Championship series - objective ranking wrap-up

I ran the Bill James Playoff Predictor for the Division Series and the Championship Series games. I also looked at the Baseball Prospectus objective ranking system. Here's the wrap-up for the post-season so far. World Series post later today or tomorrow...

Bill James Playoff Predictor was 1-1 in the Championship Series round.

Right:
Boston (100) over Cleveland (39)

Wrong:
Arizona (77) lost to Colorado (30)



Baseball Prospectus' "Secret Sauce", which was 1-3 in the DS round, was 1-1 in the CS round.


Right:

Boston (7.5) over Cleveland (48)

Wrong:

Arizona (29) lost to Colorado (57)


For the post-season so far:

• Bill James Playoff Predictor
  Boston (79) beat Los Angeles of Anaheim (33)
  Arizona (92) beat Chicago (33)
  Boston (100) over Cleveland (39)
  
  Wrong:
  New York (83) lost to Cleveland (60)
  Arizona (77) lost to Colorado (30)
  
  Tie (essentially):
  Colorado (58) beat Philadelphia (59)




• Baseball Prospectus' "Secret Sauce"
  Right (lower is better):
  Boston (7.5) beat Los Angeles of Anaheim (27.5)
  Boston (7.5) over Cleveland (48)
  
  Wrong:
  New York (32.5) lost to Cleveland (48)
  Arizona (29) beat Chicago (22)
  Colorado (57) beat Philadelphia (49)
  Colorado (57) over Arizona (29)

ALCS - thoughts on a 3-1 deficit

Cleveland's 7-3 win has the Indians on the verge of going to the World Series and the Red Sox on the verge of golfing.

And it has "idiot nation" up in arms, calling for the firing of the only manager to win a World Series for the Red Sox since Woodrow Wilson was in the White House. It is, of course, preposterous, but that's the way it goes around here.

• I just heard one of the dumbest comments ever made by a Francona bashing caller to WEEI. "He made a mistake bringing in Papelbon in game 2, because then he had no one left in the 11th."
  
  Who, exactly, does this nimrod think Francona SHOULD have brought in in the 9th? Gagne? Lopez? Lester? They pitched so well later... I suppose that would have resulted in earlier bedtimes, but it's tough to see how it would have improved Boston's chances of winning that game.
  
  Once you reach the 9th inning of a tie game, especially at home, you should be using your pitchers in DECREASED order of effectiveness. You don't save your best for later, because then you are far more likely to lose with your best pitcher still sitting in the bullpen. Barring significant match-up issues, every time you make a change, you go with the best pitcher you've got left, because that increases the chances of you scoring before you give up a run.


• Francona managed game 2 just fine, he managed game 3 just fine, he managed game 4 just fine. The players haven't gotten the job done - that's life - it happens. It's baseball. The idea that this 7 game series tells us anything about these teams that the 162 games of the regular season didn't is laughable. It determines a "champion" but it doesn't pick the best team - just the one who plays the best in 4 of 7. It is frustrating, and we would much rather see them win than lose, but the other team gets paid, too, and sometimes that's just what happens. The Red Sox certainly aren't out of this series, but the odds of them winning are not good. Whether they come back and win, or go down in game 5, this is an excellent team, with a good manager, that had a very good year. Winning the World Series is special and important, but the post-season is a crapshoot - the organization has to think in terms of putting together a team that can get there consistently, because that's how you win World Series - you get to the post-season and have a good run at the right time. Going up against a team that has that good run at the right time while you don't is no reason to fire the manager, if he isn't the proximate cause, and Francona certainly isn't, the vitriol aimed at him notwithstanding...


• I'd probably give Ellsbury a shot in center, but I don't have a problem with him not doing so. I remember a lot of abuse heaped upon Francona in 2004 for not replacing Damon and Bellhorn in the lineup while they were losing to the Yankees. That worked out. It might work out this time, and it might not, but the idea of making decisions right now based on a 3 game slump and/or losing streak is just not rational. You knew, going into the playoffs, what your best lineup consisted of - that's what you play. Period. To do otherwise is to react irrationally to a pathetically small sample of data.


• Gagne has taken a lot of grief, too. As has Drew. But neither has been the biggest problem, and no one's blaming him. Many players have failed - certainly Gagne and Drew (and Lugo and Varitek and Pedroia and Lopez and Lester and Schilling and Matsuzaka) have all failed to make significant positive contributions. The one player who has had the single biggest NEGATIVE contribution in the series thus far is Manny Delcarmen. They brought him in with a lead in the 6th inning of game two. He promptly gave up the tying run, and no one scored again until Cleveland beat Gagne, Lopez and Lester in the 11th. He came in with a 3-run deficit in game 4, and promptly gave up a 3-run HR, and another one to boot. The Red Sox scored 3 runs in the 6th. If Delcarmen does his job in game 2, the Red Sox leave Boston with a 2-0 series lead. If he does it in game 4, they're tied after 6.

League Championship Series - Objective predictors

I'll have more on the Boston-Cleveland series over the next couple of days, but I thought I'd look at a couple of objective statistical playoff prediction systems on the championship series. The Bill James Playoff Predictor is a toy, an exercise in curve-fitting that probably doesn't have much value. The Baseball Prospectus "Special Sauce" is also a curve-fitting exercise, with more in-depth sabermetric stats, and they think that it has some predictive value. (Follow the link below for more information.) But they're both objective - there's no opinion here, just plugging in of numbers into the formula.

Here's how the Bill James Playoff Predictor sees the League Championship series:

ALCS

1. 1 pt to the lead team for each half-game in the standings (EVEN)
2. 3 pts to the team that scored more runs (BOS - 3)
3. 14 pts to the team with fewer doubles (CLE - 14)
4. 12 pts to the team with more triples (BOS - 12)
5. 10 pts to the team with more home runs (CLE - 10)
6. 8 pts to the team with the lower team batting average (CLE - 8)
7. 8 pts to the team that committed fewer errors (BOS - 8)
8. 7 pts to the team that turned more double plays (CLE - 7)
9. 7 pts to the team that walked more batters (BOS - 7)
10. 19 pts to the team that had more shutouts (BOS - 19)
11. 15 pts to the team whose ERA was lower (BOS - 15)
12. 12 pts to the team that has been in postseason most recently or
    went further (BOS - 12)
13. 12 pts to the team that won season series (BOS - 12)

BOS - 100, CLE - 39


NLCS

1. 1 pt to the lead team for each half-game in the standings (ARI - 2*)
2. 3 pts to the team that scored more runs (COL - 3)
3. 14 pts to the team with fewer doubles (ARI - 14)
4. 12 pts to the team with more triples (ARI - 12)
5. 10 pts to the team with more home runs (TIE)
6. 8 pts to the team with the lower team batting average (ARI - 8)
7. 8 pts to the team that committed fewer errors (COL - 8)
8. 7 pts to the team that turned more double plays (COL - 7)
9. 7 pts to the team that walked more batters (ARI - 7)
10. 19 pts to the team that had more shutouts (ARI - 19)
11. 15 pts to the team whose ERA was lower (ARI - 15)
12. 12 pts to the team that has been in postseason most recently or
    went further (0)**
13. 12 pts to the team that won season series (COL - 12)

ARI - 77, COL - 30

* - The final standings show Arizona 1/2 game ahead of the Rockies, because Colorado played an extra game to get into the post-season. That game counts as a regular season game, but I'm not including it.

** - I've made the subjective judgement that the Diamondbacks 2002 trip to the post-season, with a different front office, different manager and (almost entirely) different players is no more relevant than than Colorado's 1995 trip.

In the two division series where the point ratio was greater than 2:1, the system was correct. That is the case for both of these series.


Here's how the Baseball Prospectus "Secret Sauce" rankings see the series. (Overall ranking is the sum of rankings in various pitching/defense categories, so lower is better)

ALCS

Boston (7.5) over Cleveland (48)

NLCS

Arizona (29) over Colorado (57)

So both of those objective systems point strongly towards Boston-Arizona in the World Series.

Odds and ends...

A couple of thoughts as the regular season winds down...

• Yankee newsgroup trolls have mocked the Red Sox every time they've celebrated reaching the post-season via Wild Card. I expect to hear less of that, given the champagne celebration the Yankees had last night on clinching the play-offs.



• The first round is almost certain to be Anaheim at Boston and New York at Cleveland. The Red Sox and Indians are 3 games ahead of the Yankees and Angels. If the season were to end this way, Boston would be the 1st seed because of head-to-head record vs. Cleveland, but the Red Sox and Yankees cannot, by rule, meet in the first round.



• I actually feel sorry (well, more sorry than usual) for Tampa Bay fans this morning (if there are any.) This is their 10th Major League season, and they've yet to win more than 70 games in any of them. They're in the most disheartening division in baseball, because they've got two of the uber-teams, teams that have both tremendous baseball acumen and massive financial resources blocking their path to the play-offs. Both of those teams play in front of essentially home crowds in the Rays' own ballpark. And now, in the span of five days, they've had to watch both of them clinch and celebrate on their own field.
  
  That stinks (more than usual) for the Devil Rays fans (if there are any.)



• There is a perception in certain quarters that the Red Sox were great early, but have struggled since. Ask anyone who has played better since the All Star break, and they'll probably list all of the other AL play-off teams, and maybe someone else.
  
  It is true that New York and Cleveland have better records since the All Star break than Boston. But they're the only AL teams for whom that's true. And if you look at run differential and pythagorean records, something interesting pops out.
  
  The Red Sox were not only the best team in the AL before the All Star break, they've been the best team in the AL since the All Star break. They've just underperformed (or, to coin a phrase, "Gagned") their projection by 5+ games while the Yankees and Indians have exceeded theirs by 3 each. From a run differential point of view, only the Yankees, who have outscored their opposition by 100 runs, have come close to the Red Sox 110 run margin. LAA's 55 is third, Cleveland's 42 is tied for fourth. Bob Ryan of the Boston Globe wrote a piece last week about the Red Sox being the fourth best of the AL play-off contenders. That's just silly.



• Anyone who believes he knows what is going to happen in the play-offs is lying to himself. Anyone who tells you that he knows what's going to happen is lying to you. These teams will have won within 3-5 of the same number of games over a 162 games schedule. The idea that anyone knows what's going to happen in a 5-game series based on that is just silly. That's the way baseball works. There is too much luck involved in any given game for a short series to mean anything. Come Sunday evening, there will be 8 teams left. The best one, Boston, has essentially the same 12.5% chance of winning the World Series as the worst one, whoever that is (Cubs? Brewers? Padres? Rockies? I don't know.) No one should be shocked or even surprised to see any given team win or lose any given series.
  
  And don't listen to anyone who later tells you, "I knew that was going to happen."
  

  One of the most systematic errors in human perception is what psychologists call hindsight bias -- the feeling, after an event happens, that we knew all along it was going to happen. Across a wide spectrum of issues, from politics to the vagaries of the stock market, experiments show that once people know something, they readily believe they knew it all along.

  
  Shankar Vedantam's article is about Iraq war opposition, but the hindsight bias comment is obviously relevant to baseball play-offs...
  
  So, you want my prediction for who will win the World Series? OK, here it is. The World Series winner will be...one of the eight teams that make the playoffs.