25 July 2008

Pausch to remember

Randy Pausch died this morning.  His book The Last Lecture is based on a lecture he gave at Carnegie Mellon after he found that he had terminal cancer.  It's full of not empty sayings and false hopes but real stories and lessons learned from a full life.  It is the only inspirational book I've read that deserves to be called inspirational.

21 July 2008

A look ahead

Last week I analyzed how the major-league standings would look if all teams were equally lucky in the way that runs fell among games and hits and outs fell among innings.  Well, unfortunately, not all teams are equally lucky, and only the real results matter.  But the real won-loss records so far don't predict the future; in fact, a team's Nostradamian record is a far better predictor of future wins and losses than past wins and losses.  So (using the All-Star break as the cutoff point) let's use each team's Nostradamian record to predict its winning percentage for the rest of the season and see how the final standings would look:

AL Eastextrapolatedreal (at ASB)
Red Sox97.4–64.6.60094057–40.587629
Rays93.1–68.9.57494655–39.585106
Yankees85.5–76.5.52766450–45.526316
Blue Jays83.0–79.0.51247247–48.494737
Orioles78.6–83.4.48493745–48.483871

AL Centralextrapolatedreal (at ASB)
White Sox94.2–67.8.58143754–40.574468
Twins85.0–77.0.52440453–42.557895
Tigers82.2–79.8.50713547–47.500000
Indians72.6–89.4.44840541–53.436170
Royals72.0–90.0.44440243–53.447917

AL Westextrapolatedreal (at ASB)
Angels89.6–72.4.55293657–38.600000
Athletics88.8–73.2.54843851–44.536842
Rangers83.1–78.9.51311950–46.520833
Mariners65.6–96.4.40481137–58.389474

NL Eastextrapolatedreal (at ASB)
Mets87.8–74.2.54200051–44.536842
Phillies87.7–74.3.54118652–44.541667
Marlins82.7–79.3.51053550–45.526316
Braves82.7–79.3.51019845–50.473684
Nationals61.0–101.0.37642636–60.375000

NL Centralextrapolatedreal (at ASB)
Cubs97.1–64.9.59964257–38.600000
Cardinals88.3–73.7.54495253–43.552083
Brewers85.8–76.2.52991552–43.547368
Reds75.2–86.8.46435646–50.479167
Astros72.4–89.6.44714344–51.463158
Pirates70.4–91.6.43433544–50.468085

NL Westextrapolatedreal (at ASB)
Diamondbacks83.0–79.0.51253847–48.494737
Dodgers79.7–82.3.49187146–49.484211
Giants69.7–92.3.43003540–55.421053
Rockies68.5–93.5.42290039–57.406250
Padres68.0–94.0.41953137–58.389474

For example, the Blue Jays have won 47 games so far and have 67 post-All-Star-break games.  I multiplied 67 by their Nostradamian record, .537618, to get 36.0, the number of those games they can expect to win, giving them an extrapolated final record of about 83–79.  Of course, this approach doesn't account for roster changes (such as injuries and the Sabathia, Harden and Blanton trades) or differences in schedule, but it beats assuming that winning percentages won't change.

So the Red Sox, White Sox, Cubs and Diamondbacks are likely to extend their division leads, while the Angels are likely to lose most of theirs and the NL East should remain very tight.  The Braves have been so unlucky so far that they're projected to finish fourth even though they have the best Nostradamian record in their division.

coolstandings.com, using different assumptions than I do, calculates not likely final standings but the rough probability that each team will make the playoffs.

15 July 2008

The breaks

The All-Star break seems like a good time to evaluate the 2008 baseball season so far.  Are the Angels and Cubs really the best teams, indicating a possible Taking Care of Business World Series matchup?  Are the Rays a fluke or a genuinely good team?  Are the Padres and Indians really as bad as their win-loss records?  Or has luck shaped the current standings?

It helps to be lucky in baseball, not just in how a team's runs are bunched game to game but in how their hits and outs are bunched inning to inning.  Using my Nostradamian approach, here's what the standings would look like now if all the luck had perfectly evened out, with "deserved" wins and losses rounded to the nearest tenth:

AL Eastdeservedreal
Red Sox60.2–36.8.62080457–40.587629
Rays52.7–41.3.56090055–39.585106
Blue Jays51.1–43.9.53761847–48.494737
Yankees50.3–44.7.52957550–45.526316
Orioles45.2–47.8.48637545–48.483871

AL Centraldeservedreal
White Sox55.6–38.4.59107154–40.574468
Tigers48.6–45.4.51699947–47.500000
Twins45.3–49.7.47691653–42.557895
Indians43.7–50.3.46531941–53.436170
Royals42.2–53.8.43929143–53.447917

AL Westdeservedreal
Athletics53.7–41.3.56488051–44.536842
Rangers48.2–47.8.50189750–46.520833
Angels46.2–48.8.48620457–38.600000
Mariners40.5–54.5.42655937–58.389474

NL Eastdeservedreal
Braves53.4–41.6.56197045–50.473684
Mets52.2–42.8.54931451–44.536842
Phillies51.9–44.1.54048752–44.541667
Marlins46.4–48.6.48815950–45.526316
Nationals36.3–59.7.37850036–60.375000

NL Centraldeservedreal
Cubs56.9–38.1.59913557–38.600000
Cardinals51.3–44.7.53457953–43.552083
Brewers48.0–47.0.50516752–43.547368
Reds42.5–53.5.44281446–50.479167
Astros40.3–54.7.42443544–51.463158
Pirates36.4–57.6.38768144–50.468085

NL Westdeservedreal
Diamondbacks51.1–43.9.53777847–48.494737
Dodgers47.8–47.2.50273246–49.484211
Padres43.9–51.1.46214937–58.389474
Rockies42.9–53.1.44711939–57.406250
Giants42.1–52.9.44277040–55.421053

So far, the luckiest 2008 teams are the Angels and Twins, who both deserve to have losing records, and the Pirates, who deserve an even worse record than they have.  The unluckiest teams are the Braves, who deserve to be atop their division instead of in 4th, and the Padres, who deserve to be 3rd rather than last.  The AL East is certainly the strongest division—it has the 1st, 6th, 10th, 12th and 18th best teams in baseball.  The NL West is the weakest but it's also by far the least lucky; the NL Central is almost as weak but has been the luckiest.  Some think the AL West race is over, but the Athletics have played well enough to have a solid lead in what is really a weak division.  They trail only because of results like their last three-game series: they outscored the Angels 13–10 but lost two of three.  The Rays have been a little lucky but are a genuinely good team, which is no surprise to some.  The Nationals are truly the worst team, and given the strength of the NL East, they probably have the lowest chance of making the playoffs.  Based on this analysis, the Red Sox and Cubs seem likeliest to make it to the Series.

The Wikipedia article on Pythagorean expectation now calls my Nostradamian approach "second-order wins" and even recommends using Base Runs, as I do.  Cool.