Concepts some posters could benefit from learning

[emoney_33]

yeah...Sabremetrics would have really went a long way in helping the Sox thwart that Bucky Dent HR...I watch sports to be entertained, not to exert my brain power...yikes

You aren't adding anything to this discussion. You are simply talking down to those who bring up statistics. You do NOT have to study stats or anything. No one is forcing you to do anything. Argue with some logic and valid points, or just ignore the thread if you don't want to discuss the topic. It's really that simple.

 

[Frezo]

well, kind of. Bill James and his ilk are in the business of predicting how age relates to player performance...age certainly is a big factor, and it was a big reason why getting rid of AV for Gost was a great decision.

however, with regard to 29 yard fg's and 34 yo AV, I just don't think the age matters much. a 45 yarder? yes, the age matters, and a lot.

Yes, I can agree with that. The linear scale I proposed, though useless without research, is nonetheless theoretically plausible and could apply to AV and the miss. It could have been that 1 out of 15 instead of the 1 out of 20.

 

[RayClay]

Thanks for the laugh. I have to read a journal article and write on the discussion board for my Research Methods class. LOL!!

 

[TomBrady'sGoat]

yeah...Sabremetrics would have really went a long way in helping the Sox thwart that Bucky Dent HR...I watch sports to be entertained, not to exert my brain power...yikes

it's funny that you say that, because bullpen management is a common topic in statistical analysis. Better bullpen usage, derived statistically, can prevent some painful losses.

 

[emoney_33]

Exactly! They're running it to make $$$$ and improve the bottom line. They're not playing it. Theo will be the first to admit that he got extremely lucky with Ortiz as he wanted to trade him to the Mets at one point for Livan Hernandez only to enjoy his "clutch" performances en route to the WS title.

No one is perfect, nothing is perfect. You bring up one example of one player to disprove things that have been studied with hundreds of players, over decades, over thousands of pitches, at bats etc...

The larger the sample size, the larger the accuracy. Ortiz' sample size was smaller, therefore the accuracy is lower.

David Ortiz is a great hitter, clutch situation or not, he's a great hitter period. He doesn't suck and then become magic in the clutch, he hits at a great rate. Some years we were more fortunate to have a lot of big hits come at the right times

 

[Joker]

it's funny that you say that, because bullpen management is a common topic in statistical analysis. Better bullpen usage, derived statistically, can prevent some painful losses.

Mike Torrez was the best option the manager had that day...bullpen management was never in question.

 

[makewayhomer]

No, you basically get the Red Sox with a lot less money.

Hypotheses are very hard to prove anyway, how many different underlying variables exists between the Indians and the Red Sox besides money? You don't know. You just invalidated the whole experiment.

I'm guessing you're not very familiar with statistical research.

Your whole second paragraph consists of throwing dollar figures and unfounded conclusions around. I'm guessing Theo isn't hiring you to analyze trends.

By the way, let me know when you can analyze the future. I could use a good day at the track.

the point I think the guy was trying to make was "lets see if this sabr run baseball stuff can work when you take away Theo's money". and I rebutted with "well, look at the Clevland Indians, who run their organization very similar but with a lot less money".

wrt to player salaries, if you don't think that diminishing returns set in at a certain point, well, that's just simply not correct and there is lots of evidence out there to back this up

 

[fair catch fryar]

No one is perfect, nothing is perfect. You bring up one example of one player to disprove things that have been studied with hundreds of players, over decades, over thousands of pitches, at bats etc...

The larger the sample size, the larger the accuracy. Ortiz' sample size was smaller, therefore the accuracy is lower.

David Ortiz is a great hitter, clutch situation or not, he's a great hitter period. He doesn't suck and then become magic in the clutch, he hits at a great rate. Some years we were more fortunate to have a lot of big hits come at the right times

I understand what you're saying and to some extent agree with all of this, but you're sending me the impression that you don't think there is an ability that can't be measured in certain people to rise up at critical times and perform. One athlete that comes to mind, purely from memory, is Claude Lemieux. The guy was an average to good player during the regular season, albeit very dirty, but for some reason, when it came to the playoffs (Montreal,NJ, and Col) he had a knack for being around the net and scored a bunch of overtime or last-minute goals.(80 playoff goals, 4 Stanley Cups, and a Conn Smythe). As much as I hated the guy, he was pretty "clutch".

 

[Joker]

You aren't adding anything to this discussion. You are simply talking down to those who bring up statistics. You do NOT have to study stats or anything. No one is forcing you to do anything. Argue with some logic and valid points, or just ignore the thread if you don't want to discuss the topic. It's really that simple.

I studied stats 3 1/2 decades ago at B.U.---never thought that years and years later I would get flame broiled by outraged statisticians...hey,you want to spend the time breaking down every single stat of every single game in every single situation, all power to you...I'm gonna go crack a beer.

 

[RayClay]

the point I think the guy was trying to make was "lets see if this sabr run baseball stuff can work when you take away Theo's money". and I rebutted with "well, look at the Clevland Indians, who run their organization very similar but with a lot less money".

wrt to player salaries, if you don't think that diminishing returns set in at a certain point, well, that's just simply not correct and there is lots of evidence out there to back this up

You can't have it both ways. If you're going to use statistics, you have to use statistical methods, otherwise your results are garbage. You're the one that stated you were going to teach posters something, not me.

Baseball lends itself to statistics much more than other sports do. Way back in the old days, managers and players kept "books" on players situations and so forth which constituted a crude form of statistical analysis. Sorry to be sarcastic, but I'm avoiding posting for my research methods class right now, which struck me as funny.

Because this latest method downplays clutch etc. etc. certainly doesn't mean it discounts it. Statistics is dry and difficult for me to understand and I'm averaging over 99%. You can bet your bottom dollar any coach doesn't need a calculator to know the ball goes to Bird in the clutch (or maybe Garnett now).

It's just another statistical method, there's nothing new under the sun and no method to replace a smart, experienced GM.

 

[emoney_33]

I understand what you're saying and to some extent agree with all of this, but you're sending me the impression that you don't think there is an ability that can't be measured in certain people to rise up at critical times and perform. One athlete that comes to mind, purely from memory, is Claude Lemieux. The guy was an average to good player during the regular season, but for some reason, when it came to the playoffs (Montreal and NJ) he had a knack for being around the net and scored a bunch of overtime or last-minute goals.

I haven't studied these things in great depth, and I'm not going to sit here and say that there are NO people ever who didn't have something special to rise up in certain situations. The human brain is an insanely complex thing and I won't for one second pretend to know everything. However, if a player does better in a certain situation, then there has to be something that can be put down on paper to show that he was better in that situation than in others.

However I am only suggesting that the "clutch" thing is exaggerated and not nearly as impactful as some make it out to be. I cannot possibly know this to be fact and don't mean to imply that I believe it as fact.

I also will mention the fact that the human brain can play tricks on us. Sometimes things seem to be something when they are not. For example, the human brain tends to notice things that are "abnormal". When many college students were asked what they believe the percentage of people who smoked that attended the university, most people vastly overrated the number. The thought is/was that the human brain noticed the abnormal actions of someone smoking rather than the more frequent people NOT smoking.


A question for you though out of curiosity, if you know or know how to find out. Did Claude score goals at a better rate in the last minutes, or overtime, than he did at other times during games over his career?

 

[emoney_33]

You can bet your bottom dollar any coach doesn't need a calculator to know the ball goes to Bird in the clutch (or maybe Garnett now).

But isn't this because their chance of something good happening is higher with the ball in the hands of Bird (or Garnett now), no matter the situation? Not just in the clutch, but in the clutch you can't make it up so with one chance you want your best player to have the ball.

 

[emoney_33]

Great discussions, off to bed...


1+1 = 1

 

[makewayhomer]

You can't have it both ways. If you're going to use statistics, you have to use statistical methods, otherwise your results are garbage. You're the one that stated you were going to teach posters something, not me.

Baseball lends itself to statistics much more than other sports do. Way back in the old days, managers and players kept "books" on players situations and so forth which constituted a crude form of statistical analysis. Sorry to be sarcastic, but I'm avoiding posting for my research methods class right now, which struck me as funny.

Because this latest method downplays clutch etc. etc. certainly doesn't mean it discounts it. Statistics is dry and difficult for me to understand and I'm averaging over 99%. You can bet your bottom dollar any coach doesn't need a calculator to know the ball goes to Bird in the clutch (or maybe Garnett now).

It's just another statistical method, there's nothing new under the sun and no method to replace a smart, experienced GM.

I'm not sure what you're saying here. my point is that performance analysis in all sports has improved lately. this argument isn't just about "clutch" it's about the ability to analyze information, come to conclusions, and improve your team.

another example, so we can get off of "clutch".

the Pats consistently go on 4th down more than any other team. this isn't by accident or b/c they have been a particularly awesome running team - it's b/c studies were done to support the idea that value could be gained by "gambling" more on 4th down, and the Pats listened.

 

[RayClay]

But isn't this because their chance of something good happening is higher with the ball in the hands of Bird (or Garnett now), no matter the situation? Not just in the clutch, but in the clutch you can't make it up so with one chance you want your best player to have the ball.

I'll substitute Dennis Johnson, then. Or Rico Petrocelli, Carl Yastrzemski. Tom Brady. Willie McGinest, Ted Bruschi, Rodney Harrison (in their prime).

Statistics record what happens, humans interpret statistics.

Dennis Johnson, widely regarded as one of the great clutch players in NBA history, averaged 14.1 points per game in the regular season during his 14-year NBA career but in 37 games in the NBA Finals he averaged 18.3 points per game.

Among the 65 players in NBA history who played at least 800 games in the regular season and 15 games in the finals, only two had a larger increase in their finals ppg average over their regular-season mark than Johnson's 4.2: Hakeem Olajuwon, who was plus 5.7 (21.8 to 27.5), and James Worthy, who was plus 4.6 (17.6 to 22.2).

Statistics can be misinterpreted. That doesn't mean the statistics are wrong.

 

[TomBrady'sGoat]

Mike Torrez was the best option the manager had that day...bullpen management was never in question.

You're limiting yourself with assumptions. Statistical analysis could lead the team to build a better bullpen. It could help a team avoid situations where their best guys aren't unavailable. It could effect the way each pitcher is utilized to make them more effective. And it might tell you that, in spite of what you believe, he is not in fact the best available.

If it can help why not be open to it? It isn't an either-or proposition.

 

[Fencer]

As long as we're going to look down noses, or whatever, I've taught statistics at one university and game theory at another. All credentials aside:

1. Judging a player based on the numbers in a continuous-action team sport is in most cases MUCH harder than judging him by the numbers in baseball. The main reason is that you don't have controlled, repeatable experiments.

2. FT shooting in basketball is the closest thing to an exception I can think of. Placekicking in football is a very partial exception at best, because the sample size is so small and conditions are more variable (notably, distance, but also the interplay between distance, weather, and field conditions).

3. I don't know to what extent we can augment the numbers by an "eyeball test." Does Ortiz look much different when batting in a clutch situation than when in any other? No. Good for him. But does A-Rod shake visibly in the clutch either? Uh, no ...

4. Players seem to think that at least in some cases there's a real difference between clutch and other situations. Vinateri says you need to control your heartbeat to have your aim be good. Brady says he gets more focused in do-or-die situations. And in basketball, you tend to have some heart rate/adrenaline issues and also the factor that in clutch situations you're probably tired at the end of games (or else not warmed up/in the flow), in a way that is not wholly consistent from game to game.

5. Coaches think so too. They try to create artificial pressure on kickers and FT shooters in practice, so as to at least partially replicate game situations.

6. The Pats certainly had a lot of good game luck early in their run. Several plays in the Raiders game could have gone either way. One or both of the ST TDs vs. the Steelers might not have happened. They might not have won all those coin flips to go first in OT.

7. The Pats certainly had lousy injury luck for a few years there. And I agree with the consensus here that they've had some bad refereeing luck recently as well.

8. BB plans for injury problems, perhaps more aggressively than other teams. That's like buying options to lessen the impact of possible unfavorable events. But despite his reputation for doing that, I don't know how much of his success with backups is personnel and how much of it is coaching and scheming. Jarvis Green is a good personnel move, as is Russ Hochstein. But Earthwind Moreland and Hank Poteat? That seems more like a matter of coaching.

9. Based on the recent Vinateri threads, I'd say a significant fraction of the people here acutely recall his imperfections and misses.

10. One place where I differ from many is that I think some football players get BETTER in the clutch. It makes sense to hypothesize as follows, and I think the evidence supports it:

They can't give their best moves, best concentration, best bursts every play. But they have an extra gear they call on when needed. That's why we see so many sacks pushing teams just out of field goal range. I think McGinest was a prime example of somebody who make some outstanding plays, but who couldn't really deliver at that level all game or all season. I guess strongly his clutchness was more than chance.

 

[fair catch fryar]

I haven't studied these things in great depth, and I'm not going to sit here and say that there are NO people ever who didn't have something special to rise up in certain situations. The human brain is an insanely complex thing and I won't for one second pretend to know everything. However, if a player does better in a certain situation, then there has to be something that can be put down on paper to show that he was better in that situation than in others.

However I am only suggesting that the "clutch" thing is exaggerated and not nearly as impactful as some make it out to be. I cannot possibly know this to be fact and don't mean to imply that I believe it as fact.

I also will mention the fact that the human brain can play tricks on us. Sometimes things seem to be something when they are not. For example, the human brain tends to notice things that are "abnormal". When many college students were asked what they believe the percentage of people who smoked that attended the university, most people vastly overrated the number. The thought is/was that the human brain noticed the abnormal actions of someone smoking rather than the more frequent people NOT smoking.


A question for you though out of curiosity, if you know or know how to find out. Did Claude score goals at a better rate in the last minutes, or overtime, than he did at other times during games over his career?

If I didn't have to get up at 5:00 AM, I might actually try to figure it out, but unfortunately, I'll only agree to disagree with you. I simply believe an a person's innate ability to be "better" at certain times than others and respect your opinion, as I do believe it has some credibility and use to the "bean-counters" when determining a player's value. Like Ray Clay said, a good GM or coach always knows which player to go to at "crunch time" and I'm guessing they don't use this theory to calculate which player has the best statistical chance to hit the last second 3-pointer based on past attempts and success rates in those particular situations. G-nite!

 

[emoney_33]

I'll substitute Dennis Johnson, then. Or Rico Petrocelli, Carl Yastrzemski. Tom Brady. Willie McGinest, Ted Bruschi, Rodney Harrison (in their prime).

Statistics record what happens, humans interpret statistics.

Dennis Johnson, widely regarded as one of the great clutch players in NBA history, averaged 14.1 points per game in the regular season during his 14-year NBA career but in 37 games in the NBA Finals he averaged 18.3 points per game.

Among the 65 players in NBA history who played at least 800 games in the regular season and 15 games in the finals, only two had a larger increase in their finals ppg average over their regular-season mark than Johnson's 4.2: Hakeem Olajuwon, who was plus 5.7 (21.8 to 27.5), and James Worthy, who was plus 4.6 (17.6 to 22.2).

Statistics can be misinterpreted. That doesn't mean the statistics are wrong.

I love this post. You are correct, statistics never lie, but are rarely interpretted correctly and never perfectly. So nowww I'm going to bed .

 

[TomBrady'sGoat]

I'll substitute Dennis Johnson, then. Or Rico Petrocelli, Carl Yastrzemski. Tom Brady. Willie McGinest, Ted Bruschi, Rodney Harrison (in their prime).

Every one of them are great players.

I agree, in clutch situations you are best served by having your best players decide the outcome. Great players are more likely to be successful in all situations, clutch included.