some amount of these stats involve math to make them look and feel objective, but they're not.

FV, you just hit on one of if not the main issue I have with advanced stats. Those who know me know that I am a numbers, math, and statistics geek. Straight A's through college calculus, someone who uses algebra to center pictures on a wall. But some things just don't make sense. I tried getting into advanced baseball stats a few years ago. I remember going to Fangraphs website and looking at their value/WAR chart... and they had Nyjer Morgan (then with the Pirates) as having more value than Ryan Braun. Damn near fell out of my chair and never went to that website again.

 

Someone made a good analogy with Yuni... yes, he's below average at the plate, you don't need to look any further than OBP to tell that. Likely he has below average range. But people bring out defensive stats - many of which such as UZR and others that have significant elements of subjectivity to them - to say that he "sucks". OK, how do you account for the fact that he has one of the strongest arms in baseball? Some of the throws I've seen him make just leave me in amazement. But that's not measured - it's what % of balls hit to him he gets to, blah, blah. How many more outs does his arm generate whereas an average arm wouldn't get the runner? The reality is that all defensive metrics have significant elements of subjectivity to them.

 

Another example is McGehee. Yes, he's having a bad year at the plate. But over his first 1000+ career plate appearances - no small sample by any means - he demonstrated being a ~.800 OPS hitter. Why the dropoff this year? I don't think many folks remember, or even caught, that a poster earlier this year said he heard through the grapevine that McGehee was having marital problems. Add to that a child with a significant disability (IIRC cerebral paulsy), and those two things combined would distract anyone. Is his off year due to being a bad player, or because of distraction caused by major stress off the field?

 

Why is Grienke so much worse on the road, when Wolf is fairly neutral and Marcum so much better on the road?

 

The reality is that stats can tell us part of the equation, but can not tell us the entire equation. They can tell what happened in the past with precision, but for many reasons are not precise in predicting the future. The bigger point of Whitlock's article, which some have caught, is that sometimes when people offer non-statistical reasoning to explain things that the stats crowd jump down their throats with stats as the end-all of the discussion. And that does tend to happen around here from time to time.

[TooLiveBrew]

OK, how do you account for the fact that he has one of the strongest arms in baseball? Some of the throws I've seen him make just leave me in amazement.

...

How many more outs does his arm generate whereas an average arm wouldn't get the runner?

 

His arm is definitely strong, probably one of the stronger ones in the league at SS. The problem people have with his defense is that his lack of range means fewer chances to put that arm to work recording outs. I have no doubt Yuni's arm allows him to make plays some SS can't, but I'd just rather see one who can get to more balls in play. I think there are more instances of range being a positive than arm. Hairston probably doesn't really have the arm for SS, but he's such a good athlete & fielder that I'd vote for him. Yuni would fit better defensively at 3B.

 

 

The bigger point of Whitlock's article, which some have caught, is that sometimes when people offer non-statistical reasoning to explain things that the stats crowd jump down their throats with stats as the end-all of the discussion. And that does tend to happen around here from time to time.

 

Ok, and the same is true in reverse. I don't see why Whitlock needed to complain about it.

[GAME05]

it's not the stats that can bug me sometimes, it's the sabrmetrician. it's become a jargon and language unto its own, and sometimes particular people (not referring to anyone on this board) will throw it in your face to highlight your ignorance or try to show themselves as the 'real' expert. it's just frustrating to have the person throw out hard numbers at you when you're countering with "but they just SEEM like..."

 

but then it's the same with anything, and not specifically Sabrmetrics. the Joe Morgan's of the world do it by waiving a hand at any number presented to them without any thought. or the coffee snobs do it by telling you that taking cream and sugar means you just don't appreciate coffee the way they do. speaking a select language can have a way of turning people into snobs.

 

but i like basic sabrmetrics and think it's added to my knowledge of players. it's fun to be a lot more accurate on how a player is probably going to progress through his career. the statheads on here calmed a lot of us down in May by telling us that Greinke's peripherals were still good and that he was probably going to be a lot better than what his ERA at the time was showing. it's great when stats can be an addition to a discussion, but frustrating when you make an observation not based on numbers and get scoffed at as if you're an idiot.

[danzig6767]

Most people associate statistical analysis with mainly just baseball, but it's also becoming more advanced in both basketball and football.

 

While i pay some attention to that analysis in football/basketball, baseball is a near ideal sport for really in depth statistical analysis given the largely individual nature to it's game compared to football and basketball which are much more team sports. In baseball, a hitters stats are clear facts and pretty much not influenced by anything else except his own abilities except RBI's. If a hitter has a very poor OBP like say Yuni, in today's baseball world, it's largely impossible now for anyone to make a believable argument that Betancourt isn't a drain on the offense, regardless of what his batting average and RBI numbers say, unlike 10-15-20 years ago. The numbers about how OBP create runs is simply to grounded in fact to be refuted. There are still gray areas like defensive stats and some having to do with evaluating pitchers actual production vs their peripherals/defense around them/park factors, but a lot of baseball stats are just really hard to argue against them without trying to explain away facts. So even though i'm not overly interested in reading up on a lot of the more newer and advanced stuff out there, even when i don't fully understand a stat, i generally believe it will have more validity to it than a person's gut feeling on a player.

 

A sport like football though is such more more team reliant. A quarterback for example can have his performance impacted by the offensive system he plays in. The quality of the line blocking for him and/or by the quality of his receiving weapons. If a running back has a bad run blocking line in front of him, that clearly will impact his performance. A good pass rusher can get fewer sacks if his fellow rushers stink, thus he'll get double teamed a lot more.

 

There are many different examples, but baseball in many cases lack these barriers to coming up with numbers that are difficult to refute and thus i think it frustrates some fans/writers that were used to analyzing baseball mostly with their eyes and gut feelings as they did with football and basketball. Even more frustrating for them is when stats show that their eyes and gut feelings about certain players to have not been as accurate as they thought. That can sometimes be hard to stomach for some because let's face it, a big part of being a fan of sports for many is thinking they instinctually know the game well. Kinda like how it feels good before a sport's draft if you say this/that guy will be a good pro and this/that guy a bust, then later you were right more than wrong.