PDA

View Full Version : Honest Question for you tRA fans??



papipapsmanny
01-06-2009, 06:42 PM
Alright so you say the stat regresses everything in order to show how each pitcher would pitch if they had the same defense.

Some of you say pitchers on the yankees would have posted better numbers if they were on a team like the Red Sox

well here is my question

Beckett is ranked 12th in tRA for the 2008 season, Lester is Ranked 28th, and Daisuke Matsuzaka is ranked 70th.

To the Casual fan the "old fashion" stats would suggest the either Matsuzaka or Lester had the best year of the three and that clearly Beckett was the worst of the 3.

Yet Beckett is suggested to be the best by tRA when everything is regressed according to defense. Well what is there to regress? They all had the same defense, how could you say the guy with the better numbers did worse than the guy with the worse numbers when they were playing with the same defense. There was nothing really to regress to because they all had the same defense behind them.

This is an honest question, not trying to start anything. (even though i still dont reallly like using the stats

blenderboy5
01-06-2009, 06:46 PM
I barely understand tRA and think people use too many stats in the game anyway.

Or judge pitchers based on stupid ones like wins.

Seamhead
01-06-2009, 06:49 PM
It doesn't take defense into account at all. That's why. Even though pitchers pitch behind the same defense, it doesn't mean that a defense will play the same for every pitcher. Or else you'd see pitchers on one team with the same BABIPs.

All tRA does is take into account everything that we know a pitcher can control. Defense is not one of those.

Ph1lly Diehard
01-06-2009, 06:49 PM
I know Seamhead and Gigants know alot about this stuff, specially seamhead who is really really into it and knows alot of his stuff (no offense Gigants)

EDIT: And Seamhead answers before I say to ask him :pity:

papipapsmanny
01-06-2009, 06:54 PM
It doesn't take defense into account at all. That's why. Even though pitchers pitch behind the same defense, it doesn't mean that a defense will play the same for every pitcher. Or else you'd see pitchers on one team with the same BABIPs.

All tRA does is take into account everything that we know a pitcher can control. Defense is not one of those.

see thats my whole thing how do you regress stats to defense than? no matter what the same defensive plays wont be made for everyone

quiksilver2491
01-06-2009, 06:57 PM
Why is it so hard to believe that Beckett pitched better then Lester and Matsuzaka last year? Matsuzaka's ERA made him appear way better then he really was last year and Lester also overperformed slightly, not nearly as much as Matsuzaka though.

Beckett on the other hand caught a ton of bad breaks and his ERA made him appear much worse then he really pitched last year, he was actually quite solid last year. Don't really understand what you are trying to get at here.

Twitchy
01-06-2009, 06:59 PM
The biggest problem with your comparison is that tRA+ doesn't take innings pitched into account.

Beckett threw 174 innings, while Lester threw 210. So despite the fact Beckett's tRA+ is higher by 7, Lester had the better season because he did it in more innings.

Just because someone has a higher tRA+ doesn't mean they're better. You have too look at sample size.

quiksilver2491
01-06-2009, 07:03 PM
The biggest problem with your comparison is that tRA+ doesn't take innings pitched into account.

Beckett threw 174 innings, while Lester threw 210. So despite the fact Beckett's tRA+ is higher by 7, Lester had the better season because he did it in more innings.

Just because someone has a higher tRA+ doesn't mean they're better. You have too look at sample size.

Actually even if you were including IP Beckett was still better last year.

2008 PRAA

Beckett: 26.3
Lester: 22.4

Seamhead
01-06-2009, 07:05 PM
The biggest problem with your comparison is that tRA+ doesn't take innings pitched into account.

Beckett threw 174 innings, while Lester threw 210. So despite the fact Beckett's tRA+ is higher by 7, Lester had the better season because he did it in more innings.

Just because someone has a higher tRA+ doesn't mean they're better. You have too look at sample size.

He was referring to effectiveness, more so than value.

Seamhead
01-06-2009, 07:06 PM
see thats my whole thing how do you regress stats to defense than?

I really don't understand what you're trying to say.



no matter what the same defensive plays wont be made for everyone

Yes, that's the point. Defense, apart from his own, which comes into play very few significant times, is something the pitcher can't control -- therefore, why should he be rewarded or hurt for it?

papipapsmanny
01-06-2009, 07:10 PM
well from my understanding tRA is a stat used to regress each pitchers's stats assuming that the exact same defensive plays are made for each pitcher

i am asking how do you do that?

quiksilver2491
01-06-2009, 07:10 PM
well from my understanding tRA is a stat used to regress each pitchers's stats assuming that the exact same defensive plays are made for each pitcher

i am asking how do you do that?

Defense isn't part of the equation in the first place so it isn't regressed, just never included as tRA is based off of BB data unlike stats like ERA which is based on runs.

Seamhead
01-06-2009, 07:17 PM
well from my understanding tRA is a stat used to regress each pitchers's stats assuming that the exact same defensive plays are made for each pitcher

i am asking how do you do that?

It's not taken into account at all. Not part of the equation. Funny how you don't like tRA, yet you don't know much about it.

There's an article at statcorner.com that explains it all for you.

papipapsmanny
01-06-2009, 07:18 PM
so why do all of you always say its a control for Defense?

im clearly not understanding something here

papipapsmanny
01-06-2009, 07:20 PM
It's not taken into account at all. Not part of the equation. Funny how you don't like tRA, yet you don't know much about it.

There's an article at statcorner.com that explains it all for you.

no the way you and others have explained it to me is that is regresses the pitchers stats assuming each pitcher has the same defense behind them

Seamhead
01-06-2009, 07:21 PM
so why do all of you always say its a conrol for Defense?

im clearly not understanding something here

It's a context-neutral metric, meaning it places every pitcher in the same context. That context includes an average defense, which is simply the average amount that a batted ball fell for a hit.

quiksilver2491
01-06-2009, 07:23 PM
no the way you and others have explained it to me is that is regresses the pitchers stats assuming each pitcher has the same defense behind them

I think what you are trying to imply is that tRA is context neutral, meaning that it wouldn't matter if you had a great or horrible defense, however that doesn't mean it has been regressed from the equation, it is just never part of it.

dcannon456
01-06-2009, 07:27 PM
so why do all of you always say its a control for Defense?

im clearly not understanding something here


from my understanding, which isnt all that well to be fair. it eliminates defense from the equation. its 100% soley about what the pitcher actually impacts. and i can see why people might confuse that for having equal defenses.

papipapsmanny
01-06-2009, 07:38 PM
^^ but then comes the question what about pitchers who pitch to contact? are they punished by this stat?

and it still comes with many flaws, their are different defensive shifts (DP depth, In, or shaded left or right for certain players) and yes a pitcher can slightly influence where the ball is hit by either throwing in or away.

And So what i am getting now tRA is a stats based soley on Strikouts and Walks? is that it?

ComeBackGeorge
01-06-2009, 07:42 PM
Pitching to contact is not really good though unless they are an extreme groundball pitcher. Pitchers have very little to no control of the balls put into play. Except of course GB/FB ratio. Does this stat account for that ratio? And am I correct in saying it mostly neutralizes BABIP?

quiksilver2491
01-06-2009, 07:57 PM
^^ but then comes the question what about pitchers who pitch to contact? are they punished by this stat?

and it still comes with many flaws, their are different defensive shifts (DP depth, In, or shaded left or right for certain players) and yes a pitcher can slightly influence where the ball is hit by either throwing in or away.

And So what i am getting now tRA is a stats based soley on Strikouts and Walks? is that it?

Basically it's based off of BB data and things a pitcher can control (yes that includes walks and strikeouts). Instead of asking so many questions you should take 5 minutes out of your life and read the tRA Primer, it goes into full detail about what tRA entails and exactly how it is calculated.

http://www.statcorner.com/tRAabout.html

quiksilver2491
01-06-2009, 08:00 PM
Pitching to contact is not really good though unless they are an extreme groundball pitcher. Pitchers have very little to no control of the balls put into play. Except of course GB/FB ratio. Does this stat account for that ratio? And am I correct in saying it mostly neutralizes BABIP?

Yes, tRA accounts for both GB% and FB%. Also it is neutral contexted so yes it does pretty much neutralize BABIP so all pitchers are placed at the same level.

x_notorious
01-06-2009, 08:06 PM
My question is..why is John Lackey's tRA+ 94? With that statistic, he slots as a below average number 3 starter according to this:

http://www.statcorner.com/blog/2008/10/rotation-slots-by-tra.html

ugafan
01-06-2009, 08:10 PM
My question is..why is John Lackey's tRA+ 94? With that statistic, he slots as a below average number 3 starter according to this:

http://www.statcorner.com/blog/2008/10/rotation-slots-by-tra.html

What makes you think otherwise?

That's only based on last year, he didn't pitch as well last year as he did the year before(and other years).

poodski
01-06-2009, 08:22 PM
One reason why Dice K and Lesters ERA could be better while Becketts would be worse is LOB%

Dice K was at 80% while Beckett was at 71%. Not a huge reason but a reason none the less.

papipapsmanny
01-06-2009, 08:27 PM
here is the thing, the averages the stats use change each and every year so i dont understand how it shows the true talent of a pitcher, when it seems like this stats are more based on what the batters do and not what the pitchers do. Which means the level of talent the pitchers play is huge factor of the tRA, yet it is literally not accounted for by the stat.

Also i dont like how they use the mean to regress the stats, because the mean is largely affected by outliers, if they wanted to make tRA (which im not convinced of yet) more accurate they should regress everything according to the median of each thing, as the median isnt really affected by such outliers (like extreme highs or lows).

Even if the stat were deemed to be most accurate, i would still take Matsuzaka over Beckett for the 2009 season based on 2008 performances

x_notorious
01-06-2009, 08:28 PM
What makes you think otherwise?

That's only based on last year, he didn't pitch as well last year as he did the year before(and other years).

His 'normal' statistics seem OK to me. Not that bad actually considering he came back from injury. But to be slotted as a number 4 starter?

Riddle me this then:


#1 STARTERS: 130 tRA+
#2 STARTERS: 112 tRA+
#3 STARTERS: 100 tRA+
#4 STARTERS: 91 tRA+
#5 STARTERS: 76 tRA+

By using that chart, you are telling me that only 9 starters in the league were considered number 1 pitchers? Including recent Giants' f/a pick up, Randy Johnson? (not that I'm complaining)

Going back to Lackey now..John Lackey has a lower ERA, WHIP, higher ERA+ and both have identical BAA's. Yet the difference between their tRA+ is 36 points?

Explain that.

Gigantes4Life
01-06-2009, 08:39 PM
no the way you and others have explained it to me is that is regresses the pitchers stats assuming each pitcher has the same defense behind them

It doesn't take defense into account.

I guess they do have the same defense if you may, since the final number is an ERA equivalent (-.35 or +.35, I forget) with an average defense.


My question is..why is John Lackey's tRA+ 94? With that statistic, he slots as a below average number 3 starter according to this:

http://www.statcorner.com/blog/2008/10/rotation-slots-by-tra.html

Small-sample size. He was hurt, so it may have lowered his effectiveness.

Gigantes4Life
01-06-2009, 08:43 PM
His 'normal' statistics OK to me. Not that bad actually considering he came back from injury. But to be slotted as a number 4 starter?

Riddle me this then:



By using that chart, you are telling me that only 9 starters in the league were considered number 1 pitchers? Including recent Giants' f/a pick up, Randy Johnson?

Going back to Lackey now..John Lackey has a lower ERA, WHIP, higher ERA+ and both have identical BAA's. Yet the difference between their tRA+ is 36 points?

Explain that.

http://www.statcorner.com/tRAabout.html

Look at the run values (the 2nd box).

HR cost the most runs obviously, and Lackey gave up 26 of them in only 163 IP.

Randy gave up 24 in 186 IP. Then, Arizona is a hitters-park, and LA is the opposite. So Randy's HR run values are lowered, and Lackey's are raised.

That's not a very accurate explanation, but it should make some sense.

Just look at the run values and then compare Randy's and Lackey's GB%, LD%, FB%, etc.

Freel for prez
01-06-2009, 08:45 PM
Tell me the TRA for Edinson Volquez if he had a good D behind him

25YrsNcounting
01-06-2009, 08:48 PM
I go by wins and era...that other stuff is for nerds.

SouthSideIrish
01-06-2009, 08:56 PM
Alright so you say the stat regresses everything in order to show how each pitcher would pitch if they had the same defense.

Some of you say pitchers on the yankees would have posted better numbers if they were on a team like the Red Sox

well here is my question

Beckett is ranked 12th in tRA for the 2008 season, Lester is Ranked 28th, and Daisuke Matsuzaka is ranked 70th.

To the Casual fan the "old fashion" stats would suggest the either Matsuzaka or Lester had the best year of the three and that clearly Beckett was the worst of the 3.

Yet Beckett is suggested to be the best by tRA when everything is regressed according to defense. Well what is there to regress? They all had the same defense, how could you say the guy with the better numbers did worse than the guy with the worse numbers when they were playing with the same defense. There was nothing really to regress to because they all had the same defense behind them.

This is an honest question, not trying to start anything. (even though i still dont reallly like using the stats

Most stats are for full retards. If you think a player is good, roll with it. You may be wrong, but **** anyone who cares. None of us are real GMs and nothing we say really matters. But people keep throwing them out there to "prove" their baseball knowledge even though it's nothing more than utter reliance on a quantitative analysis that someone else developed. It's mindless and lame.

Gigantes4Life
01-06-2009, 09:04 PM
Tell me the TRA for Edinson Volquez if he had a good D behind him

tRA can't be adjusted with a good D. There's only one tRA for a pitcher.

Volquez was a 3.78, 121 tRA+, which is of course ace-caliber.


I go by wins and era...that other stuff is for nerds.

Definitely. Wins FTW!


Most stats are for full retards. If you think a player is good, roll with it. You may be wrong, but **** anyone who cares. None of us are real GMs and nothing we say really matters. But people keep throwing them out there to "prove" their baseball knowledge even though it's nothing more than utter reliance on a quantitative analysis that someone else developed. It's mindless and lame.

You're so right.

Seamhead
01-06-2009, 09:41 PM
I go by wins and era...that other stuff is for nerds.

Hell yeah. Only badasses use wins and ERA. If you nail chicks and do drugs, you use ERA and wins.

Seamhead
01-06-2009, 09:50 PM
Most stats are for full retards.

Yeah. Like Tom Tango, Dan Fox, Billy Beane, MGL, Paul Depodesta, and Matt Antonelli, who all have pretty sweet front office gigs.

Not believing in stats, on the other hand, is for geniuses such as Billy Bavasi, Woody Paige, and Bill Plaschke.


If you think a player is good, roll with it. You may be wrong, but **** anyone who cares.[/quote]

Hell yeah. Be a ****ing badass. I may be contradicted by facts, but I don't give a ****!

(Honestly, what the hell are you talking about?)


None of us are real GMs and nothing we say really matters.

This is pretty irrelevant to everything and anything. Hell, everything you've said is pretty irrelevant, but you're a ****ing badass....so it doesn't matter.


But people keep throwing them out there to "prove" their baseball knowledge even though it's nothing more than utter reliance on a quantitative analysis that someone else developed.

It takes some understanding of the game to be able to understand and explain all of these quantitative analysis. Some people do blindly believe in some things, but that's a pretty off-center generalization considering how much debate there is between saberheads everywhere, and even in this forum.


It's mindless and lame.

Uh-huh, because everyone that doesn't believe in sabermetrics is super-original. Hahaha, think about how much sense that doesn't make.

Everyone that believes RBIs is a super metric does it because it was passed on from someone else. Everyone that believes in grit, heart, and leadership heard it from someone on ESPN, their coach, or their dad.

Everyone that believes in clutch hitting had it shoved down their throats by their favorite analysts, broadcaster, beat writer, father, etc.


Everyone learns everything from someone else. Think before you post or you might find yourself in a huge pile of hypocrisy and double standards.

Seamhead
01-06-2009, 10:00 PM
here is the thing, the averages the stats use change each and every year so i dont understand how it shows the true talent of a pitcher, when it seems like this stats are more based on what the batters do and not what the pitchers do.

It doesn't really vary from season to season. Same as most other league average figures. Plus, I think they use a constant from season to season based on a larger sample size.


Which means the level of talent the pitchers play is huge factor of the tRA, yet it is literally not accounted for by the stat.

Seriously?

It takes everything we KNOW a pitcher has control over it, and accounts for it in a context-neutral situation.

Read the primer again. And actually read this time. Next question you have, ask it statcorner.com. I'm done repeatedly answering the same questions.


Also i dont like how they use the mean to regress the stats, because the mean is largely affected by outliers, if they wanted to make tRA (which im not convinced of yet) more accurate they should regress everything according to the median of each thing, as the median isnt really affected by such outliers (like extreme highs or lows).

Considering how large of a sample size they're drawing from, it's not much of a problem.



Even if the stat were deemed to be most accurate, i would still take Matsuzaka over Beckett for the 2009 season based on 2008 performances

Ok?

papipapsmanny
01-06-2009, 10:46 PM
It doesn't really vary from season to season. Same as most other league average figures. Plus, I think they use a constant from season to season based on a larger sample size.



Seriously?

It takes everything we KNOW a pitcher has control over it, and accounts for it in a context-neutral situation.
Read the primer again. And actually read this time. Next question you have, ask it statcorner.com. I'm done repeatedly answering the same questions.



Considering how large of a sample size they're drawing from, it's not much of a problem.




Ok?

all the pitcher really has control over if your going to say things like that is what pitch he throws, and whether he throws a strile or not. Its really up to the talents of the hitter whether he can hit it or not.

There is defintely a difference between pitching to pedroia (who rarely strikes out) and guy like dunn (who always strikes out).

To me now it looks like the pitcher only gets credit for outs recorded by a strikeout, and gives no credit to out made by the defense, which i believe is obsurd. You say the stat tries to show how maybe some pitchers get luckier than others, but how can you base a stats on that. You cant prove luck exists, and many people dont even believe in luck.

and i know tRA is the pitchers expected ERA per nine innings. Well it didnt happen, what is the point of showing what there expected to do when ERA shows what they just did that season?

and i just dont like how the stat puts more stock into certain types of outs or hits than others.

It can be explained by the simplest baseball saying. An out is an Out, a hit is a hit.

A single is no better than any other type of single no matter whether it was a linedrive, groundball or bloop hit.

I just cant agree with it.

Seamhead
01-06-2009, 11:03 PM
all the pitcher really has control over if your going to say things like that is what pitch he throws, and whether he throws a strile or not. Its really up to the talents of the hitter whether he can hit it or not.

Which is why all the stats which all of the stats that tRA incorporates have strong year to year correlation, right? They are a measure of skill.



There is defintely a difference between pitching to pedroia (who rarely strikes out) and guy like dunn (who always strikes out).


To me now it looks like the pitcher only gets credit for outs recorded by a strikeout, and gives no credit to out made by the defense, which i believe is obsurd.

No. Once again, for the 1000th, batted balls are taken into account; they're just inserted into a context-neutral situation. Certain types of batted balls fall for hits more than others, and pitchers have shown the ability to induce those types of batted balls.


You say the stat tries to show how maybe some pitchers get luckier than others, but how can you base a stats on that. You cant prove luck exists, and many people dont even believe in luck.

Are you a brain dead moron (don't take offense to that; it's just rhetoric)? I've literally explained that part to you like 20 times. Not doing it again.


and i know tRA is the pitchers expected ERA per nine innings. Well it didnt happen, what is the point of showing what there expected to do when ERA shows what they just did that season?

:confused:


and i just dont like how the stat puts more stock into certain types of outs or hits than others.

Again -- what? I don't think you have the slightest idea of what you are talking about.


I
t can be explained by the simplest baseball saying. An out is an Out, a hit is a hit.

Right. A triple is a single.


A single is no better than any other type of single no matter whether it was a linedrive, groundball or bloop hit.

What does this have to do with tRA? You're confusing me now...


I just cant agree with it.

It's impossible to when you still don't know anything about it.

Gigantes4Life
01-06-2009, 11:13 PM
It doesn't really vary from season to season. Same as most other league average figures. Plus, I think they use a constant from season to season based on a larger sample size.



Seriously?

It takes everything we KNOW a pitcher has control over it, and accounts for it in a context-neutral situation.

Read the primer again. And actually read this time. Next question you have, ask it statcorner.com. I'm done repeatedly answering the same questions.



Considering how large of a sample size they're drawing from, it's not much of a problem.




Ok?

I think he meant that tRA doesn't take into account level of competition.

Gigantes4Life
01-06-2009, 11:14 PM
all the pitcher really has control over if your going to say things like that is what pitch he throws, and whether he throws a strile or not. Its really up to the talents of the hitter whether he can hit it or not.

There is defintely a difference between pitching to pedroia (who rarely strikes out) and guy like dunn (who always strikes out).

To me now it looks like the pitcher only gets credit for outs recorded by a strikeout, and gives no credit to out made by the defense, which i believe is obsurd. You say the stat tries to show how maybe some pitchers get luckier than others, but how can you base a stats on that. You cant prove luck exists, and many people dont even believe in luck.

and i know tRA is the pitchers expected ERA per nine innings. Well it didnt happen, what is the point of showing what there expected to do when ERA shows what they just did that season?

and i just dont like how the stat puts more stock into certain types of outs or hits than others.

It can be explained by the simplest baseball saying. An out is an Out, a hit is a hit.

A single is no better than any other type of single no matter whether it was a linedrive, groundball or bloop hit.

I just cant agree with it.

Yet balls in play are much more likely to become hits if it's a linedrive.

papipapsmanny
01-06-2009, 11:24 PM
Which is why all the stats which all of the stats that tRA incorporates have strong year to year correlation, right? They are a measure of skill.






No. Once again, for the 1000th, batted balls are taken into account; they're just inserted into a context-neutral situation. Certain types of batted balls fall for hits more than others, and pitchers have shown the ability to induce those types of batted balls.



Are you a brain dead moron (don't take offense to that; it's just rhetoric)? I've literally explained that part to you like 20 times. Not doing it again.



:confused:



Again -- what? I don't think you have the slightest idea of what you are talking about.


I

Right. A triple is a single.



What does this have to do with tRA? You're confusing me now...



It's impossible to when you still don't know anything about it.

no i understand

alright i should have put this a much easier way. I dont like the fact that a pitcher is hurt because of a higher linedrive%. Or i dont like how an out on a line drive is perceived to be luckier than out recorded by a groundball.


As for the hits a hit thing that was more me whining about the us of BABIP.


Let me rephrase what i also said about the luck thing, and "actually happened". tRA looks at what actually happened and then regresses these numbers as if the pitcher were pitching with a league average defense behind him. Through that it gets the expected ERA. I just dont see the point of it when its after the fact.

papipapsmanny
01-06-2009, 11:28 PM
I think he meant that tRA doesn't take into account level of competition.

yes thats precisely what i meant, and it seems that it has to play a major role in the outcome of the stat.


Yes i know i could have made that a lot easier to infer also that i was asking that.

Gigantes4Life
01-06-2009, 11:32 PM
no i understand

alright i should have put this a much easier way. I dont like the fact that a pitcher is hurt because of a higher linedrive%. Or i dont like how an out on a line drive is perceived to be luckier than out recorded by a groundball.


As for the hits a hit thing that was more me whining about the us of BABIP.


Let me rephrase what i also said about the luck thing, and "actually happened". tRA looks at what actually happened and then regresses these numbers as if the pitcher were pitching with a league average defense behind him. Through that it gets the expected ERA. I just dont see the point of it when its after the fact.

How do you not get the point after?

What do you think is the point of ERA? I'm not being sarcastic, answer that the best you can in your own words.

papipapsmanny
01-06-2009, 11:36 PM
Yet balls in play are much more likely to become hits if it's a linedrive.

yes, but that doesnt mean a pitcher's stats should be hurt for linedrive outs given up and perceived as lucky out.

papipapsmanny
01-06-2009, 11:42 PM
How do you not get the point after?

What do you think is the point of ERA? I'm not being sarcastic, answer that the best you can in your own words.

to me its pretty self explanatory, ERA shows how many runs per 9 innings the pitcher gives up.

I dont see the point at the end of the season seeing their ERA, and then using that to show what their expected ERA was, when their actual ERA is right in front of you

I just don't get it

Gigantes4Life
01-06-2009, 11:49 PM
yes, but that doesnt mean a pitcher's stats should be hurt for linedrive outs given up and perceived as lucky out.

That's not what happens. You don't seem to get it.

Do you or do you not agree that if you give up 10 line drives in a row, you're likely to give up more runs than if you were to give up 10 groundballs?

It's based off of historical measurements if you may; a linedrive is worth .384 runs. A groundball .053, and a flyball .046 (-.096 in the infield).

Homeruns, walks, K's and HBP also have a run value.

That is how tRA is calculated.

Gigantes4Life
01-06-2009, 11:50 PM
to me its pretty self explanatory, ERA shows how many runs per 9 innings the pitcher gives up.

I dont see the point at the end of the season seeing their ERA, and then using that to show what their expected ERA was, when their actual ERA is right in front of you

I just don't get it

Because their actual ERA is inaccurate.

There are park factors, defense factors and luck factors.

Seamhead
01-07-2009, 05:17 PM
to me its pretty self explanatory, ERA shows how many runs per 9 innings the pitcher gives up.

I dont see the point at the end of the season seeing their ERA, and then using that to show what their expected ERA was, when their actual ERA is right in front of you

I just don't get it

Their actual ERA isn't going to be a really precise measure of their actual talent, as it involves way too many factors that create some sort of bias, whether it'd be defense or park. tRA isolates what we are certain a pitcher can control.

By the way, I've been a dick throughout the thread (s), but I was just trying to get the point across and hammer it in. Thanks for asking questions, being open minded, and not coming out with any stupid stat nerd jokes.

Gigantes4Life
01-07-2009, 06:56 PM
If it makes you feel better, sometimes your responses are hilarious :)

ugottabjoshinme
01-08-2009, 06:07 PM
i know absolutely nothing about sabermetrics, but i would like to. where do you start? is there a book or website that you guys all started at to learn the theory behind all this stuff?

Adalbjorg
01-08-2009, 06:18 PM
Alright so you say the stat regresses everything in order to show how each pitcher would pitch if they had the same defense.

Some of you say pitchers on the yankees would have posted better numbers if they were on a team like the Red Sox

well here is my question

Beckett is ranked 12th in tRA for the 2008 season, Lester is Ranked 28th, and Daisuke Matsuzaka is ranked 70th.

To the Casual fan the "old fashion" stats would suggest the either Matsuzaka or Lester had the best year of the three and that clearly Beckett was the worst of the 3.

Yet Beckett is suggested to be the best by tRA when everything is regressed according to defense. Well what is there to regress? They all had the same defense, how could you say the guy with the better numbers did worse than the guy with the worse numbers when they were playing with the same defense. There was nothing really to regress to because they all had the same defense behind them.

This is an honest question, not trying to start anything. (even though i still dont reallly like using the stats

That's correct. tRA suggests that Beckett was very unlucky last year and Daisuke was very lucky. This is compared to 2006, when tRA suggested that Beckett wasn't unlucky, he just pitched badly. tRA and tRA+ are also coupled with measures of a pitcher's defensive support behind him. Two of these measures, O-xO (Outs Minus Expected Outs) and xRR (Expected Runs Minus Recorded runs) both suggest that Beckett had significantly poor defense/bad luck with where balls fell behind him.

But be careful, these stats also suggest the opposite for Matsuzaka. So, if tRA correctly predicts next year's performances, I'd expect Matsuzaka to be the one with a 4.00 ERA and Beckett to be the one with an ERA in the low-3's.

But that's if tRA correctly predicts next year's season. I'm not saying it will or won't.

Adalbjorg
01-08-2009, 06:19 PM
Their actual ERA isn't going to be a really precise measure of their actual talent, as it involves way too many factors that create some sort of bias, whether it'd be defense or park. tRA isolates what we are certain a pitcher can control.

By the way, I've been a dick throughout the thread (s), but I was just trying to get the point across and hammer it in. Thanks for asking questions, being open minded, and not coming out with any stupid stat nerd jokes.

You're a stat nerd! JK, I agree with pretty much everything you've said in this thread.

Ph1lly Diehard
01-08-2009, 06:49 PM
i know absolutely nothing about sabermetrics, but i would like to. where do you start? is there a book or website that you guys all started at to learn the theory behind all this stuff?

Website that explains everything is HERE (http://www.statcorner.com/)

Gigantes4Life
01-08-2009, 06:53 PM
i know absolutely nothing about sabermetrics, but i would like to. where do you start? is there a book or website that you guys all started at to learn the theory behind all this stuff?

I'd suggest buying The Book (http://www.insidethebook.com/) then.

A good place to go would be http://www.fangraphs.com/ and read some of their articles.

Maybe search for one of your favorite players, and they have the most recent articles on them at the top. Read those, that way you know the player well so you can make sense of what they're saying.

If you don't understand some terms, they may explain them or you can just check the glossary.

Myself, I got interested in it from another member on the site, and once it happened, I couldn't stop. I just keep looking stuff up, it's really fascinating.


Website that explains everything is HERE (http://www.statcorner.com/)

It's impossible for anyone site to explain anything.

Statcorner is a bit more complex than the average site, so I wouldn't start there.

Seamhead
01-08-2009, 07:07 PM
i know absolutely nothing about sabermetrics, but i would like to. where do you start? is there a book or website that you guys all started at to learn the theory behind all this stuff?

Books:

Read Moneyball by Michael Lewis. Doesn't actually explain too much, but it's what seems to turn the light on for most people.

Pick up Baseball Between the Numbers, a Bill James book, The Numbers Game by Alan Schawrz. Total Baseball's a good beginner's book. After you understand the basics, read The Book: Playing the Percentage. Hell, you can go ahead and pick that one up, as it the end of the chapter, it presents/summarizes in a way that's easy to understand.

Read blogs. Watch baseball. Think about all of it while you watch or even play (PFFT LIKE THOSE STUPID NERDS CAN EVEN THROW!!)

Gigantes4Life
01-08-2009, 07:28 PM
Just curious seamhead, how much of The Book have you read?

Seamhead
01-08-2009, 07:30 PM
Just curious seamhead, how much of The Book have you read?

Hmmm...let's see...approximately 0% of it?

Gigantes4Life
01-08-2009, 07:42 PM
Hmmm...let's see...approximately 0% of it?

:laugh:

I honestly wasn't expecting that.

I figured you'd say like 10% maybe.

Well, you learn something new every day :p

quiksilver2491
01-08-2009, 07:55 PM
Hmmm...let's see...approximately 0% of it?

:speechless:

I would of thought that you would of read that one? I borrowed it from a friend who wasn't into it and it was the single driving force that got me into saber stats and things of the sort, great book.

Seamhead
01-08-2009, 08:03 PM
:speechless:

I would of thought that you would of read that one? I borrowed it from a friend who wasn't into it and it was the single driving force that got me into saber stats and things of the sort, great book.

Never had the means (parents won't let me use credit card) to buy it, and I've been too lazy to go to Barnes and Nobles and pick up via the ISBN. Plus, I like to read different topics, since I spent 65% of my life doing baseball related things, whether on the internet or not. I might go get it later tonight, though. I have like $20 that are eating a hole in my pocket.

I read the preview chapter on Amazon and the reviews, though.:o

Gigantes4Life
01-08-2009, 08:06 PM
I've read parts of it, but don't own it.

I'll probably buy it eventually as well.

natepro
01-08-2009, 08:17 PM
Never had the means (parents won't let me use credit card) to buy it, and I've been too lazy to go to Barnes and Nobles and pick up via the ISBN. Plus, I like to read different topics, since I spent 65% of my life doing baseball related things, whether on the internet or not. I might go get it later tonight, though. I have like $20 that are eating a hole in my pocket.

I read the preview chapter on Amazon and the reviews, though.:o

I actually just got it from the library and read it that way. :nod:



Watch baseball.

I assume by "watch baseball" you mean "watch what happens when you plug numbers into your spreadsheet and baseball stats pop out as if a game was played somewhere, at sometime, by some people." Cause... who watches baseball anymore? :confused:

Seamhead
01-08-2009, 08:31 PM
I actually just got it from the library and read it that way. :nod:




I assume by "watch baseball" you mean "watch what happens when you plug numbers into your spreadsheet and baseball stats pop out as if a game was played somewhere, at sometime, by some people." Cause... who watches baseball anymore? :confused:

No. I actually have a program that interprets the baseball stuff thingy occurring on the field and translates it into 1's and 0's, so I can understand what's going on.

natepro
01-08-2009, 08:41 PM
No. I actually have a program that interprets the baseball stuff thingy occurring on the field and translates it into 1's and 0's, so I can understand what's going on.

You actually still use the things that people do? :laugh2:


Don't you know the game is played on paper and not on a field by real people? :rolleyes:

Seriously, back into your mother's basement. There's nothing out there for you but people and there nasty intangibles.

papipapsmanny
01-08-2009, 09:11 PM
Their actual ERA isn't going to be a really precise measure of their actual talent, as it involves way too many factors that create some sort of bias, whether it'd be defense or park. tRA isolates what we are certain a pitcher can control.

By the way, I've been a dick throughout the thread (s), but I was just trying to get the point across and hammer it in. Thanks for asking questions, being open minded, and not coming out with any stupid stat nerd jokes.

alright after reading things up, and looking at the different sites i am going to have to agree that tRA has its great use, and may in fact be the best stat to see the pitcher's true talent level, and expected results.

But despite that belief I believe it is poor in determining who had a better season. Although Beckett may have gotten unluckier I would still say lester and matsuzaka had a better year for the 2008 season. I dont know if that really makes sense. To me tRA is suggesting that Beckett is the best bet to have a good season next season and that lester and matsuzaka are good bets to regress, but still i do believe lester and matsuzaka had better years last year than beckett. You have any clue what i am trying to say.

Gigantes4Life
01-08-2009, 10:22 PM
From a traditional standpoint, they did have better seasons.

I disagree with your disagreement to be honest.

There's nothing to suggest that Lester will necessarily regress, since he's young and improving. Beckett is what, 28? So I could see a better year for him, but not necessarily.

You may be able to say that Lester had a better year than Beckett, but Dice-K just isn't that good.

Seamhead
01-08-2009, 10:23 PM
alright after reading things up, and looking at the different sites i am going to have to agree that tRA has its great use, and may in fact be the best stat to see the pitcher's true talent level, and expected results.

But despite that belief I believe it is poor in determining who had a better season. Although Beckett may have gotten unluckier I would still say lester and matsuzaka had a better year for the 2008 season. I dont know if that really makes sense. To me tRA is suggesting that Beckett is the best bet to have a good season next season and that lester and matsuzaka are good bets to regress, but still i do believe lester and matsuzaka had better years last year than beckett. You have any clue what i am trying to say.

Yeah, I do see what you're saying, and it is somewhat understandable. I don't agree with it, but I've felt similarly about different metrics at times (like when I first read the DIPS theory).

Their stats might look more appealing, but it doesn't mean they were better pitchers in 2008. I'm not going to reward a guy for something that wasn't really within his control.

papipapsmanny
01-08-2009, 10:38 PM
here is the thing your going to take lesters production over becketts production last year. Whether he aquired it by luck or not he was still produced the better numbers.

For some reason its hard to explain what i am trying to say

papipapsmanny
01-08-2009, 10:39 PM
Yeah, I do see what you're saying, and it is somewhat understandable. I don't agree with it, but I've felt similarly about different metrics at times (like when I first read the DIPS theory).

Their stats might look more appealing, but it doesn't mean they were better pitchers in 2008. I'm not going to reward a guy for something that wasn't really within his control.

i am not going to say they were better in the sense of their abilities, but i am still going to take their production over beckett's last year.

Gigantes4Life
01-08-2009, 11:06 PM
I get what you're saying.

You're glad the way it turned out.

Just like the Angels and Rays are glad with their final records, even though they both overperformed.

No hitter ever says "wow, I really didn't deserve that bloop single. Here, I'm out now."