ARLINGTON, VA (Bottom of the Fourth) - In recent years, the baseball website FanGraphs has become a destination for cutting-edge baseball statistics and analysis. So cutting-edge, it seems based on news released this morning, that it has found the end of baseball statistics.
Founder David Appelman says he came upon the end of baseball stats accidentally while looking for his slide rule. "I checked under the bed, in the fridge, behind all my Bill James posters, but it was nowhere. So I started looking in spreadsheets, and got into some really complex non-linear weight stuff based on ZiPS projections."
Appelman says he didn't find his slide rule there, but soon completely forgot he was even looking for it. "Something caught my eye when I ran an anti-regression model on the fractal matrix of Luke Gregerson's slider", he explained. "I didn't believe it at first, but I've checked it to 99.99999% confidence. It was the end of baseball statistics, there's no doubt about it." Appelman went on to add that there was a restaurant there, and that the food was "above-average".
With the discovery, FanGraphs has decided against attempting to improve on existing statistics, as it believes such an endeavour would be fruitless. Instead, the website believes the only way to move forward is to go back to the beginning. "It's like an old-school video game," commented editor Dave Cameron, "when your dude reaches the edge of the screen, he appears on the opposite edge. We're taking that philosophy and really running with it."
The first implementation of FanGraphs' new methodology has players ranked by games played. Consequently, the title of best player in 2011 is currently a 13-way tie between Torii Hunter, Adrian Beltre, Alex Gonzalez, Michael Young, Miguel Cabrera, Dan Uggla, James Loney, Martin Prado, Howard Kendrick, Matt Kemp, Peter Bourjos, Austin Jackson and Jason Heyward.
But the team of analysts is already furiously working to push the boundaries. Despite attempts to control information flow, an anonymous source has leaked that the next statistic to appear on the site will be called either "ats" or "bats".
Showing posts with label Xavermetrics. Show all posts
Showing posts with label Xavermetrics. Show all posts
Tuesday, May 10, 2011
Monday, April 25, 2011
Smaller Samples Actually Most Accurate: Fuld
TAMPA BAY (Bottom of the Fourth) - Many expected the Tampa Bay Rays to regress in 2011 after losing stars Carl Crawford and Carlos Pena to free agency and Evan Longoria to an early-season injury, but the team is currently sitting second in the AL East thanks to some unlikely contributors, chief among them the previously unheralded Sam Fuld.
While early-season success is often dismissed as a small sample-size fluke, Fuld doesn't believe this to be logical. "I took a statistics class in university," the Stanford grad told Bottom of the Fourth, "and these stat geeks have it backward. The smaller the sample, the more accurate."
Fuld elaborated, explaining that the "binomial nature" of the "central limit theorem" produces a "chirality" that makes it seem as if larger samples are more predictive, but that this is simply a red herring. Upon observing the blank stares on the faces of the media throng surrounding him, Fuld aborted his attempt at explanation, instead asking reporters to "just trust me. I went to Stanford."
While early-season success is often dismissed as a small sample-size fluke, Fuld doesn't believe this to be logical. "I took a statistics class in university," the Stanford grad told Bottom of the Fourth, "and these stat geeks have it backward. The smaller the sample, the more accurate."
Fuld elaborated, explaining that the "binomial nature" of the "central limit theorem" produces a "chirality" that makes it seem as if larger samples are more predictive, but that this is simply a red herring. Upon observing the blank stares on the faces of the media throng surrounding him, Fuld aborted his attempt at explanation, instead asking reporters to "just trust me. I went to Stanford."
 |
| Fuld's teammates have taken to calling him 'Professor' |
Saturday, October 16, 2010
Could Philadelphia Win Game 1 With 0 Runs?
In their respective National League Division Series, Roy Halladay and Tim Lincecum each pitched the first playoff game of their career. Halladay, as you probably know, threw a no-hitter, the second in playoff history. The next day, Lincecum hurled a two-hit shutout while striking out 14 Atlanta Braves. His game score, a statistic designed as a fun way to compare individual games thrown by starting pitchers, was 96, and was actually slightly higher than Halladay's (95). Most years, a game score that high would rank somewhere in the top three games pitched all season, and it happened twice within two days in these 2010 playoffs.
Halladay and Lincecum are scheduled to face each other in game one of the National League Championship Series on Saturday, and ever since that match-up became a reality, fans have been salivating in anticipation. What might happen? Simultaneous no-hitters? 54 strikeouts combined? Will any hitter even make contact? The hyperbole has gotten out of hand.
(Note: the aforementioned out-of-hand hyperbole refers exclusively to that put forth on this blog.)
But here at Bottom of the Fourth, we think the hyperbole is completely justified - that's how good Halladay and Lincecum are. In fact, we believe it isn't yet out-of-hand enough.
So the question must be asked: could the Phillies get shut out and still win this game?
It seems like an impossible task, one that's never been done before. But just because something has never been done doesn't mean it will never happen. Neil Armstrong never said "well, nobody's ever walked on the moon, it must be impossible." No. He went out and walked on the fucking moon. Did the Wright brothers ever think "hey, maybe we shouldn't spend all this time building something that nobody's ever built before"? Of course not. They kept building and flew a fucking plane. And just because no-one had ever filmed a recorder version of My Heart Will Go On by candlelight, that didn't fucking stop Matt Mulholland.
So yeah, I believe the answer to the question is yes, the Phillies could get shut out and still win this game. Now, you may be wondering: why the Phillies and not the Giants? And it's a fair question - it certainly could happen both ways. I just think the Phillies superior offense will enable them to score fewer negative runs.
Which leads us to another question: negative runs? How does that work?
It's actually a little-known rule that has been in the book for many years. It hasn't been applied in recent memory because, well, who actually reads all the rules? That tome is thicker than Ken Griffey Jr.'s head in the Simpsons. Major league umpires simply aren't aware of the rule, so it never gets applied.
That doesn't explain how it works, though, so here is a short primer. Whenever a pitcher induces a batter to strike out in a manner so foolish that an umpire judges it to have been at least "ridiculous", and up to and including "absurd", the batter receives a Reverse Walk. He is sent to third base and is to advance backwards on the base-paths. He doesn't run on balls in play (at least he isn't required to, but he may if he wants, at great disadvantage to his team); he only advances whenever another Reverse Walk is issued, in which case the batter does not go to third base; only the current runner moves up. (i.e. there is only one Reverse Runner on the bases at any one time.)
When a Reverse Runner reaches home plate, his team scores a negative run. This run is taken away from the team's current score, and if that score is already zero, it dips below. Now, the best case scenario for a pitcher in one inning is to induce three strikeouts* greater than or equal to "ridiculous". In this case, the Reverse Runner first goes to third, then second, then first, and cannot reach home plate. However, unlike regular base-runners, Reverse Runners are carried over to the next inning, so a team would be gradually accumulating negative runs throughout the game, if the rule were applied properly.
*Note: like regular strikeouts, it is in fact possible to accumulate four Reverse Walks, and therefore a negative run, in the same inning; it simply requires one of the "ridiculous" strikeout victims to reach first base on a dropped ball by the catcher. This results in the unlikely scenario of runners going to first and third on the same strikeout. A pinch runner is used as the Reverse Runner.
Teams would of course need to be aware of which runners are which. Regular rules don't allow two runners to be at the same base, but that scenario is sometimes unavoidable when Reverse Runners come into play. A team may have two runners at the same base, as long as they're traveling in different directions, but need to be extra cognizant of the situation to ensure that no runner travels in the wrong direction, resulting in the runner being called automatically out. Runners may also not make contact, including anything from incidental base-running-related contact to high-fiving and other celebratory gestures. This temptation can be difficult to avoid in a case such as a walk-off home run because of the finality of the play; however, in this case, a flagrant high-five or pat on the butt by the Reverse Runner results in a reversal of the home run and a penalty out, possibly enough to turn a win immediately into a loss through a simple union of palms.
Both teams in the NLCS have their fair share of Ridiculous Strikers Out (RSOs). Pat Burrell (145), Aaron Rowand (129) and Andres Torres (128) of the Giants were each in the top 10 in the NL in RSO+, a statistic designed to compare the rate of Ridiculous Strikeouts (average is 100). But Philadelphia has a secret weapon. Ryan Howard, while merely very bad as measured by RSO+ (135), is the most prolific Preposterous Striker Out in the history of the game.
Preposterous Strikeouts (PSOs) are a notch above Ridiculous on the Ridiculous-Absurd scale, and statisticians have found that PSOs are much more predictive of future ridiculosity than RSOs. Ryan Howard has led the majors in PSOs each of the last five years, and even holds the major league record for PSOs in a season (63 in 2008) So despite San Francisco's incredibly Ridiculous 2010 outfield, if there's one player who will make a (negative) difference with his bat in this series, it's Howard.
Halladay and Lincecum are scheduled to face each other in game one of the National League Championship Series on Saturday, and ever since that match-up became a reality, fans have been salivating in anticipation. What might happen? Simultaneous no-hitters? 54 strikeouts combined? Will any hitter even make contact? The hyperbole has gotten out of hand.
(Note: the aforementioned out-of-hand hyperbole refers exclusively to that put forth on this blog.)
But here at Bottom of the Fourth, we think the hyperbole is completely justified - that's how good Halladay and Lincecum are. In fact, we believe it isn't yet out-of-hand enough.
So the question must be asked: could the Phillies get shut out and still win this game?
It seems like an impossible task, one that's never been done before. But just because something has never been done doesn't mean it will never happen. Neil Armstrong never said "well, nobody's ever walked on the moon, it must be impossible." No. He went out and walked on the fucking moon. Did the Wright brothers ever think "hey, maybe we shouldn't spend all this time building something that nobody's ever built before"? Of course not. They kept building and flew a fucking plane. And just because no-one had ever filmed a recorder version of My Heart Will Go On by candlelight, that didn't fucking stop Matt Mulholland.
So yeah, I believe the answer to the question is yes, the Phillies could get shut out and still win this game. Now, you may be wondering: why the Phillies and not the Giants? And it's a fair question - it certainly could happen both ways. I just think the Phillies superior offense will enable them to score fewer negative runs.
Which leads us to another question: negative runs? How does that work?
It's actually a little-known rule that has been in the book for many years. It hasn't been applied in recent memory because, well, who actually reads all the rules? That tome is thicker than Ken Griffey Jr.'s head in the Simpsons. Major league umpires simply aren't aware of the rule, so it never gets applied.
That doesn't explain how it works, though, so here is a short primer. Whenever a pitcher induces a batter to strike out in a manner so foolish that an umpire judges it to have been at least "ridiculous", and up to and including "absurd", the batter receives a Reverse Walk. He is sent to third base and is to advance backwards on the base-paths. He doesn't run on balls in play (at least he isn't required to, but he may if he wants, at great disadvantage to his team); he only advances whenever another Reverse Walk is issued, in which case the batter does not go to third base; only the current runner moves up. (i.e. there is only one Reverse Runner on the bases at any one time.)
 |
| An example of a strikeout that is at least "ridiculous" |
When a Reverse Runner reaches home plate, his team scores a negative run. This run is taken away from the team's current score, and if that score is already zero, it dips below. Now, the best case scenario for a pitcher in one inning is to induce three strikeouts* greater than or equal to "ridiculous". In this case, the Reverse Runner first goes to third, then second, then first, and cannot reach home plate. However, unlike regular base-runners, Reverse Runners are carried over to the next inning, so a team would be gradually accumulating negative runs throughout the game, if the rule were applied properly.
*Note: like regular strikeouts, it is in fact possible to accumulate four Reverse Walks, and therefore a negative run, in the same inning; it simply requires one of the "ridiculous" strikeout victims to reach first base on a dropped ball by the catcher. This results in the unlikely scenario of runners going to first and third on the same strikeout. A pinch runner is used as the Reverse Runner.
Teams would of course need to be aware of which runners are which. Regular rules don't allow two runners to be at the same base, but that scenario is sometimes unavoidable when Reverse Runners come into play. A team may have two runners at the same base, as long as they're traveling in different directions, but need to be extra cognizant of the situation to ensure that no runner travels in the wrong direction, resulting in the runner being called automatically out. Runners may also not make contact, including anything from incidental base-running-related contact to high-fiving and other celebratory gestures. This temptation can be difficult to avoid in a case such as a walk-off home run because of the finality of the play; however, in this case, a flagrant high-five or pat on the butt by the Reverse Runner results in a reversal of the home run and a penalty out, possibly enough to turn a win immediately into a loss through a simple union of palms.
Both teams in the NLCS have their fair share of Ridiculous Strikers Out (RSOs). Pat Burrell (145), Aaron Rowand (129) and Andres Torres (128) of the Giants were each in the top 10 in the NL in RSO+, a statistic designed to compare the rate of Ridiculous Strikeouts (average is 100). But Philadelphia has a secret weapon. Ryan Howard, while merely very bad as measured by RSO+ (135), is the most prolific Preposterous Striker Out in the history of the game.
Preposterous Strikeouts (PSOs) are a notch above Ridiculous on the Ridiculous-Absurd scale, and statisticians have found that PSOs are much more predictive of future ridiculosity than RSOs. Ryan Howard has led the majors in PSOs each of the last five years, and even holds the major league record for PSOs in a season (63 in 2008) So despite San Francisco's incredibly Ridiculous 2010 outfield, if there's one player who will make a (negative) difference with his bat in this series, it's Howard.
Subscribe to:
Posts (Atom)