A win clinches you a spot in the playoffs...

[RickD]

If you throw stats out the window then your point (not to sound rude which I promise I am not intending this to be) becomes moot. Reason being without the stats who is to say the manager is not using his best AVAILABLE releiver in that situation. Maybe a guy is tired, sore, etc and someone else is ready and he is the best that day available! See the point? Any manager has to look beyond stats to see if a guy is ready that day. THis is why managers will give guys a day off, they notice a slight variance in swings, favoring legs, arms, etc and decide the guy needs a rest.

[HoustonGM]

If you throw stats out the window then your point (not to sound rude which I promise I am not intending this to be) becomes moot. Reason being without the stats who is to say the manager is not using his best AVAILABLE releiver in that situation. Maybe a guy is tired, sore, etc and someone else is ready and he is the best that day available! See the point? Any manager has to look beyond stats to see if a guy is ready that day. THis is why managers will give guys a day off, they notice a slight variance in swings, favoring legs, arms, etc and decide the guy needs a rest.

I understand that. But he's saying that I'm somehow using the stats that are only due to the way he's used to justify why he should be used differently, which isn't true, because the stats are due mostly to the player's ability, not how he's used. I'm using the player's ability to justify why he should be used differently.

I'm in no way saying that managers shouldn't give their players rest or anything. That'd be crazy. I'm just saying that the best relievers should be used in the highest leverage situations, and vice versa. I'm not understanding why that is such a difficult concept to grasp.

[FRENCHREDSOX]

I understand that. But he's saying that I'm somehow using the stats that are only due to the way he's used to justify why he should be used differently, which isn't true, because the stats are due mostly to the player's ability, not how he's used. I'm using the player's ability to justify why he should be used differently.

I'm in no way saying that managers shouldn't give their players rest or anything. That'd be crazy. I'm just saying that the best relievers should be used in the highest leverage situations, and vice versa. I'm not understanding why that is such a difficult concept to grasp.

No,you are having the difficulty to grasp that "leverage" analysis in itself is a totally subjective theorem.

ltt is simply saying,that a manager's choice is "mathematically" the only true factual statistic,as Manager chooses pitcher at the PRECISE moment,the pitcher pitches & that produces a single,unique OUTCOME.This OUTCOME leads to the stats which are the only true mathematical stats.1 Result ONLY.

Leverage analysis is "what if" after the actual event WHICH in itself is not a fact & thus is a hypothetical supposition.

Also leverage analysis does not & cannot compare exact situations,because you cannot (as of today at least in this universe) have 2 pitchers throw to the same batter at the same instant & obtain a result,thus it non mathematical as it cannot be proved/disproved.This is what llt "means" when he says

.

It is precisely the way Torre uses Rivera that produces the stats that are then used to insist he should use him otherwise.

.THAT choice is a free will act which provokes a mathematical factual & unique result.

Lastly,different situations lead to different results.A player may have a high K,low BB & thus a low ERA & another be a "worse" leveraged pitcher but they were pitching at different times with different stress factors.The former may be a middle reliever who is a "great" pitcher but may not be able to handle to 8th or 9th innings.Leverage does not show this.

[WATERY]

It's like punching a brick wall, guys.

[HoustonGM]

Okay, fine, I'm completely crazy for thinking that the best pitchers should be used in the most important situations, and the worst pitchers should be used in unimportant situations.

[RickD]

Okay, fine, I'm completely crazy for thinking that the best pitchers should be used in the most important situations, and the worst pitchers should be used in unimportant situations.

HGM you are not wrong in thinking that....I think that as well....however you HAVE to factor in conditioning, soreness, and a host of other factors beyond just stats. Your stats are great when taken in context but not EVERY manager is wrong. If they mostly all do this there is probably good resons beyond stats.

[HoustonGM]

HGM you are not wrong in thinking that....I think that as well....however you HAVE to factor in conditioning, soreness, and a host of other factors beyond just stats. Your stats are great when taken in context but not EVERY manager is wrong. If they mostly all do this there is probably good resons beyond stats.

I've never claimed to not factor in conditioning, soreness, etc. Obviously, if a pitcher is unable to pitch, the manager can't and shouldn't use him. That explains single incidents. It does not explain that managers, overall, over full seasons, don't practice very good leveraging of relievers. They'd rather save their closer (usually the best reliever) for a save situation, than use him when it matters most. Just because most managers do it does not mean that it's the most beneficial way of doing things.

[FRENCHREDSOX]

HGM you are not wrong in thinking that....I think that as well....however you HAVE to factor in conditioning, soreness, and a host of other factors beyond just stats. Your stats are great when taken in context but not EVERY manager is wrong. If they mostly all do this there is probably good resons beyond stats.

The problem is that "leverage" as pointed out above is a SUBJECTIVE analysis,sheesh,it is that simple.It is a "what if" hypothesis,of INCOMPARABLE players because their stats that they have garnered through out the season or career ARE simply INDIVIDUAL & UNIQUE situations & thus are each every time DIFFERENT, full stop.

It cannot be "proved" that X would have done better than Y because X did NOT pitch,whereas Y did.It is a SUPPOSITION of a POSSIBLE event.

Whereas Manager A's decision to pitch Y LEADS directly to an outcome,results & thus OBJECTIVE real stats.The only thing that can be garnered is that Y was either successful or unsuccessful & the next time Mr A.,will have to take his last action into consideration,IF a similar set of circumstances come into play.

The original question remains why did Torre do what he did,the posters,IMO,have shown that he did this to "test" certain players under a "stressful" situation in order to garner information for the PLAYOFFS not just for that 1 win/loss & all THIS information is used by himself,his Bullpen crew & the GM staff to obtain the "best" post season group.(in THEIR opinion.)

One has to remember that in Post season,most teams actually use LESS players than in regular season (rightly or wrongly) & go with veterans rather than "rookies".However,teams need to know if X,Y or Z is capable to handle the demands IF ASKED.

[HoustonGM]

And again, I said the reason I posted this was really beyond one situation. It's a larger problem that happens year round. While there are reasons to excuse Torre for this specific instance, it's not as if Torre properly leverages his relievers, on the whole. Barely any manager does.

And no, leverage is not subjective. A tie game in the 7th inning is higher leverage than a 6 run lead in the 9th.

And players aren't incomparable just because they've pitched in different situations. There are obvious better pitchers and obvious worse pitchers. For example, Mariano Rivera is obviously better than Jeff Karstens. I don't care what situation it is. Mariano Rivera is a better pitcher than Jeff Karstens. If you're going to argue otherwise, you better be prepared to give some good facts to back up that claim.

[ltt]

And again, I said the reason I posted this was really beyond one situation. It's a larger problem that happens year round. While there are reasons to excuse Torre for this specific instance, it's not as if Torre properly leverages his relievers, on the whole. Barely any manager does.

but the example you used is demonstrative of a universal truth - there are reasons for Torre's decisions in EVERY instance. Your attempt to distill a statistical truth from human decisions is, imo, misguided, because even when a manager doesn't use what you would call his "best" pitcher in what you would call a "high leverage" situation, he does so to maximize his entire roster's chance for success over 162 games and the playoffs. Your conclusion that their actions in general do not succeed in doing this is not an empirically provable position, as you keep claiming, but rather your opinion.

And no, leverage is not subjective. A tie game in the 7th inning is higher leverage than a 6 run lead in the 9th.

yes it is. no two tie games in the 7th are the same. no two 6-run ninth-inning leads are the same. no two "best" relievers are the same. you're creating a class from object that are not alike.

And players aren't incomparable just because they've pitched in different situations. There are obvious better pitchers and obvious worse pitchers. For example, Mariano Rivera is obviously better than Jeff Karstens. I don't care what situation it is. Mariano Rivera is a better pitcher than Jeff Karstens. If you're going to argue otherwise, you better be prepared to give some good facts to back up that claim.

that's not what I am, or anyone is, arguing.

What I'm arguing is that you can't use the reductive argument you've chosen to analyze pitcher use, because in every single case there are myriad considerations a manager must evaluate besides the one you are insisting be paramount.

[HoustonGM]

but the example you used is demonstrative of a universal truth - there are reasons for Torre's decisions in EVERY instance. Your attempt to distill a statistical truth from human decisions is, imo, misguided, because even when a manager doesn't use what you would call his "best" pitcher in what you would call a "high leverage" situation, he does so to maximize his entire roster's chance for success over 162 games and the playoffs.

Does he? Managers can't make wrong decisions?

Multiple managers have stated that they've chosen to go with a lesser reliever in a more important situation, such as a tie game in the 9th in order to save their better reliever IN CASE they get their lead. That is NOT in the best interest of winning.

yes it is. no two tie games in the 7th are the same. no two 6-run ninth-inning leads are the same. no two "best" relievers are the same. you're creating a class from object that are not alike.

A six run lead is lower leverage than a tie game. Leverage is the importance of a given situation. With a six run lead in the 9th, the probability of your team winning is extremely high. Thus, low leverage. A tie game in the 7th, the probability of winning is much less than that of a six run lead in the ninth. It is higher leverage.

When is a tie game ever a less dire situation than a six run lead? There are times in the game that are more important than others. When you have a huge lead, the situations are less important than when you're in a tie game. I don't see how you can even dispute that.

that's not what I am, or anyone is, arguing.

What I'm arguing is that you can't use the reductive argument you've chosen to analyze pitcher use, because in every single case there are myriad considerations a manager must evaluate besides the one you are insisting be paramount.

In every single game, it is always best to maximize your chances of winning. Ignoring the obvious consideration of injury and rest and other things that affect availability, the better reliever should nearly always be chosen over the lesser reliever in a highly important situation. Okay, fine, testing for the playoffs, whatever, that doesn't happen in April, and that is what I'm talking about. Managers make these decisions to go with lesser relievers in more important situations ALL SEASON LONG, and that is not in the best interest of winning.

While every situation is different, I am speaking in OVERALL terms.

Stop acting as if managers are infallible beings that can never make incorrect decisions.

[FRENCHREDSOX]

but the example you used is demonstrative of a universal truth - there are reasons for Torre's decisions in EVERY instance. Your attempt to distill a statistical truth from human decisions is, imo, misguided, because even when a manager doesn't use what you would call his "best" pitcher in what you would call a "high leverage" situation, he does so to maximize his entire roster's chance for success over 162 games and the playoffs. Your conclusion that their actions in general do not succeed in doing this is not an empirically provable position, as you keep claiming, but rather your opinion.



yes it is. no two tie games in the 7th are the same. no two 6-run ninth-inning leads are the same. no two "best" relievers are the same. you're creating a class from object that are not alike.



that's not what I am, or anyone is, arguing.

What I'm arguing is that you can't use the reductive argument you've chosen to analyze pitcher use, because in every single case there are myriad considerations a manager must evaluate besides the one you are insisting be paramount.

So true but when one does not even compound the differentiation between objectivity nor subjectivity then a classical "circular argument" is induced.But aparrently some cannot even see this............

End of story.

[ltt]

Does he? Managers can't make wrong decisions?

not in the sense you mean - i.e., scientifically. baseball isn't euclidean geometry, you can't "prove" your position the same way I can "prove" the interior angles of any triangle add up to 180 degrees.

Multiple managers have stated that they've chosen to go with a lesser reliever in a more important situation, such as a tie game in the 9th in order to save their better reliever IN CASE they get their lead. That is NOT in the best interest of winning.

YOU say it's not in the best interests of winning. They disagree.

A six run lead is lower leverage than a tie game. Leverage is the importance of a given situation.

and "importance" is a subjective valuation.

With a six run lead in the 9th, the probability of your team winning is extremely high. Thus, low leverage. A tie game in the 7th, the probability of winning is much less than that of a six run lead in the ninth. It is higher leverage.

look, a couple of you guys have to get over the notion that the only way anyone could possible disagree with you is because they don't understand you. Let me make this clear - I understand what you mean by "leverage," and I think it's an interesting and probably useful concept. But I also understand that actual managers are faced with much more complicated decisions than simply calculating win probability in a given situation when dealing with a team comprised of 25-40 human beings. In my opinion, that means your position is just one way of thinking about baseball, not the right way.

When is a tie game ever a less dire situation than a six run lead?

Never. That has zero bearing, however, on how a manager should use his pitching staff.

There are times in the game that are more important than others. When you have a huge lead, the situations are less important than when you're in a tie game. I don't see how you can even dispute that.

I'm not disputing it, I'm disputing whether that means every time a manager doesn't use the pitcher you think he should he's doing the wrong thing.

In every single game, it is always best to maximize your chances of winning.

this is false...

Ignoring the obvious consideration of injury and rest and other things that affect availability,

...because this can NEVER be ignored

the better reliever should nearly always be chosen over the lesser reliever in a highly important situation.

assumes facts not in evidence

Okay, fine, testing for the playoffs, whatever, that doesn't happen in April, and that is what I'm talking about.

of course it does. any manager worth his salt is CONSTANTLY evaluating his squad. Following your suggestion, Torre only uses half his bullpen in blowouts. Rivera gets injured tomorrow and whathisname get injured tomorrow. Under your method, Torre has ZERO game-situation information to help him manage the relievers he has left.

Managers make these decisions to go with lesser relievers in more important situations ALL SEASON LONG, and that is not in the best interest of winning.

you can not EVER demonstrate this. As FRS has pointed out over and over again, you are treating a probability based on statistics as though it were the predictable and guaranteed outcome of a change in behaviour.

While every situation is different, I am speaking in OVERALL terms.

yes, but your overall terms are completely inapplicable to individuals cases, and therefore wholly suspect.

Stop acting as if managers are infallible beings that can never make incorrect decisions.

stop acting as if there is a scientifically provable theory of baseball.

[HoustonGM]

look, a couple of you guys have to get over the notion that the only way anyone could possible disagree with you is because they don't understand you. Let me make this clear - I understand what you mean by "leverage," and I think it's an interesting and probably useful concept. But I also understand that actual managers are faced with much more complicated decisions than simply calculating win probability in a given situation when dealing with a team comprised of 25-40 human beings. In my opinion, that means your position is just one way of thinking about baseball, not the right way.

You believe that there are 6-run leads in the 9th that are actually higher leverage situations than tie games in the 7th or 8th?

Never. That has zero bearing, however, on how a manager should use his pitching staff.

What? The score of the game should have no bearing on how a manager should use his pitching staff? Wow.

That is just completely baffling. I don't even know what to say to that. The score of the game most definitely matters. In a tie game in the bottom of the 9th, you absolutely cannot allow a run, or you lose. With a 6 run lead in the top of the 9th, you have much more leeway. How in heck's name should that not have an effect on how the manager uses his pitching staff?

I'm not disputing it, I'm disputing whether that means every time a manager doesn't use the pitcher you think he should he's doing the wrong thing.

I already said that it isn't ALWAYS the wrong decision - availibility of the pitchers matters. As you've brought up, testing for playoffs, etc. I've said that ON THE WHOLE a better pitcher should be used in a more important situation.

Game 7 of the World Series. Would you rather have Josh Beckett on the mound, or Jason Marquis?

this is false...

Oh. I guess I didn't know that the goal of a baseball game wasn't to win.

...because this can NEVER be ignored

I never said that you should ignore player's injuries or rest and always pitch them. I'm saying that if you have two pitchers, both equally rested and uninjured (thus, ignoring rest/injury), the better one should be used in the more important situations.

of course it does. any manager worth his salt is CONSTANTLY evaluating his squad. Following your suggestion, Torre only uses half his bullpen in blowouts. Rivera gets injured tomorrow and whathisname get injured tomorrow. Under your method, Torre has ZERO game-situation information to help him manage the relievers he has left.

Torre has no information on the talent of his pitchers?

There are ways to evaluate your squad without lessening your chances at winning ball games.



Jeez. I never would have thought that the idea of using better players in more important situations would ever draw such criticism. I guess we should just completely ignore the talent of players, and just play whoever. You have a must-win game. There's no reason to go with your best pitcher and best possible lineup, right? Apparantly.

The best way to win is to play the best possible combination of players, and maximize the chances of scoring and preventing runs in important situations by using the best possible players for the given situation. Why is that such a crazy idea?

[WATERY]

Wow. Just wow. Carry on.