+ Reply to Thread
Page 133 of 133 FirstFirst ... 33 83 123 130 131 132 133
Results 3,301 to 3,324 of 3324
  1. #3301
    NYYF Legend

    mbn007's Avatar
    Join Date
    Jul 2004
    Location
    Brooklyn, NY

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by Carsten Charles View Post
    Romine was one of the biggest negatives in this game, which is to be expected from one of the worst players in baseball.
    The way Gray pitched, I think it's a big stretch to say Romine was a big negative.

  2. #3302
    Better than you teknetic's Avatar
    Join Date
    May 2007
    Location
    NJ

    Re: 2017 Joe Girardi Performance Thread

    Gray kept them off the board, but it's not like he was lights out. He labored quite a bit and Romine's terrible pitch framing failed to steal a few strikes in that final inning.

  3. #3303

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by ClownPickle View Post
    This is why Joe has to go:

    @Buster_ESPN
    Fangraphs: Yankees' win probability never dipped below 98.8% after 6th inning in Monday's game. But Green, Chapman warmed up; Kahnle pitched
    Just wanted to point out that the Fangraphs win probability for the Yankees was 5.8% after they were down 4-0 last night. I find these projection statistics to be a particularly annoying arm of sabermetrics. It's patently ridiculous to think that you could play out that scenario 17 times or so and they'd only win the game once.
    /sarcasm
    KayNOTForPresident

  4. #3304
    Suspended
    Join Date
    Aug 2013
    Location
    San Diego

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by KayForPresident View Post
    Just wanted to point out that the Fangraphs win probability for the Yankees was 5.8% after they were down 4-0 last night. I find these projection statistics to be a particularly annoying arm of sabermetrics. It's patently ridiculous to think that you could play out that scenario 17 times or so and they'd only win the game once.
    Calmer than you are.

  5. #3305
    NYYF Legend

    Yankee Tripper's Avatar
    Join Date
    Nov 2007
    Location
    Left coast

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by teknetic View Post
    Gray kept them off the board, but it's not like he was lights out. He labored quite a bit and Romine's terrible pitch framing failed to steal a few strikes in that final inning.
    It's funny everyone was creaming over McCullers but saying Gray was laboring.

    Through 5IP, Gray was 80P 1H, 1 BB, 4K, 0R
    Through 5IP, McC was 76P 1H, 2BB, 3K, 0R

    The only difference was McCullers was able to pitch around 2 lead off walks in the 4th & 5th while Gray was pulled after BB, CI in the 6th.
    Baseball is life;
    the rest is just details.

  6. #3306
    NYYF Legend

    Yankee Tripper's Avatar
    Join Date
    Nov 2007
    Location
    Left coast

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by KayForPresident View Post
    Just wanted to point out that the Fangraphs win probability for the Yankees was 5.8% after they were down 4-0 last night. I find these projection statistics to be a particularly annoying arm of sabermetrics. It's patently ridiculous to think that you could play out that scenario 17 times or so and they'd only win the game once.
    Well I think they'd lost the last 27 games when trailing by 4 that late in the game so 1 out of 17 seems reasonable to me.
    Baseball is life;
    the rest is just details.

  7. #3307

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by Yankee Tripper View Post
    It's funny everyone was creaming over McCullers but saying Gray was laboring.

    Through 5IP, Gray was 80P 1H, 1 BB, 4K, 0R
    Through 5IP, McC was 76P 1H, 2BB, 3K, 0R

    The only difference was McCullers was able to pitch around 2 lead off walks in the 4th & 5th while Gray was pulled after BB, CI in the 6th.
    What was their balls to strike ratio? I know that Sonny threw a lot of balls last night...not sure about McCullers though.
    "Owning the Yankees," Steinbrenner once said, "is like owning the Mona Lisa."

  8. #3308

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by Yankee Tripper View Post
    Well I think they'd lost the last 27 games when trailing by 4 that late in the game so 1 out of 17 seems reasonable to me.
    Regular season is completely different animal.
    /sarcasm
    KayNOTForPresident

  9. #3309
    NYYF Legend

    Yankee Tripper's Avatar
    Join Date
    Nov 2007
    Location
    Left coast

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by b_joseph View Post
    What was their balls to strike ratio? I know that Sonny threw a lot of balls last night...not sure about McCullers though.
    Gray threw a lot of pitches in the first 2 innings but was very economical after.

    For the game BR tells me McCullers was 81P, 50 strikes, Gray was 84P 46 strikes. In looking at the 6th Gray threw 7 Balls and 2 strikes so through 5 he was 75P, 44 Strikes. He lost the zone in the 6th.
    Baseball is life;
    the rest is just details.

  10. #3310
    NYYF Legend

    Yankee Tripper's Avatar
    Join Date
    Nov 2007
    Location
    Left coast

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by KayForPresident View Post
    Regular season is completely different animal.
    Yeah you'd expect to win at an even lower rate since you won't ever be facing the back end of someones bull pen in that situation like you might in the regular season.
    Baseball is life;
    the rest is just details.

  11. #3311

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by ClownPickle View Post
    Calmer than you are.
    /sarcasm
    KayNOTForPresident

  12. #3312
    Tends to be difficult JL25and3's Avatar
    Join Date
    Jul 2006

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by mbn007 View Post
    I don't know what Fangraphs is talking about. Chapman tossed a ball around in the pen. He did not "get hot", or warmup to come in.


    Not sure about Green, but I would bet it was the same deal.
    Green was throwing for real.
    A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines.
    - Barry Manilow

  13. #3313

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by KayForPresident View Post
    Just wanted to point out that the Fangraphs win probability for the Yankees was 5.8% after they were down 4-0 last night. I find these projection statistics to be a particularly annoying arm of sabermetrics. It's patently ridiculous to think that you could play out that scenario 17 times or so and they'd only win the game once.
    The same people making these projections still think Hillary is going to win. I agree 100% with you. There's an arm of the statistics community that try to quantify everything. It's very unscientific and most ppl don't realize that the models are not actually predictive but rather just extrapolate from previous results. There's no point added information by saying x team has a 5.8432 percent chance of winning than simply saying team x is up 4 in th 6th. The former is preferred by people who like to think of themselves as smart but don't really understand science or significant figures.

  14. #3314

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by billyhoyle View Post
    The same people making these projections still think Hillary is going to win. I agree 100% with you. There's an arm of the statistics community that try to quantify everything. It's very unscientific and most ppl don't realize that the models are not actually predictive but rather just extrapolate from previous results. There's no point added information by saying x team has a 5.8432 percent chance of winning than simply saying team x is up 4 in th 6th. The former is preferred by people who like to think of themselves as smart but don't really understand science or significant figures.
    I think you're referencing yourself. People who understand probability are fine with those numbers.
    Something that has a 5% chance of happening will still happen. We just got to see one of the 5%...of course everyone disregards the numerous games that are won by a team up 4 that late in a game because its not a notable event.

    If someone is up 3-0 in the LCS, the odds are barely worth reading. But that doesnt mean that once it happens, it makes those odds meaningless. It just means that the rarest of rare events occurred at the right time for some and the wrong for others.
    "Owning the Yankees," Steinbrenner once said, "is like owning the Mona Lisa."

  15. #3315

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by b_joseph View Post
    I think you're referencing yourself. People who understand probability are fine with those numbers.
    Something that has a 5% chance of happening will still happen. We just got to see one of the 5%...of course everyone disregards the numerous games that are won by a team up 4 that late in a game because its not a notable event.

    If someone is up 3-0 in the LCS, the odds are barely worth reading. But that doesnt mean that once it happens, it makes those odds meaningless. It just means that the rarest of rare events occurred at the right time for some and the wrong for others.
    I do not have a problem with saying that a team up 4 has won X percent of the time. Or that a team up 3-0 LCS has won 99/100 times. The problem is that these are not actually predictive indicators for individual games because they ignore every detail other than the score and inning. For example, 538 tweeted out something like the falcons had a 97% of chance of winning during the super bowl. However, that "percent" ignores the fact that Tom Brady is the quarterback. There is no added value from providing these real time updates on percentages.

    So when should people use statistical analyses? There are a number of reasons where it provides value. The first, and most common, is to demonstrate significance/correlation for a set of data. The second is to simplify a complex set of data and "fish" out particular components that are interesting.

    Creating "win probabilities" reported to the decimal of a percentage is not actually a predictor of future events. It's a description of the past. That is what most people don't understand, and you clearly don't either.

  16. #3316
    Brien "The Incident" just-blaze's Avatar
    Join Date
    Nov 2004
    Location
    Jax

    Re: 2017 Joe Girardi Performance Thread

    I just want to state for the record that I want Joe back for at least a year.

    This man has been a boss aside from the Romine incident. And that might have been the turning point in the Kraken reawakening.

  17. #3317
    Suspended
    Join Date
    Aug 2013
    Location
    San Diego

    Re: 2017 Joe Girardi Performance Thread

    Nice call on Headley over Holliday today.

    If we go any further they need to remove Holliday from the roster.

  18. #3318
    NYYF Legend


    Join Date
    Apr 2004
    Location
    I live by the Sound

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by ClownPickle View Post
    If we go any further they need to remove Holliday from the roster.
    Yeah, I don't mean to be cold but Holliday is basically a fossil remnant at this point. His hitting never recovered from the viral illness and T. Frazier has pretty much pushed him out as the "veteran leader" guy. Can't lead from the bench.

  19. #3319

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by billyhoyle View Post



    Creating "win probabilities" reported to the decimal of a percentage is not actually a predictor of future events. It's a description of the past. That is what most people don't understand, and you clearly don't either.
    But it is predicting future events ( as much as anything actually can ). If something is 99% likely...then on the flip side, something is 1% likely.
    In said Superbowl, we witnessed the 1% likely. In game 4, we witnessed the 5% likely.

    Yes, if Brady is the QB, your chances are better than if Joe Average is your QB. That's when your interpretation of the data comes to play. You could then look at what Brady has done in those situations...How often has Brady been behind by 25+ on the road/neutral stadium? The answer would be zero before that awful night in February.

    Would you prefer them just to say '' hey, its monumentally unlikely that Team A makes a comeback in this game. Team B losing from a lead like this rarely ever happens ''.
    "Owning the Yankees," Steinbrenner once said, "is like owning the Mona Lisa."

  20. #3320

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by billyhoyle View Post
    I do not have a problem with saying that a team up 4 has won X percent of the time. Or that a team up 3-0 LCS has won 99/100 times. The problem is that these are not actually predictive indicators for individual games because they ignore every detail other than the score and inning. For example, 538 tweeted out something like the falcons had a 97% of chance of winning during the super bowl. However, that "percent" ignores the fact that Tom Brady is the quarterback. There is no added value from providing these real time updates on percentages.

    So when should people use statistical analyses? There are a number of reasons where it provides value. The first, and most common, is to demonstrate significance/correlation for a set of data. The second is to simplify a complex set of data and "fish" out particular components that are interesting.

    Creating "win probabilities" reported to the decimal of a percentage is not actually a predictor of future events. It's a description of the past. That is what most people don't understand, and you clearly don't either.
    good post
    the 1890's socialist Player's League, founded by workers:
    socialistappeal.org/history-theory/us-history/959-the-1890-players-league.html

  21. #3321

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by b_joseph View Post
    Would you prefer them just to say '' hey, its monumentally unlikely that Team A makes a comeback in this game. Team B losing from a lead like this rarely ever happens ''.
    Yes, exactly. I would very much prefer that. I don't have a problem with somebody giving the statistics if it's framed in a way that clearly describes past occurrences. Here's an example. If the yankees are down 4 in the 7th inning, and they've only comeback to win one time out of 100 in the past 10 years. The following statements I don't mind:

    1. "When down by 4 runs, the Yankees have lost 99 out of their last 100 games."

    2. (as you said) "It is monumentally unlikely that the Yankees come back for this."

    What I despise is when websites publish that the Yankees have a 1% chance of winning the game, which then ticks up to 1.5% when they get a single, and 1.55% when a ball is thrown to the next batter.

    This is the scam that specifically 538 plays on their readers. They build "predictive" models and provide quantification on virtually everything, when most of the time these models only give the appearance of precision. In reality, they are plagued by systematic error and are no more meaningful than simply saying (as you point out) that something is very unlikely to happen.

    It's popular pseudoscience. Here is a hero of mine describing this phenomenon: https://vimeo.com/118188988

  22. #3322

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by billyhoyle View Post
    Yes, exactly. I would very much prefer that. I don't have a problem with somebody giving the statistics if it's framed in a way that clearly describes past occurrences. Here's an example. If the yankees are down 4 in the 7th inning, and they've only comeback to win one time out of 100 in the past 10 years. The following statements I don't mind:

    1. "When down by 4 runs, the Yankees have lost 99 out of their last 100 games."

    2. (as you said) "It is monumentally unlikely that the Yankees come back for this."

    What I despise is when websites publish that the Yankees have a 1% chance of winning the game, which then ticks up to 1.5% when they get a single, and 1.55% when a ball is thrown to the next batter.

    This is the scam that specifically 538 plays on their readers. They build "predictive" models and provide quantification on virtually everything, when most of the time these models only give the appearance of precision. In reality, they are plagued by systematic error and are no more meaningful than simply saying (as you point out) that something is very unlikely to happen.

    It's popular pseudoscience. Here is a hero of mine describing this phenomenon: https://vimeo.com/118188988
    Right on the money. Thanks for the link.

  23. #3323

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by billyhoyle View Post
    ...
    But take it further. How do you come to the conclusion that something is highly unlikely? You would come to it by reviewing a situation and seeing how a similar situation played out in the past.
    Once doing that, you can put it to number because you can put anything in the world to number.

    So if we say something is a coin flip. We all know what number that represents...from there, I don't think its unreasonable to put the next stages to number too.

    Thanks for the link btw...I've seen it before but never hurts to hear his voice. He is a icon to us all.
    "Owning the Yankees," Steinbrenner once said, "is like owning the Mona Lisa."

  24. #3324

    Re: 2017 Joe Girardi Performance Thread

    Quote Originally Posted by b_joseph View Post
    But take it further. How do you come to the conclusion that something is highly unlikely? You would come to it by reviewing a situation and seeing how a similar situation played out in the past.
    Once doing that, you can put it to number because you can put anything in the world to number.

    So if we say something is a coin flip. We all know what number that represents...from there, I don't think its unreasonable to put the next stages to number too.

    Thanks for the link btw...I've seen it before but never hurts to hear his voice. He is a icon to us all.
    I think the coin flip analogy is a good one. It's precisely the example that was used in the beginning of my first statistics course to explain probability theory.

    Coin flips can be repeated in exactly the same way whether they're repeated 10 times, or 100 times, or a thousand times. As a history of coin flips is accumulated, it's easy to determine the likelihood of various outcomes based on the history.

    The problem with probability is that most things in life aren't repeated under the exact same conditions as coin flips are. Yet statistics are often used as though are. The outcomes of baseball games are the result of a complex interaction of many factors under a variety of conditions. Coin flips are not.

    One of my favorite cases is the 2004 Red Sox come back. The claim was that it had never been done before. Well, it had. Any number of times. What the Red Sox did was to win 4 straight in a postseason series. The claim treated that case as though the likelihood of it happening changed after the opponent won the first 3 games. It didn't. 4 straight is still 4 straight regardless of whether the opponent has won no games, 1 game, 2 games, or even 3 games first. Are the odds of getting 4 straight heads in a coin flip changed by what's happened in the few coin flips immediately preceding the 4 straight? It's a great example of how bogus probabilities are created and supported by bogus probability numbers just to make for interesting reading material on TV or internet blogs.

Thread Information

Users Browsing this Thread

There are currently 8 users browsing this thread. (0 members and 8 guests)

     

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts