Lies, damn lies, and statistics

Jaime Garcia was pulled in the fifth inning last night, which marked the fifth consecutive game - all of them in the NLCS - in which the St. Louis starter did not pitch more than five innings (and only in Game Three, when Chris Carpenter went five exactly, did they even hit that mark). As a result, the talking heads have been abuzz about the fact that no team has ever won a postseason series when its starters have failed to pitch more than five innings in any of the first five games.

#1: This isn't too surprising, since in MOST cases this would indicate that the starters were getting absolutely shelled.

#2: It's not like this happens very often.

With the exception of their six-spot that chased Garcia in the fifth inning of Game One, the Brewers haven't scored more than two runs in any inning in the series, and in Games Two through Five they've topped out at four runs. The ERA for the Cardinals' starters is not good, mostly because they aren't pitching a lot of innings, but check it out:

Cardinals starters: 22.1 IP, 15 ER, 6.04 ERA

Brewers starters: 27.2 IP, 19 ER, 6.18 ERA

And that's with Randy Wolf allowing just 2 ER in 7 IP in the Game Four win. The other four games, Zack Greinke (x2), Shaun Marcum and Yovani Gallardo have combined to go 20.2 innings (barely over five a game) and allow 17 earned runs for a combined ERA of 7.40.

So why are the Cardinals winning? Because neither team's starters are pitching well, but the Cardinals' are pitching slightly better, in spite of LaRussa's quick hook. Really, this is a total non-story, and if Carpenter gets one more out in Game Three there isn't even anything to talk about.

And, of course, because St. Louis' bullpen has thrown 21.1 IP in this series and allowed four earned runs, whereas Milwaukee's bullpen has thrown 15.1 IP and allowed 9 ER. 1.69 bullpen ERA vs. 5.28 bullpen ERA... hmm. I wonder how they're doing it?

The irony is that St. Louis' bullpen was not very good this year. Really, St. Louis didn't pitch that well in general, finishing 8th in the NL in ERA - there's a reason they were the NL's highest-scoring team and yet barely snuck into the playoffs. But just ask the 2005 White Sox how scalding-hot pitching (even if it is mostly your relievers in this case) can carry you in October.

By the way, the last team to have its starters go five innings or less in each of the first five games of a series? The 1984 Padres, in the World Series against Detroit, which they lost in five games. The five starters (Ed Whitson, Tim Lollar, Eric Show, and Mark Thurmond twice) went a COMBINED 10.1 innings in the series, with Whitson (in Game Two - which San Diego still won) and Thurmond (in Game Five) both getting yanked in the first inning, Lollar failing to make it out of the second in Game Three and Show getting pulled in the third in Game Four. Only Thurmond's Game One start saw a starter complete the fifth. The combined ERA for those four starters over five games? 13.94. St. Louis' starters are doing just a LITTLE better this series.

My fellow nerds and I will retire to the nerdery with our calculators

Earlier today, I saw the following piece of trivia on ESPN.com's mobile site: "Derrick Rose is the first Chicago Bulls player to score 40 points in a game against the Spurs since Michael Jordan."

My first thought was: not bad.

My second thought was: how many Bulls players have scored 40 points in a game since Jordan left at all?

As it turns out, not very many. The Bulls, from the start of the 1998-99 season until now, have had just ten regular season games in which one of their players has scored 40 points. Rose's is one; Jalen Rose, Elton Brand and Luol Deng had one each; Jamal Crawford had two; and Ben Gordon had four. That's the entire list. Ten games, six guys. So it seemed like the fact that Rose was the first Bull to score 40 on the Spurs since Jordan wasn't too surprising given that very few Bulls since Jordan had scored 40 on anybody. (To put it another way, in Jordan's final 1997-98 season alone, he scored 40 points or more twelve times, plus twice more in the playoffs. Even adding the playoffs to every 1998-99 to 2010-11 Bull only gets them to eleven.) I posted as much on Facebook.

But then I started to wonder if that was unfair in a way, because 40 points is a pretty rare mark to hit unless you're a really elite scorer. In other words, it seemed fairly likely that most teams' records over the last thirteen seasons wouldn't look that much different from the Bulls', especially for teams that rarely had elite scorers and didn't spend the entire thirteen seasons or close to it with playoff caliber teams (as the Bulls were pretty lousy for almost half that span).

And so, thanks to the Basketball-Reference play index, I came up with the following numbers:

Most 40-point games: Lakers, 129

Not terribly surprising - 106 of those games belong to Kobe Bryant, and he also owns the highest-scoring single game in that span, with the infamous 81-point game.

Second-most 40-point games: 76ers, 70

All 70 belong to Allen Iverson. The 76ers have the second-most 40-point games total but are also the only team with just one player to score 40 in a game for them during the span.

Other teams with more than 40: Cavaliers 49, Wizards 43

42 of Cleveland's 49 belong to LeBron James, unsurprisingly. Gilbert Arenas leads the Wizards with 28, but 8 of the 43 are Michael Jordan - meaning that he almost had as many 40-point games as anyone who played for the Bulls even though he was at the end of his career and kind of washed up.

Mean 40-point games: 26.4
Median 40-point games: 20

129 is sort of an extreme outlier, so I also did the median, which gives you a more realistic number. In other words, the Bulls had half as many 40-point games as an "average" franchise during this span.

Teams with fewer 40-point games than the Bulls: Pacers (9), Jazz (9), Grizzlies (7), Bobcats (6), Clippers (5)

The Bobcats haven't existed the whole time, so it's a little unfair to them. The Clippers' number of five is even more pathetic when you consider that Blake Griffin has two of them just this year - and no one else has more than one. So in a 12-season span, there were just three Clippers games in which someone on their team scored 40 points. That is awful.

Most different players to score 40: Warriors, 10

Not really surprising when you think about the Warriors' style of play. The Warriors also had the most total 40-point games (29) for a team that didn't have any one player reach double figures (Monta Ellis had the most with seven).

Players to lead more than one team in 40-point games: Tracy McGrady (30 for Orlando, 15 for Houston) and Vince Carter (17 for New Jersey, 14 for Toronto)

McGrady had a 62-point game with the Magic, the highest single-game output during the period of anyone not named Kobe.

Is this the kind of thing that's only interesting to me and Joe Posnanski? Probably. But I'm posting it anyway, because what else am I going to do with it?

Here are the stats for every team, since I already bothered to look them up.

Team / 40 pt Games / Different players to score 40 / Player with most / Player with highest scoring game

Lakers / 129 / 4 / Bryant, 106 / Bryant, 81
76ers / 70 / 1 / Iverson, 70 / Iverson, 60
Cavaliers / 49 / 5 / James, 42 / James, 56
Wizards / 43 / 7 / Arenas, 28 / Arenas, 60
Thunder / 36 / 6 / Durant, 17 / R. Allen, 54
Heat / 36 / 4 / Wade, 31 / Wade, 55
Nuggets / 34 / 6 / Anthony, 19 / Iverson, 51
Magic / 34 / 4 / McGrady, 30 / McGrady, 62
Suns / 30 / 9 / Stoudemire, 15 / Delk, 53
Warriors / 29 / 10 / Ellis, 7 / Jamison, 51
Nets / 27 / 5 / Carter, 17 / Carter, 51
Rockets / 25 / 7 / McGrady, 15 / McGrady, 48
Raptors / 25 / 4 / Carter, 14 / Carter, 51
Knicks / 24 / 8 / Marbury, 5 / Houston, 53
Celtics / 21 / 2 / Pierce, 19 / Pierce, 50
Bucks / 19 / 6 / Redd, 11 / Redd, 57
Pistons / 19 / 4 / Stackhouse, 10 / Stackhouse, 57
Mavericks / 19 / 4 / Nowitzki, 16 / Nowitzki, 53
Kings / 16 / 4 / Kevin Martin, 6 / Webber, 51
Blazers / 13 / 7 / Roy, 4 / D. Stoudamire, 54
Hornets / 13 / 4 / Paul and West, 5 / Mashburn, 50
Spurs / 13 / 3 / Ginobili, 6 / Parker, 55
T-Wolves / 11 / 7 / Garnett, 4 / Garnett, 47
Hawks / 11 / 4 / J. Johnson, 6 / Abdur-Rahim, 50
Bulls / 10 / 6 / Gordon, 4 / Crawford, 50
Pacers / 9 / 5 / Granger, 4 / J. O'Neal, 55
Jazz / 9 / 5 / K. Malone, 3 / Millsap, 46
Grizzlies / 7 / 5 / Gasol and M. Miller, 2 / M. Miller, 45
Bobcats / 6 / 3 / G. Wallace, 4 / S. Jackson, 43
Clippers / 5 / 4 / Griffin, 2 / Griffin, 47

The Amazing Rays

You might not know it, but the Rays and Rangers are making history with their first-round series, one which hopefully leaves the winner with Yankee-killing momentum. How so, you ask?

There have been 96 best-of-five series in baseball history prior to this year, between the League Championship Series from 1969 to 1984, the 1981 Division Series, and the modern Division Series from 1995 until 2009. Of those 96, only 26 - 27% - went the full five games. And of those 26, ten - 38% - saw one team go down 2-0, then rally back to force a fifth game. They were:

1972 ALCS: Oakland 3, Detroit 2
1981 NLDS: LA Dodgers 3, Houston 2
1981 NLDS: Montreal 3, Philadelphia 2
1981 ALDS: NY Yankees 3, Milwaukee 2
1982 ALCS: Milwaukee 3, California 2
1984 NLCS: San Diego 3, Chicago Cubs 2
1995 ALDS: Seattle 3, NY Yankees 2
1999 ALDS: Boston 3, Cleveland 2
2001 ALDS: NY Yankees 3, Oakland 2
2003 ALDS: Boston 3, Oakland 2

A couple of interesting things here. First, in not all of these cases did the team rallying to force a fifth game win the series. The '81 Brewers and Phillies, along with the '72 Tigers, were able to force a fifth game but still lost the series. The other seven, however, all won three straight. That bodes well for the Rays, who evened up their series 2-2 with their second win in Texas today. History gives them a 70% chance of winning on that alone.

Consider this, though: prior to 1998, best-of-five series formats went 2-3, with the team having home field getting the last three games. So seven of the ten series above were played like that, and of those seven, four involved the home team winning all five games - i.e. the winning team going down 2-0 but then winning three straight at home (the '81 Dodgers, '82 Brewers, '84 Padres and '95 Mariners). So in some respects we can't really judge based on these, because series are now played 2-2-1.

And that is really where we get into history. Prior to now, only two teams ever opened a five-game series with two road wins, then lost the next two games at home. They were the '01 A's and the '81 Yankees.

The '81 Yankees won the first two games of their series in Milwaukee, then had three straight at home in which to clinch it - they dropped the first two, but won Game Five.

The '01 A's won the first two games of their series in New York, but after dropping Games Three and Four in Oakland, they had to go back to New York for Game Five... which they lost.

In other words, there has never been a five-game series in which the road team won all five games. The only analogous series to what the Rays/Rangers have done is the '01 ALDS, and if that "pattern" holds, the Rays will win Game Five. If the Rangers win, not only will they be the first team to win a five-game series in which the road team won every game, but they will be just the third team (after the '72 A's and '81 Expos) to take a 2-0 lead in a five-game series, blow the lead, but then win the fifth game on the road to take the series anyway (and just the fourth team to come back to win after letting the series go from 2-0 to 2-2).

So things would seem to look pretty good for the Rays (except for the part where they have to face Cliff Lee in Game Five). One other thing to consider, though. Out of the 26 series that went to five games, however they got there, the deciding Game Five has been won by the home team 14 times and the road team 12 times (including five of the last six). So overall, the chances of the home team winning Game Five are 54% - not much better than even.

Why We Fight

I imagine some of you wonder why I get so annoyed by bad MVP votes and the like. It's for the same reason why I don't think intelligent design deserves to be taught in public schools alongside evolution. We have scientific methods and data that can tell us things we're not always capable of guessing or perceiving on our own - but some people actively disdain these methods. Saying that people who believe in the importance of stats don't like or can't adequately appreciate baseball, as many anti-stat types do, is really no more intellectually valid than saying that a belief in evolution means I don't like or can't adequately appreciate the majesty of the earth. It's ridiculous. And, more specifically, because sometimes there is an MVP ballot so monumentally dumb that it suggests that its associated voter should not only be stripped of his right to vote for postseason awards, but dragged out into the street and beaten like Carlo Rizzi. Ladies and gentlemen, Tom Haudricourt of the Milwaukee Journal-Sentinel's ballot for 2008 NL MVP:

1. Ryan Howard, Phil
2. CC Sabathia, Mil
3. Manny Ramirez, LA
4. Carlos Delgado, NY
5. Aramis Ramirez, Chi
6. Prince Fielder, Mil
7. Albert Pujols, Stl
8. Ryan Ludwick, Stl
9. Ryan Braun, Mil
10. David Wright, NY

My first thought: Tom Haudricourt was awakened from a coma on July 1st.

1. Ryan Howard, Phil

I mostly went over this one yesterday. Ryan Howard is the MVP only if you believe one of the following three things:

(1) There is nothing more important than hitting a lot of home runs
(2) There is nothing more important than having a ton of RBIs
(3) There is nothing more important than getting hot in September

(3) is maybe, kinda, sorta true, although if you were absolute shit for the first three months of the season, and then again in August, I don't think a hot September should outweigh that. (1) and (2) are nonsense.

2. CC Sabathia, Mil

Sabathia pitched extremely well for the Brewers. He also did not pitch for them before July. Unless you put up full-season stats in half a season, you cannot be an MVP for half a season. There are no such things as extrapolation points.

3. Manny Ramirez, LA

See above. Ramirez played for the Dodgers for two months. He was a really, really good hitter during that time, but he still only played for them for two months. Mercifully, Ramirez finished "only" fourth in the balloting this year, suggesting that at least most writers were only so swayed by the "Manny has turned this team around!" narrative.

4. Carlos Delgado, NY

Remember the first half of the season, when Delgado was hitting .248 with a .328 OBP and everyone was talking about how washed up he was? Tom Haudricourt doesn't.

5. Aramis Ramirez, Chi

Dear God. At this point there can be no doubt that Haudricourt is focused primarily on team accomplishments; Ramirez had a nice season, of course, but it doesn't compare with Albert Pujols' even if you love counting stats.

6. Prince Fielder, Mil

Decent year. Better year than Albert Pujols?

7. Albert Pujols, Stl

And finally we get to the real MVP. The guy who Tom Haudricourt does not even think was a top five most valuable player candidate because his team was mediocre. Tom Haudricourt was voting for the MVPOATTAMTPOALGCTTCDBOEHCDGOHB Award, the Most Valuable Player on a Team That Also Made the Playoffs or At Least Got Closer Than the Cardinals Did Because Otherwise Explain How Carlos Delgado Got on His Ballot Award.

Who's ready for some stupid justification?

I had an MVP ballot and voted for Howard first because he almost single-handedly carried the Phillies to the playoffs by batting .352 with 11 homers and 32 RBI in September. I like to weight my voting to teams in the playoff hunt because I think that puts more pressure on players and separates the men from the boys. There's little pressure on players having big years if their teams aren't playing for anything at the end.

First of all: no. Okay, Howard had 32 RBI in September. Even if you give him full credit, as an individual, for driving in all of those important runs, the Phillies scored 138 runs. That's less than a quarter that were driven in by Howard. Maybe the other 77% of runs that were knocked in were also kind of important? Also, maybe if Howard hadn't been such a non-factor for four months, he wouldn't have needed to have such an enormous September just to "carry" the Phillies to the playoffs. Maybe?

With the Cardinals finishing fourth, I voted Pujols seventh on my ballot. I don't consider MVP to be "the most outstanding player" award and therefore don't just go by who had the best stats. I like to credit players for lifting their teams to the post-season or at least keeping them in the race until the very end.

At least he, unlike Thomas Boswell, admits that Pujols had the best stats. Then he goes ahead and ruins it by pretending that individual players are capable of "lifting" otherwise mediocre teams to the postseason. Remove Pujols from the Cardinals and they lose 13 wins if you just stick any AAA guy in his place; rather than finishing four out of the wild card - a respectable if not successful season given how low the team's expectations were in March - they finish seventeen out of the wild card and drift behind the Reds into fifth place in the Central. Why Pujols is to be punished because the rest of his team wasn't that good is beyond me. Certainly writers like Haudricourt do not seem to be able to grasp the idea that elevating any team's performance by 13 wins is "value," and it's not his fault that the Cardinals had a shitty bullpen that blew literally dozens of leads, and had it blown five fewer the Cardinals would have made the playoffs and guys like Haudricourt would be lining up to tell us how amazing a season Pujols had. But because Jason Isringhausen blew five saves that turned into losses, Pujols finishes seventh on Haudricourt's ballot. Are you capable of seeing why this makes no fucking sense? At all? Pushing the guy with the best stats to seventh on your ballot because his team's bullpen sucked is madness.

I understand that the Cardinals would not have been even close to the wild-card berth without Pujols, but I still like players who elevate their game in crunch time and lift their teams to new heights. And I thought Ryan Ludwick had just as much to do with keeping the Cards in the hunt as Pujols did. St. Louis did stay in the wild card race until mid-September, but mainly because the Brewers and Mets were gagging at the time.

I swear to God, if I ever run into Tom Haudricourt on the street, I am going to punch him in the face. Albert Pujols' on-base percentage was .462. With RISP he hit .339/.523/.678. In "late and close" situations he hit .314/.444/.600. In "high leverage" spots as defined by WPA, he hit .392/.492/.725. For the month of August, as the Cards were desperately trying to stay afloat in the playoff race, he hit .398/.491/.745. Oh, but I guess he only hit .321/.427/.702 in September. What a fucking choke artist. (Ryan Ludwick: Nice season, especially considering. But he gives up 87 points of OBP to Pujols, and was no doubt aided by the fact that he spent half the season hitting in front of Pujols, giving him better pitches to hit, and the other half hitting behind Pujols, boosting his RBI chances since Pujols was on base twice a game.)

Also, none of this explains why Carlos Delgado finished three spots above Pujols, considering that he was teammates with David Wright, Carlos Beltran and Jose Reyes, all of whom are at least as valuable as Ryan Fucking Ludwick, and his team also failed to make the playoffs. If Pujols didn't elevate his game in the last two months it's only because he played at a ridiculously high level all season. So what Tom Haudricourt is saying is: Playing great all year is less valuable than playing shitty for half the year and great for two, three months tops. And also, playing great for a team that doesn't have a lot of other great players is somehow less valuable than playing pretty well for a team that does have a lot of other great players. Jesus Christ. This is the dumbest shit ever. Tom Haudricourt wrote this down and thought, "This will make sense when people read it."

It's a subjective vote and every writer has his own preferences. That's why I voted for Sabathia second and Ramirez third because even though they played in the league only half a season they were primarily responsible for putting their teams in the playoffs.

Some preferences may be stupider than others. Like yours, for example. Also, don't lean on subjectivity as an excuse. The vote should not be nearly as subjective as you've made it. And when you put two other Brewers in your top ten, that damages the argument that Sabathia was "primarily" responsible for putting the Brewers into the playoffs. How many guys can be that valuable for a team that slipped in by the skin of its teeth (and then in large part because their final weekend opponent had long since clinched)?

I voted Fielder higher than Braun because Fielder had a much better September when the Brewers were clawing to get in the playoffs. Braun was ailing, as we discovered, and did have the homer that put the Brewers in the playoffs, but I just felt Fielder did more down the stretch.

I know in this last paragraph he's comparing Fielder and Braun. But remember that he voted Fielder 6th and Pujols 7th as well.

Prince Fielder, September 2008: .316/.398/.600
Albert Pujols, September 2008: .321/.427/.702

Prince Fielder's team, September 2008: 10-16
Albert Pujols' team, September 2008: 12-13

Prince Fielder: more valuable than Albert Pujols.

Tom Haudricourt: IQ of a can of garbanzo beans.

Post-election stats wonkery

I'm a big nerd. So here are some stats I found interesting relating to the election:

The most Democratic county in the nation was Prince George's County, Maryland. Fully 89.1% of the vote in PG went to Obama, beating even the 88.7% of Shannon County, South Dakota, which was the most Democratic county in the nation in the 2004 election. (92.9% of the District of Columbia voted for Obama, but that's not really a county since it's not a segment of a state.)

The only state without a Democratic county was Oklahoma. The New York Times, where I got the stats, doesn't have county-by-county breakdowns for Alaska, but among the lower 48 and Hawaii, only Oklahoma didn't have a single county turn blue. By contrast, six states - Hawaii, Massachusetts, Connecticut, Rhode Island, Vermont and New Hampshire - didn't have any red counties.

The county map still shows mostly red, but that doesn't matter. After the 2004 election, I recall seeing a map showing how most of the counties in America were red and how this was somehow proof that the Republicans were dominating the country. While a look at the '92 and '96 electoral county maps does show that many more counties are red now than were red even just 12 years ago, it's kind of important to know which counties are red. And the fact is that most populous counties went to Obama. Of the 49 states for which the Times has county results, Obama won the most populous county in 40 of them, and often by huge margins: he won Los Angeles County by 40.5%, Denver County by 52.1%, Cook County by 53.2%, Orleans Parish by 60.1%, Kings County (Brooklyn) by 58.7%, and Philadelphia County by 66.7%. All told, of his 40, Obama won 25 of them by at least 20 points. McCain's nine were the biggest counties in Arizona, Idaho, Kansas, Montana, Oklahoma, South Carolina, Utah, West Virginia and Wyoming; only two of them he won by more than 20 points (Greenville County, SC and Laramie County, WY). Even in some of the states that Obama lost, he won the biggest counties by sizable margins, including Orleans in Louisiana; Hinds County, MS (which he won by more than 40 points); Shelby County, TN (27.6 points); St. Louis County, MO (almost 20 points); and Fulton County, GA (35 points).

What's more, of the 50 most populous counties in the entire country (which, oddly, is the same as the list of counties with 900,000 people or more), Obama won 46. The only four he didn't? Maricopa County, AZ (home to McCain's base of Phoenix), Orange County, CA (rich people), Tarrant County, TX, and Salt Lake County, UT (reliably Republican Mormons, although McCain won this county with less than 50% of the vote and by just 0.5 points). Obama also won many of the 46 quite handily, taking at least 60% of the vote in 25 of the 46 and more than 70% of the vote in ten of them. Even in McCain's most commanding win of his four, Tarrant County, he took just 55.6% of the vote.

And even besides that, Obama's most motivated counties were more populous than McCain's in most states. In fact, the county that gave Obama the highest percentage of its vote among the counties in its state was often the largest or one of the few largest in its state (this includes Cook in IL, Wayne in MI, Philadelphia in PA, Bronx in NY, Suffolk in MA, San Francisco in CA, Multnomah in OR, King in WA, Ramsey in MN, Essex in NJ, Cuyahoga in OH, and Shelby in TN), whereas McCain's most motivated supporters were usually found in more rural and/or less populated counties. In only seven states was McCain's most motivated county more populous than Obama's, and only in Hawaii - where the most "staunchly" McCain county, Honolulu, gave him a full 29% of the vote - was the county that gave its biggest percentage of the vote to McCain the biggest county in its state.

The greatest percentage of votes in one county was 93.2%, going to McCain in King County, Texas. I love this one mostly because of how few voters there were in King County (the third-smallest county by population in the US). 151 people voted for McCain, and eight voted for Obama. Eight??? I love it. Who are these eight people and why do they live there?

Nearly every state got more Democratic. Indiana went from +21 GOP to +1 for the Democrats, amazingly. And it's not just the ones that switched - already blue states got bluer. California, for example, went from +10 for Kerry in 2004 to +24 for Obama. Hawaii went from +9 for Kerry to +45 for Obama. Even Kerry's home state of Massachusetts went from +25 to +26. The only states that got redder? Arkansas (+10 to +20), Louisiana (+15 to +19), and Tennessee (+14 to +15). Oklahoma and West Virginia stayed the same.

Obama got 52.6% of the popular vote. It's the most of any candidate since Bush I in 1988, and the most for a Democrat since Lyndon Johnson in 1964, also the last time the Democrats won Indiana and Virginia. Of course, this part you probably already knew. I bring it up more to point out how hilarious it is that conservatives have jumped right on the "This isn't a mandate for the Democrats" wagon. This is the same party that bragged about Bush's "political capital" after he won 2% less of the popular vote and almost 80 fewer electoral votes than Obama got this year. (In case you didn't buy that the results of the election aren't a mandate for Obama, some conservatives have started to argue that he's "center-right" and that his policies resemble Dwight Eisenhower's. Anything to avoid the fact that this election was an absolute destruction of the Republican party.)

January 20 can't come soon enough.