2007 Astros Related Baseball Blog

September 27, 2007: Ed Wade: Good for the Astros?
September 20, 2007: The 500 HR Club and the HoF: Jim Thome, Frank Thomas and Sammy Sosa
September 13, 2007: Astros: What was Mr. Purpura thinking?
August 27, 2007: Looking Back at the 1991 Houston Astros: The Magnificent Five
August 20, 2007: Hall of Fame: Bagwell's easy case
Summer Break
July 2, 2007: Hall of Fame: Biggio vs. Joe Morgan
June 21, 2007: Chan Ho Park: Does he have a chance with The Astros?
June 18, 2007: The Astros and the Leadoff Hitter Spot
June 7, 2007: Woody Williams, Coors Field and Mr. Garner's 7th inning decision
June 3, 2007: Craig Biggio: Limping to 3,000?
May 24, 2007: Observations on the 2007 Astros, after 1/4 of the season
May 17, 2007: The Press, the HR Record and Barry Bonds
May 10, 2007: Roger Clemens: Unusual finish
May 3, 2007: Woody Williams and the Astros: Why bring a flyball pitcher?

Ed Wade: Good for the Astros?

Date: September 27, 2007.

Last week the Astros hired Ed Wade as the new GM. Is it a good decision for the Astros? Evaluating GMs is a difficult task, unless their tenure is very long, due to small sample problems. But, I'll take a look at his GM record (98-05) in Philladelphia, along with the Phillies record in the three previous season (95-97) and the following two seasons (06-07).

Season	W-L (finish)	GB	RS-RA	ERA+	OPS+	Promoted	Memorable Move
1995	69-75 (5th)	21	615-658	102	92	-	-
1996	67-95 (5th)	29	650-790	98	91	-	-
1997	68-94 (5th)	33	668-840	87	90	-	-
1998	75-87 (3rd)	31	713-808	94	93	Bobby Estalella	Traded for Bobby Abreu
1999	77-85 (3rh)	26	841-846	94	101	Marlon Anderson	Signed Ron Gant
2000	65-97 (5th)	30	708-830	99	85	Pat Burrell	Traded C. Schilling
2001	86-76 (2nd)	2	746-719	103	97	Jimmy Rollins	Hired Larry Bowa
2002	80-81 (3rd)	21	710-724	91	113	Brett Myers	Traded Scott Rolen
2003	86-76 (3rd)	15	791-697	101	109	Chase Utley	Signed J. Thome
2004	86-76 (2nd)	10	840-781	98	107	Ryan Madson	Traded for Billy Wagner
2005	88-74 (2nd)	2	807-726	107	102	Ryan Howard	Hired Charlie Manuel
2006	88-74 (2nd)	12	867-812	100	107	-	-
2007	86-72 (2nd)	1	872-812	94	112	-	-

Reputation. From reading online and newspaper articles, I can summarize Wade "reputation."

1. Wade built a contender in Philly, mainly, by promoting players from the minors. My Evaluation: True
Wade received a bad team, that finished 5th in the NL East the three years before he became the Phillies GM. Philadelphia showed improved hitting during and after Wade's tenure in Philly, becoming an above average hitting team. Phillies pitching, on average, was average during and after Wade's tenure. Wade did promote minor leaguers to the majors, something that shows today in the Phillies lineup, with five Phillies regulars drafted or promoted during Wade's tenure. (Not including Shane Victorino, a Rule 5 draftee of Wade). Not a bad record. Wade's trade history is average, maybe marginally negative, but it does not look signficantly negative. (Good trades: Bobby Abreu, Willy Wagner. Bad trades: Schilling, Rolen.) With the exception of Jim Thome, Wade was able to attract good (Corey Liddle, Jon Lieber, Kevin Millwood, Kenny Lofton), but not top talent.

2. Wade cannot take a team to the "next level" -i.e., to the playoffs. My Evaluation: False
Well, the fact behind statement is true. But, a real evaluation should separate skill from luck. From this perspective, it's very difficult to judge the statement. Wade lived with the fantastic streak (and very good baseball teams) of the Atlanta Braves during his tenure. If winning the division is the yardstick, only Atlanta's GM did a good job in the NL East during Wade's tenure as GM. Wade's Phillies finished two games behind Atlanta in 2001 and 2005. Not that bad. In 2005, the Phillies missed the NL wildcard by one game. Difficult to separate distinguish a second and first position when the difference is so small. Statistically speaking, the difference can be just bad luck. BTW, in 2004, according to David Gassko, Ed Wade ranked number 5th as GM in 2004, when the Phillies finished ten games behind Atlanta. (John Schuerholz, Atlanta's GM ranked 13th.)

3. Wade's choices of managers are questionable. My Evaluation: True
Over Wade's tenure, following Bradford Doolite's system to rank managers, Phillies managers are on the negative side of added wins.

Good for Houston?: The Astros have a poor recent record of promoting players from the minors. The Astros have poor hitting. Wade's past history is good regarding these issues. But, we should be aware that Wade received a decent minor league system, something that the Astros do not have. Thus, to contend in the immediate future, the Astros will need to attract talent. The Astros will not expand payroll significantly, as the Phillies did during Wade's tenure as GM. In all likelihood, Wade will be able to attract decent (but not top) talent, as he did in Philly. Overall, Wade does not seem a bad choice. Based on past history, the only negative I see is the Astros need to find a new manager for 2008. Wade's past choices do not bode well for the Astros.

The 500 HR Club and the HoF: Jim Thome, Frank Thomas and Sammy Sosa

Date: September 20, 2007.

Since the Astros are no going anywhere this season and are not doing anything interesting, let's talk about something else: Jim Thome. (BTW, the Astros are still playing Biggio a lot in the top of the lineup... Why stop milking the cow, now?) Jim Thome's 500 HR started a minor debate: Is Jim Thome an automatic HoFer? There was an online article on ESPN.com, by the respected Rob Neyer, mentioning that he might need another 100 HRs to get to the HoF. Rob Neyer's article focused on MVP voting, not on production. I will analyze the distribution of the annual OPS+ of the 500 HR club members (451 seasons), along with the HOFers voted by the BBWAA members (1,355 seasons). The distribution of interest is the first one, since Jim Thome will be judged as an offensive player, a home-run hitter.

OPS+ Distribution	10%	25%	Median	75%	90%	95%	99%
500 HR Club member	97	125	150	169	175	212	258
HOF BBWAA-voted member	85	108	130	153	175	189	219

Next, I will place Jim Thome's production in the 500 HR Club member OPS+ distribution. I will also place recent 500 HR club member Frank Thomas, along with recent 600 HR Club member Sammy Sosa. (I only take into account seasons with more than 100 AB.) We should keep in mind that, with the exception of Mark McGwire, all the players with at least 500 homers who are eligible for the Baseball Hall of Fame have been inducted to the HoF by the BBWAA.

OPS+ Distribution

Player (Seasons)	OPS+	< 125	125-150	150-175	175+	Best (Year)
Jim Thome (16)	149	2	6	7	1	191 (02)
Frank Thomas (17)	157	2	5	3	7	212 (94)
Sammy Sosa (18)	128	10	4	3	1	201 (01)
Harmon Killebrew (17)	143	3	8	5	1	179 (69)
Reggie Jackson (20)	139	5	8	6	1	187 (69)

Jim Thome's numbers fit nicely in the OPS+ of the 500 HR club members. Jim Thome has eight seasons with an OPS+ under the median, and eight seasons with an OPS+ over the median. Thome has only two seasons with OPS+ under the 25% percentile of the 500 HR club distribution. Clearly, Jim Thome is an average member of the 500 HR club, with consistent production. While not in Frank Thomas'class, Jim Thome's numbers compare well and, as a matter of fact, are slightly better than Reggie Jackon's and Harmon Killebrew's, just to name two deserving HoFer. BTW, Frank Thomas, as Rob Neyer mentions in the above mentioned article, should be a lock for the HoF. (Frank Thomas has 7 seasons in the top 90% percentile of the 500 HR club distribution. Wow!)

Now, let me go back to Rob Neyer's argument. Rob Neyer argues that Jim Thome might need another 100 HRs to get to the HOF. This argument should apply to Sammy Sosa. By looking at the OPS+ numbers, Sammy Sosa is clearly not in the same class as Jim Thome. Sammy Sosa's OPS+ numbers are well under the median of the 500 HR club, with 14 seasons under the median -and 10 season under the 25% percentile. A simple t-statistic on Sammy Sosa's annual OPS+ confirms that his production is statistically significantly lower than 150 -i.e., the median OPS+ for the 500 HR Club members.

Given the recent backlash against Mark McGwire, it is safe to assume that, because of the steroid allegations, Sammy Sosa won't get into the HoF; which is good for the HoF. Otherwise, since BBWAA voters tend to emphasize overall career numbers and subjective measures (MVP, Gold Gloves, etc.) over consistent relative production, I am afraid that we may see Sammy Sosa in and Jim Thome out. This scenario would not reflect their respective values to their teams.

Astros: What was Mr. Purpura thinking?

Date: September 13, 2007.

Finally, on August 28, manager Mr. Garner and GM Tim Purpura were fired by the Astros. Although I am not a big fan of Mr. Garner, as a manager, I have mixed feelings about his firing. Mr. Garner's loyalty towards Biggio and other veterans -Woody Williams, Jason Lane, Brad Ausmus- throughout the season bothered me, but overall he understands the percentage game. Mr. Garner made some mistakes during games, may have subtracted one or two games over the win column, but his mistakes ended up as footnotes in this bad Astros season.

Tim Purpura, on the other hand, handed Mr. Garner a very weak roster, with poor hitting and below average pitching. Mr. Purpura should be considered the father of the 2007 creature. Let's go back to the 2006 season and post-season.

In 2006, the Astros finished 82-80, very much what was expected given their 2006 production. The Astros scored 735 runs and allowed 719 runs. Pretty even. Team OPS+ 93, Team ERA+ 111. Mediocre hitting, above average pitching. Purpura marginally changed the make-up of the team, bringing essentially the same average team back to 2007. Mr. Purpura did a couple of roster movements. Let's look back at them.

Hitting
Additions: Carlos Lee, Mark Loretta. Lee, a career 116 OPS+ producer. Last year, Carlos Lee had a 125 OPS+. Not a bad addition, but not a Jim Thome or Albert Pujols offensive blockbuster addition. Mark Loretta's career OPS+ is 99 -almost average. Last year, Loretta had an 82 OPS+, below average. But, nobody will consider Mark Loretta a terrible addition to any ball club, especially as a bench player. Both, Lee and Loretta, are considered below average defensive players.
Subtractions: Willy Taveras. Taveras played only two seasons before being traded to Colorado. His OPS+ numbers in those seasons 77 and 72. But, Taveras is considered an excellent defensive CF, with FRAAs of 17 and 22 during those two seasons.

Overall, Lee and Loretta have produced as expected, with 121 OPS+ and 87 OPS+ respectively. Also, both have produced below average defense as expected. Taveras was traded and this season has a 90 OPS+ and 13 FRAAs. Not a bad year for Willy Taveras in both offense and defense.Letting Taveras go was a bad bet, especially given his stellar defense and his speed. (The Astros had no legitimate CF going into the 2007 season and had nobody that can consistently steal a base if needed.) Hunter Pence, who was called up early in the season, has been a very nice, productive surprise. But, Pence was not in the 2007 plans of the Astros! Keep in mind that half the projected 2007 lineup was very weak: Biggio, Everett, Lane, and Ausmus, with 83 OPS+, 63 OPS+, 80 OPS+, and 53 OPS+, respectively, in 2006.

Pitching
Additions: Jason Jennings, Woody Williams. Jennings has a career 99 ERA+ (about average). His 2006 numbers were very good (127 ERA+), but his 2003-2005 ERA+ numbers were 93, 92, 94. Not earth shattering. On the positive side, Jason Jennings is a groundball pitcher, with a 1.21 GO/AO in 2006. Woody Williams, as stated in the first entry at the bottom of this blog, was not expected to do well in Houston.
Subtractions: Andy Pettitte (& Roger Clemens), Jason Hirsh. Andy Pettitte didn't have a great 2006, with a 108 ERA+, but he is a lefty, groundball pitcher, with a 1.89 GO/AO in 2006. Andy Pettitte was willing to stay at a discount, but Mr. Purpura refused to give him an option for a second year, as the Yankees did. His good friend, Roger Clemens followed him to the Yankees. Jason Hirsh was a valued prospect. So far, in his first season as a starter, Hirsh has been average (99 ERA+).

Overall, very difficult to understand why Pettitte was let go (and by association Clemens) and Jennings was brought to replace him. Taking the 2006 numbers as 2007 projections, one may think that replacing Pettitte for Jennings was not such a bad idea. But, one has to wonder if a good season (Jennings' 2006) is enough to justify that thinking. (Pettitte, with the exception of the 1999 season, has always been above average.) In addition, bringing Jennings, from Colarado, was costly: Taveras and Hirsh. Very difficult to justify. Bringing Woody Williams, who ended up replacing Roger Clemens, was baffling.

Evaluation
2007 Team OPS+ 95, 2007 Team ERA+ 94.While Lance Berkman's decrease in production (124 OPS+, relative to a 146 career OPS+) was unexpected, the productive season of Hunter Pence (124 OPS+) was also unexpected. I'd say a wash there. The mediocre team hitting was not unexpected with half the lineup consisting of Biggio, Everett, Lane and Ausmus, bad as expected (with 67 OPS+, 53 OPS+, 54 OPS+, and 67 OPS+, respectively, so far, in 2007. Ouch!). The drop in pitching was not unexpected. Firing Mr. Purpura was appropriate.

Looking Back at the 1991 Houston Astros: Ranking the Magnificent Five

Date: August 27, 2007.

Yesterday, I went to Jeff Bagwell's number retirement ceremony. Very moving. Bagwell was the Rookie of the Year in 1991 and had an elite career that, as pointed out in a previous entry, should land him in the HoF. But, looking back at Jeff Bagwell's rookie year, we have to admire the careers of five of the 1991 regulars: Craig Biggio (C); Jeff Bagwell (1B), himself; Ken Caminiti (3B); Steve Finley (OF); and Luis Gonzalez (OF). These guys played together for only 4 seasons, but I always think of them as a group.

I am going to take a look at the career numbers of these magnificent five members of the 1991 Houston Astros lineup. They produced a career average of 325 HRs, 230 SBs, and 2,420 hits. Not bad. And, perhaps more important for the players, they all played in at least one World Series (Luis Gonzalez and Steve Finley won it in 2001).

Player (MLB seasons)	Age in 91	Rookie Year	OBP (91)	OPS+ (91)	SB (SB%)	Hits	FRAA-season
Jeff Bagwell (15)	23	1991	.408 (.387)	150 (138)	202 (72%)	2,314	8.07*
Craig Biggio (20)	25	1989	.364 (.358)	112 (112)	414 (77%)	3,040	-6.65*
Ken Caminiti (15)	28	1987	.347 (.312)	116 (100)	88 (69%)	1,710	4.31*
Steve Finley (19)	26	1989	.331 (.332)	105 (113)	320 (73%)	2,548	-0.95
Luis Gonzalez (18)	23	1991	.368 (.320)	120 (116)	127 (60%)	2,486	4.74*

OPS+ Distribution	OPS+	100-120	120-140	140+	Best (Year)
Jeff Bagwell (15)	150	1	4	9	213 (94)
Craig Biggio (20)	112	7	6	2	143 (97)
Ken Caminiti (15)	116	1	5	3	173 (96)
Steve Finley (19)	105	10	3	0	135 (96)
Luis Gonzalez (18)	120	7	7	2	176 (01)

Note: * significantly different than average at the 5% level, using a t-test.

Jeff Bagwell is clearly the best of the bunch. Not a big surprise. With the exception of his last two years, when Bagwell's injured shoulder started to slow him down, Bagwell was, year in year out, among the elite offensive players in baseball. He had 12 great years (OPS+ of 130 or more), with a fantastic 1994. Recall that 130 OPS+ is the median OPS+ of all BBWAA inducted HoF members. Amazing offensive production. In addition, Bagwell's defense was above average, fielding a statistically significant 8.1 runs above the average CF.

Luis Gonzalez, Ken Caminiti and Craig Biggio are in the second group of excellence. They all have a decent number of very good years (OPS+ between 120-140, the center of the interval is 130), with a couple of great years (OPS+ over 140). Though clearly not in Bagwell's level of production, both Gonzalez and Biggio have had careers worthy of HoF consideration. (As pointed out in a previous article, I believe that Biggio will get in. I suspect that Luis Gonzalez, who has better OPS+ than Biggio, will not get in.) Biggio's stolen bases, not reflected in his OPS+, should help to consider his offensive performance over Ken Caminiti's and Luis Gonzalez's better OPS+ numbers. On defense, Ken Caminiti (4.3 FRAA per season) and Luis Gonzalez (4.7 FRAA per season) have a clear edge over Biggio (-6.7 FRAA per season). As the numbers show, Biggio was significantly below average in all the positions he played.

Finally, Steve Finley, who is one of the 6 members of the 300-300 club (i.e., players who have hit at least 300 HRs and collected at least 300 stolen bases in their careers), comes in last. No shame here, considering the excellence of the group. Finley was considered a very able centerfielder, though his FRAA, for a CF, are on the negative side (-0.95 FRAA per season), though, not significantly different than the average CF.

Unfortunately for Houston Astros fans, management dismantled this group, who in 1991 was already looking very promising. (None of the players in 1991 had an OPS+ lower than 100.) By the way, that 1991 group also included young pitchers Curt Schilling, Pete Harnish, and Darryl Kyle. Not a bad core.

Hall of Fame: Bagwell's easy case

Date: August 20, 2007.

Next Sunday, the Astros will retire Jeff Bagwell's number. They will show a clip with highlights of Bagwell's career. Unfortunately, the highlights will not be enough to capture the excellent career of Jeff Bagwell. To measure Jeff Bagwell's career I am going to analyze the OPS+ empirical distribution of all the seasons of all the BBWAA voted HOF members. We have 104 BBWAA voted HOF members, with a total of 1,355 seasons.

OPS+ Distribution	10%	25%	Median	75%	90%	95%	99%
HOF BBWAA-voted member	85	108	130	153	175	189	219

Now, let's insert Bagwell's OPS+ production into the empirical distribution.

Bagwell's seasons with OPS+ over 130 -i.e., above the median: 12 (only his last three seasons are under the median.)
Bagwell's seasons with OPS+ over 153 -i.e., above the 75%: 5 (94, 96, 97, 98, and 99. BTW, Bagwell had a 152 OPS+ in 2000!)
Bagwell's seasons with OPS+ over 175 -i.e., above the 90%: 2 (94, and 96)

We should note that only 25 BBWAA inducted HOF members have 12 or better seasons with an OPS+ higher than the median (130). But only 4 of the last 20 BBWAA inducted HOF members have 12 seasons with OPS+ greater than 130, and none of them has a better percentage of season with above the median OPS+ production than Bagwell (80%).

Bagwell's career OPS+ is 150. Only 15 members of the HOF have a better OPS+ than Bagwell's 150 OPS+. (Nap Lajoie and Honus Wagner share a tie with Bagwell's OPS+.) And since Hank Aaron's induction, there has not been an inducted HOFer with a better OPS+ than Bagwell's.

Bagwell's best two years place in the top 90% of the distribution. Bagwell's OPS+ production in 1994, 213, places in the top performance 98.4% of the distribution; that is, only 21 seasons of HOF BBWAA inducted members were better. Using the WARP-3(Wins Above Replacement Player, level 3) measure, Bagwell's 1994 is the best individual season by a first baseman in the history of baseball. Bagwell's 1996 OPS+ production places in the top 91.6%.

In addition to his OPS+, Bagwell was considered an above average defensive player. Using Baseball Prospectus� FRAA (Fielding Runs Above Average), Bagwell collected 127 FRAAs, which represents around 50 fielded runs over the median first baseman. (Bagwell also had some speed, stealing 202, but his 72% effectiveness is neat to the break-even point. Therefore, Bagwell's SBs were, overall, not net run producing.)

Based on production, Bagwell's case for the HOF should be a slam dunk. Some BBWAA voters, however, do not emphasize enough relative excellence, and look at longevity -i.e., career totals-, or subjective measures like MVPs, All-Star appearances, etc. Thus, I would not be surprised if Bagwell does not get in. It'd be another sad mistake by the BBWAA voters.

Hall of Fame: Biggio vs. Joe Morgan

Date: July 2, 2007.

I've just heard something on our local, Houston, radio that was outrageous and funny at the same. On our local, Houston, sports radio 790 AM station, Mr. John Lopez hosts the early afternoon show, which is a phone call driven show. Today, a caller, answering the question whether Craig Biggio is a Hall of Famer, said that, after looking at the numbers, "Biggio is no Joe Morgan". Mr. Lopez, who regards Biggio as a lock for the HoF, answered that "(by looking at some numbers) Biggio is better than Joe Morgan." As somebody that knows baseball, Mr. Lopez should be ashamed. But, in Houston, we know that Mr. Lopez is a biased homer. After all, John Lopez is the writter who casted the lone HoF vote for ex-Astros pitcher Jim Deshaies, who had a career ERA+ of 91.

Let's look at Craig Biggio's and Joe Morgan's numbers. (I don't take into account subjective measures such as All-Star appearances, MVP votes, Gold Gloves, etc.) I'll count only seasons with more than 100 ABs, erasing 3 short seasons for Joe Morgan: 1963, 1964, and 1968.

Player (seasons)	AB	OPS+	OPS+ under 100	OPS+ between 130-149	OPS+ over 150	SB	SB %	OBP	SLG
Craig Biggio (20)	10,654	113	7	5	0	413	77%	.365	.435
Joe Morgan (19)	9,277	132	0	6	4	689	81%	.392	.427
Ryne Sandberg (15)	8,385	114	6	6	0	344	76%	.344	.452

The comparison between Craig Biggio and Joe Morgan is not fair. Joe Morgan is considered one of the top 100 baseball players of all time, a consideration the numbers clearly support. (BTW, on baseballfever.com, there is an informative Joe Morgan vs. A-Rod thread.) Biggio has an advantage in slugging, but this number is skewed in favor of Biggio by the era in which Biggio played. And, of course, Biggio will end up with 1,400 more at-bats than Joe Morgan.

The comparison with Ryne Sandberg is probably more appropriate for Biggio. The overall numbers are very similar, though longevity should count on Biggio's side -and great defense should count on Ryne Sandberg's side. Ryne Sandberg is a third ballot Hall of Famer. Biggio will probably be a second ballot Hall of Famer. At worst, Biggio should be a third ballot HoFer, like Ryne Sandberg and, among others, Joe DiMaggio. And that's nothing to be ashamed of.

Note: Another interesting comparison is with Roberto Alomar, who was considered a lock for the HoF during his first 14 seasons. Alomar's three last seasons, after being traded to the NY Mets, were underpar. Alomar's career OPS+ is 116, with only four seasons with an OPS+ under 100 and five seasons with an OPS+ between 130 and 149. Alomar stole 474 SB, with a success percentage of 81%. In addition, for many years, Alomar was considered an outstanding defensive second baseman.

Chan Ho Park: Does he have a chance with The Astros?

Date: June 21, 2007.

The Houston Astros have recently signed Chan Ho Park to a minor league contract. Does he have a chance to pitch and be successful for the Astros? Let's look at Chan Ho Park's recent pitching numbers.

Season	IP	ERA+	BB/9	K/9	HR/9	GO	AO	GO/AO
2004 (TEX)	95.2	93	3.10	5.93	2.08	117	101	1.27
2005 (TEX/SDP)	155.3	75	4.64	6.55	0.64	203	155	1.68
2006 (SDP)	136.2	88	2.90	6.32	1.31	162	152	1.13
Career (14 yrs)	1750.2	97	4.23	7.77	1.05	1339	1200	1.22

Chan Ho Park's performance has been in the decline since he left the L.A. Dodgers in 2001. (Park's ERA+ has not been over 93 since 2001.) His numbers in terms of groundball-to-flyball outs (GO/AO), however, seem to make Chan Ho Park a good fit for the Astros and Minute Maid Park. Though not reported, his HR/9 (1.57) while playing for the Texas Rangers, who play in another hitter's park, should be a concern. (I consider an outlier his 2005 HR/9 number.) Another concern for the Astros should be Chan Ho Park's career BB/9 -close to Wandy Rodriguez's numbers in 2006.

At this stage in Park's career -and the prospect of pitching at Minute Maid Park- Chan Ho Park's overall numbers look similar to T. Buchholz's 2006 numbers. Not very appealing.

The Astros and the Leadoff Hitter Spot

Date: June 18, 2007.

Craig Biggio has been asked, inexplicably, to bat in the leadoff spot in 2007. As easily noticed (and pointed out below, in a previous entry), Biggio's production has been in the decline for a couple of years, but still his production is acceptable for an end of the lineup hitter. To everybody's surprise, Biggio is consistently asked to do things he used to do, but he can no longer deliver. To get a perspective of how bad the Astros have mismanaged the leadoff position, let's compare Biggio's production with other NL leadoff hitters.

Before I briefly analyze these numbers, keep in mind the importance of OBP. According to Tango Tiger on Baseball , with a man on first and nobody out in an inning, the average runs scored that inning are 0.953, while with one out and nobody on, the average runs scored are 0.297. In probability terms, and comparing the two same situations, getting on base while leading off increases the chances of scoring a run by a factor of 3 (from 43.7% probability with man on first and none out, to 17.3% probability with none on and one out).

I collected the batting average, OBP, SLG, and OPS+ for all the NL leadoff hitters that started a game on June 18 and June 17. They are not necessarily "the leadoff" hitters of the club; for example, for the Saint Louis Cardinals, So Taguchi batted leadoff, while the usual leadoff hitter, David Eckstein, is on the D.L. But, this sample will give us an idea of what's expected from an NL hitter. Below, I report the 2007 season numbers for the 16 leadoff hitters in the NL, along with Biggio's numbers. (Biggio's rank is in parenthesis.)

Season	AVG.	OBP.	SLG.	OPS.+
16 NL Leadoff Hitters
2007	.284	.348	.428	105.63
2006	.273	.339	.421	94.81
Biggio
2007	.235 (15th)	.277 (16th)	.396 (10th)	75 (15th)
2006	.246 (13th)	.308 (15th)	.422 (8th)	83 (12th)

Biggio's offensive production (OPS+, OPS measured against the league average, and adjusted for ballpark factors) is at the bottom of all the NL leadoffs hitters in the sample. Only Norris Hopper, used sparingly by the Cincinatti Reds, has a worst OPS+ (72) than Biggio's. But, what is particular troublesome for a leadoff hitter is Biggio's OBP. Biggio's OBP is the worst of all. Second worst OBP belongs to Dave Roberts, which is 28 points higher than Biggio's. Again, Biggio's slugging is not shameful, but it is not middle of the pack, as it was last year.

To put Biggio's numbers in a more practical light, I calculate a team's expected runs scored production when the batters in this sample leadoff an inning. For this purpose, I use Tango Tiger on Baseball average run scores per inning for the different outcomes for a leadoff hitter. I use the 2007 production of the batters in the sample to compute what I will call Leadoff expected runs scored (ERS). The average 2007 Leadoff ERS for this sample is .544. The best 2007 Leadoff ERS belongs to Jose Reyes with .586 ERS, while the worst Leadoff ERS belongs to Biggio with .503 ERS. Assuming 2 plate appearances starting an inning per game, Biggio's numbers translate over the course of a season to 163 runs, this is 13 runs less than the average NL leadoff hitter (and 27 runs less than Jose Reyes!).

That Biggio keeps batting leadoff is very difficult to justify. (I suspect that Drayton McLane's hand is somewhere there and, of course, Tim Purpura's poor team building and poor understanding of the importance of OBP also show up here.) Then, the question for Mr. Garner is: who can bat leadoff for the Astros instead of Biggio?

As the offensive numbers show for this season, the Astros have a below average lineup. Let's use Biggio's numbers to find Biggio's replacement. The Astros would like to have a player with higher OBP and lower SLG, with more than 100 at-bats. Candidates (OBP and SLG are in parenthesis): Ensberg (.314/.378), Burke (.350/.371), Ausmus (.351/.372), and Berkman (.388/.385). Given this choices, and expecting Lance Berkman to recover, Chris Burke is the best candidate. Last season his OPB and SLG numbers were .347 and .418. (As an aside, in May, also inexplicably, Chris Burke was sent to AAA, while Jason Lane's .175 and .354 stayed in the team.)

A final point regarding Biggio's path to 3,000. As pointed out below, in the June 3, 2007 entry, the Astros should give Craig Biggio his best chance to be succesful and appreciated, by batting him in the bottom of the lineup. By consistently giving Biggio the leadoff spot, his very poor OBP makes fans dread Biggio's turn in the lineup. That's not fair to Biggio and to his fans.

Woody Williams, Coors Field and Mr. Garner's 7th inning decision

Date: June 7, 2007.

Situation. Bottom of the 7th inning. Astros 5 - Rockies 4. Woody Williams gave 9 hits, 2 walks and threw 110 pitches, getting 14 flyouts. Kaz Matsui, who bats right, will bat at the bottom of the inning. Matsui is 1-2, with a single, a SB and a BB. Chad Qualls the usual 7th inning reliever is available in the bullpen. Dave Borkowski is also available.

Question for Mr. Garner. Should the Astros keep Woody Williams in the game or bring a reliever to face Matsui?

The Players.
Starting Pitcher: Woody Williams. Woody Williams is not having a good year. Let's look at some of Woody Williams numbers:
Season ERA: 5.50 (Career ERA: 4.15). Season ERA+: 72 (Career ERA+: 104)
Season BAA: .304 -against righties 3.21. (Career BAA: .259 -against righties .267 )
Season WHIP: 1.58 -against righties 1.57. (Career WHIP: 1.32)
Season HR/9: 1.57 (Career HR/9: 1.23)
Season's stats after 76-90: IP 9, H 16, HR 4, BB 3, BAA .364, ERA 9.00
Season's stats after 90-105: IP 8.1, H 8, HR 0, BB 4, BAA .235, ERA 9.72
Season's stats after 106-120: IP 2, H 4, HR 1, BB 0, BAA .500, ERA 9.00
Season's none on/out: .387 .410 (Career BAA .272, OBP .315)
Lifetime ERA at Coors Field: 14.90 ERA (in seven career starts, prior to yesterday's game)

Batter: Kazuo Matsui. Kazuo Matsui is having a good year. As expected, Matsui is no slugger, with a .440 slugging percentage and no homeruns (at that point). But, for the year, Matsui is batting .321 with a .345 OBP. At Coors Field, Matsui is batting .364, with a .404 OBP. Matsui hits righties well. His batting average is .342. He is also hitting .350 with none on/out.

Usual 7th inning reliever: Chad Qualls. Qualls is having just an average year. His ERA is 4.40 (ERA+ 97), BAA is .255, WHIP is 1.26, and HR/9 is 1.88. (The homerun has been Qualls' undoing this year). Righties are hitting .282 against him. This season with none on/out, Chad Qualls has pitched 5 innings, with a BAA .261, a WHIP 1.80 and two 2 HR.

Alternative reliever: Dave Borkowski. Borkowski is having a very good year. His ERA is 3.24 (ERA+ 129), BAA is .242, WHIP is 1.20, and HR/9 is .72. Righties are hitting .233. With none on/out, in 4.2 innings, Borkowski gave no homeruns, but his BAA is .333. Borkowski has pitched extremely well lately. Borkowski should be in the late innings mix.

Decision. The numbers involved in the situation called for taking Williams out of the game. The relievers, except Lidge and Borkowski, have not been very good lately. The usual reliever in the 7th inning, Qualls, has not been very effective this year, but he has better numbers than Williams' numbers at that point in the game. But, if Mr. Garner does not trust Qualls, he should've gone with Borkowski, who has clearly better numbers than Williams -or Qualls. Mr. Garner decided to let Williams face Matsui. A gut call, probably. A nod to Woody Williams' better than expected start at Coors Field. But, not the right probabilistic call.

Resolution. Unfortunately for Mr. Garner, the "gut" failed him big. Matsui, instead of hitting the usual single, hits an unlikely homer. Unexpected homer? Yes, extremely. Unexpected that Matsui got on base? No, not at all. Bad call. (This season is not turning into a good season for Mr. Garner.)

Craig Biggio: Limping to 3,000?

Date: June 3, 2007.

Craig Biggio is having a bad season for his standards. Very bad. It's been said that he has not been playing the last year and a half, instead that he has been "hanging in," just to get to 3,000 hits. It may be true, but to check the validity of the statement, we should compare Biggio's path to 3,000 hits with the same path of the members of the 3,000 hits club. I look at Biggio's last two seasons along with the current season, relative to the last two seasons of the members of the 3,000 hits club prior to the season they got the 3,000 hit, along with the 3,000 hits season. (The numbers here are skewed against Biggio's numbers, since many 3,000 hits club members were not at the very end of their careers when they got their 3,000th hit.)

Season	AVG.	OBP.	SLG.	OPS.
Averages for 3,000 hits club
-2	.315	.384	.471	.856
-1	.294	.368	.431	.799
0 (3,000 hits year)	.297	.369	.435	.804
Career	.310	.382	.468	.851
Biggio's numbers
Season	AVG.	OBP.	SLG.	OPS.
-2	.264 (0.09)	.325 (0.04)	.477 (0.45)	.792 (0.20)
-1	.246 (0.07)	.306 (0.05)	.422 (0.44)	.727 (0.24)
0 (3,000 hits year)	.234 (0.00)	.285 (0.00)	.385 (0.27)	.670 (0.07)
Career	.282 (0.11)	.365 (0.22)	.435 (0.22)	.800 (0.14)

First, the average 3,000 hits member had an average season two years before their 3,000 hits landmark season -i.e., season -2. But, means do not show the whole picture. For example, the empirical distribution of the averages shows that Tris Speaker had the higgest OBP on season -2, with an OBP of .469, 41 points higher than his career OBP. On the other hand, during the same season -2, five players had OBP under .34, with names like Lou Brock, Robin Yount and Cal Ripken, among them. Lou Brock's OPS was 26 points lower than his career OBP of .343. Second, during the season before the 3,000 hits season -i.e., during season -1-, the representative member of the 3,000 hits club hit well under his career numbers. Same situation occurs for the 3,000 hits season -i.e., season 0. Therefore, a decline in offensive production is expected for a player approaching the 3,000 hits landmark.

Now, I examine Biggio's offensive production. The numbers in parenthesis are the percentile of Biggio's numbers in the respective empirical distribution of the 3,000 hits club members. For example, Biggio's batting average in 2005 (Season -2) was .264. That would rank him in the bottom 9% of the 3,000 hits club for Season -2. It is clear that Biggio's batting average and OBP are at the very bottom for all the seasons (-2, 1 and 0). Moreover, for this season -Biggio's 3,000 hits season-, no member of the 3,000 hits club has worst batting average and OBP than Biggio's. That is, Biggio will expand the empirical distribution to the left. This is something that baseball purists should consider disturbing.

On the last two columns, looking at slugging percentage or OPS, there is a different view of Biggio's place among his future peers. Biggio's numbers are not that bad. During seasons -2 and -1, from a SLG point of view, Biggio was very average for a 3,000 hits player. His OPS was nothing to brag about, but he was in the bottom 20% and 24%, respectively. Not embarrassing, given the quality of the group. Robin Yount, Al Kaline, Rickey Henderson, Rod Carew, Lou Brock and Paul Waner had lower OPS than Biggio during season -1. And if I focus on the 2007 season, Biggio's season 0, Biggio's SLG and OPS are below average, but are clearly not unheard of for a player approaching his 3,000 hit.

Is Biggio limping to 3,000? As pointed out above, the average member of the 3,000 hits club also was playing below his average in seasons -1 and 0. Thus, the relevant question is how far below their career numbers did the members of the 3,000 hits club produce? To answer this question, I look into the decline on OBP and SLG. The following table reports the median, bottom 10% and lowest percentage change in OBP and SLG relative to career numbers.

Season	Median OBP.	Bottom 10% OBP.	Lowest OBP.	Median SLG.	Bottom 10% SLG.	Lowest SLG.
-2	.002	-.086	-.129	-.008	-.127	-.201
-1	-.034	-.154	-.233	-.100	-.261	-.453
0 (3,000 hit year)	-.038	-.105	-.189	-.042	-.267	-.347

The median 3,000 hits club member had an OBP and SLG during season -2 that was very similar to his career average, with a slight improvement of 0.2% over his career OBP and a slight decline of 0.8% over his career SLG. But there is a 3,000 hits club member, Paul Waner, that had declines of 12.9% and 20.1% in OBP and SLG, respectively, over career numbers. For those who don't have a recollection of Paul Waner, the second highest decline in SLG during season -2 was Cal Ripken Jr.'s. Cal Ripken had declines of 2.7% and 13% in OBP and SLG, respectively, over career numbers. As pointed out before, declines are the norm for seasons -1 and 0, and declines of 10% or more, especially in SLG, are common.

Below, I report Biggio's change in production in his last two seasons, along with the current season, relative to his career numbers. (Again, numbers in parenthesis represent the percentile in the respective distribution of the 3,000 hits club members.)

Season	OBP.	SLG.
-2	-.110 (0.03)	.076 (0.74)
-1	-.162 (0.04)	-.030 (0.62)
0 (3,000 hit year)	-.219 (0.00)	-.115 (0.38)

Biggio's decline in OBP, relative to his career numbers, puts him at the bottom of the distribution, especially for the current season. But, again, slugging percentage-wise, Biggio's decline is very decent for a 3,000 hits club member. For example, even for his season 0 -i.e., the 2007 season-, his decline would place him in the bottom 37% of the distribution. Not bad.

So, is Biggio limping towards 3,000 hits? For a leadoff hitter, where the OBP is valued, he is limping a lot relative to his future peers. But, at his age and with his production numbers, Biggio should never be hitting leadoff. It is a shame that Mr. Garner keeps using Biggio as a leadoff hitter, instead of using him as a number 6 or number 7 hitter, where he'd be valued more for his slugging, than for his OBP.

Observations on the 2007 Astros, after 1/4 of the season

Date: May 24, 2007.

Poor lineup. Team batting average: Bottom 5 in MLB. Team slugging: Bottom 10 in MLB. Not good. Let's look at one recent lineup.
HOU 1 - SF 9 (Wed, May 23).

C. Biggio 2b, 0-3, 1 BB, .237
M. Ensberg 3b, 1-4, 1HR, .216
M. Loretta 1b, 0-4, .329
C. Lee lf, 0-4, .307
H. Pence cf, 2-3, 1 BB, .341
J. Lane rf, 0-3, .176
A. Everett ss, 0-3, .192
H. Quintero c, 0-3, .216
M. Albers p, 1-2, .222

Pathetic. Three position layers hitting under .220. Five players -not including the pitcher- hitting under .240. No wonder the Astros are going nowhere.

Pitching still decent. Team ERA around 4.3. About average for MLB. Similar story for Team WHIP. Oswalt, as expected, very good. Sampson, a nice surprise. Even Wandy Rodriguez has been decent (knock on wood), so far. Jennings (DL) in his two starts was fine. The main relievers (Wheeler, Qualls, Lidge), along with Borkowski, have been very solid. Woody Williams? As expected, bad (see May 3, 2007 entry below). Woody Williams has always been prone to the HR, but he was pitching in low HR parks (PETCO and Busch) during the past couple of years. Now, at above average Minute Maid, Williams is in trouble. The rest of the pitching staff? Also as expected, bad.

No Bench. After Loretta and Lamb, who should be regulars, the Astros have no bench. This is also inexcusable and it's the fault of the GM, Tim Purpura. (By the way, Mr. Purpura has still to justify replacing Mr. Hunsicker as GM three years ago. It should be noted that, according to David Gassko, Hunsicker ranked number 2 as GM in 2004.) Take a look at Orlando Palmeiro. Yesterday, Orlando Palmeiro entered the game hitting .138 (I know, it's only 32 at-bats and Palmeiro's career OBP is decent (Hunsiker signed him). But, Palmeiro is 38 and last year's OBP was .294.) Palmeiro had a very decent season in 2005. Andujar Cedeno had a very decent season in 1993. Mr. Garner, since he has a very weak bench, should consider letting the pitchers hit, especially if the pitcher is Woody Williams (a .204 career hitter).

Adam Everett. Considered one of the best, if not the best (see John Dewan's The Fielding Bible), defensive shortstops in the game, Adam Everett has been terrible at the plate. A .245 career hitter, nobody was expecting Miguel Tejada's numbers from Adam Everett. But, hitting under .200? Walt Weiss, another defensive wizard, had poor hitting seasons with the Oakland A's; but never this poor. Nobody would complain if Adam Everett were batting ninth for the Boston Red Sox (or for Walt Weiss' Oakland A's). But this is the "poor hitting and avearge pitching" Houston Astros. With the current Astros lineup, Everett should be used more like Tim Bogar. A platoon starter (for example, when Woody Williams starts, Everett's defense is not that relevant) and a late inning defensive replacement.

Craig Biggio. What is Biggio doing batting leadoff for the Astros? Just to give him thirty or forty more at bats in the first half? (At his current pace, that's less than 10 hits!) He is going to get -one may say, painfully- to 3000 hits, regardless. If he gets 400 at-bats and hits .230, he'll get to 3,022 hits. What's the hurry? Biggio is hurting the team batting leadoff. Biggio knows it. Mr. Garner knows it. The opponents know it. Biggio's OBP is under .300. He is not a threat to steal a base anymore. Unfortunately, Biggio hurt the Astros during the second half of last year, when he was correctly demoted to hitting sixth. And recall that one single game was the difference between the Cardinals and the Astros! It makes no sense.

Phil Garner. Biggio's batting leadoff is puzzling. I also don't understand the makeup of yesterday lineup. I understand that managers try to keep the players fresh, and that Mr. Garner is working with what Mr. Purpura is giving him. But, yesterday's lineup chances of wining are very low in any day. Was Mike Lamb unavailabe? Lance Berkman? Mr. Garner is just turning into a mediocre manager, below average. Sabermetricians say that a good manager adds three or four wins over the course of a season. This season, Mr. Garner, probably, subtracts one or two. Strange, before coming to Houston, Mr. Garner was considered a good manager; that is, a manager that adds wins to his team. (See Bradford Doolite's article on wins added by MLB managers.)

The Press, the HR Record and Barry Bonds

Date: May 17, 2007.

No numbers today. Just an observation on how the average media outlet is treating Barry Bonds, as he gets closer to Hank Aaron's HR record. Barry Bonds is continuously bashed by the press, presumably because of his steroids use. The fans also bash him, but no surprise here. Mr. Bonds has been used to the routine boos at visitor's ballparks for a long time. So, let's focus on the press.

According to comments made by the press, Mr. Bonds has an unfriendly attitude towards the press and the fans. With some exceptions (Peter Gammons being one), the press has been very tough towards Barry Bonds. I've heard two of the local, Houston, sports experts on the radio called Mr. Bonds a "jerk ." Today, another local radio sports "personality," John Lopez, called Mr. Bonds a "one tool" player, closing his statement about Mr. Bonds by saying "now, he is a tool." BTW, I thought that batting for average and hitting for power was "two tools," Mr. Lopez. But, we can't do much here. In general, local radio guys are fairly unaccountable for what they say, unless is a racial comment.

The press knows perfectly well that Barry Bonds was/is not the only MLB player to have used steroids. According to Ken Caminiti, 50% of MLB players used steroids during his playing days. Jose Canseco, in his book, implicates many of his teammates as steroids users. There are hundreds of baseball players who changed their bodies from one year to the next and whose numbers went up substantially. But, the press keeps the pressure up on Bonds, Barry Bonds.

Jason Giambi seems to have been forgiven by the press. Derek Turnbow, who along Termel Sledge tested positive for performance enhancing drugs while playing for USA Baseball in January 2004, never gets into the steroids discussion. Here, in Houston, Roger Clemens, the local hero, seldom gets mentioned in the mix, even though he has been recently implicated in the steroids saga, along with the other hometown pitcher, Andy Pettite.

Barry Bonds has committed two sins. One, he used steroids. Two, he has been unfriendly towards the press. It's my impression that sin number two is the one that the press cannot stand. Albert Belle can attest to the importance of the second sin.

Moral of the story: Be friendly to the press. Be very friendly. And your sins will be overlooked and, eventually, completely forgotten.

Reference: The Steroid Era Blog

Roger Clemens: Unusual finish

Date: May 10, 2007.

Last week, Roger Clemens got a very generous contract from NY Yankees, reportedly making a $4.5 million per month. Unusual amount. But, what is really unusual is the last two years of Roger Clemens' career. Below, I report some of Roger Clemens numbers (IP, Strikeouts, ERA and ERA+) for the last 7 years of his illustrious career.

Season	IP	SO	ERA	ERA+	ERA+/Career ERA+
2000	204	188	3.70	137	0.951
2001	220.1	213	3.51	128	0.889
2002	180	192	4.35	101	0.701
2003	211.2	190	3.91	112	0.777
2004	214.1	218	2.98	145	1.007
2005	211.1	185	1.87	221	1.535
2006	113.1	102	2.30	197	1.368
Career	4817.2	4604	3.10	144	1

Roger Clemens' numbers were tailing down from 2000 to 2003. His ERA and ERA+ were under his career averages. Nothing strange there. That is the usual pattern for a pitcher, even a Hall of Famer, at the end of his career. For example, during the last four years of their careers, the members of the 300 wins club had a mean ERA+ of 125, 117, 107, and 93, with a career ERA+ of 125. Clemens retired at the end of the 2003 seasons.

Suddenly, in 2004, Clemens unretires and jumps to the National League to pitch for the Houston Astros. This jump to the NL may partly explain the jump in Clemens' ERA+ and reduction in Clemens' ERA. According to an article by Jason Scwartz, from the NY Times, a pitcher that goes from the AL to the NL can see a difference of 13% in his ERA+ and a decrease of 0.85 in his ERA. Then, the incredible happens. Clemens produces, at 43 and 44, two of his best four seasons, measured by ERA+. Moreover, Clemens' last two seasons rank in the top 5% of ERA+ ever produced by a 300 wins club member (plus future member Tom Glavine) in a season. Wow!

But, to get a better measure of how unusual the last two years of Roger Clemens are, I compare Roger Clemens' performance towards the end of his career, with the performance of his peers in the 300 wins club at the end of their respective careers. I look at the last years of all the pitchers that won 300 games plus future member Tom Glavine. For this analysis, I use the last 30% years of MLB service. That is, for a pitcher that pitched 23 years, like Roger Clemens, I take the last 7 years (23 years x .30 = 6.9 years). Using this cut-off measure, I gather a sample of 127 observations. To measure performance, I focus on the ERA+, relative to career ERA+. Roger Clemens' numbers are shown in the table above, in the last column. Below, I present the empirical relative ERA+ distribution of the 300 wins club member plus Tom Glavine.

(ERA+)/(Career ERA+) Distribution	Largest	Top 95%	Top 90%	Median	Bottom 10%	Bottom 5%	Lowest
Last 30% games (300 games winners)	1.535	1.245	1.154	0.925	0.656	0.577	0.187

In 2005, Roger Clemens had the best season of any 300 wins club member at the end of their careers. His ERA+ was 53.5% better than his career ERA+. To put things in perspective, two years before retirement, Tim Keefe at 35, in 1892, had the second best season of the group, with an ERA+ of 138, 10.4% better than his career ERA+ of 125.

In 2006, Roger Clemens' ERA+ was 36.7% better than his career ERA+. This relative performace is the fourth best in the whole sample. (Though, it can be argued that Early Wynn's 1.5 relative ERA+ to career should not count, since he started only 5 games, pitching 55.1 innings in 1963, his last year.) After Wynn and Clemens, the best relative ERA+ during their last MLB season belongs to Eddie Plank, at 41, in 1917, with a 19.7% ERA+ improvement over career ERA+. Clemens' numbers are nothing but amazing, especially for a pitcher that is at the end of his career. Barry Bond-esque.

Can the NY Yankees expect Clemens to repeat his 2005/2006 performance? (I guess they are paying him more than any other palyer in MLB with the expectation of exceptional performance.) First, with the change in leagues, it should be expected a significant decay in his ERA+ and a significant increase in his ERA. Second, given the usual history for a 300 wins club member at the end of his career, the NY Yankees cannot expect a 2005/2006 year from Roger Clemens.

At Roger Clemens age, another performance similar to 2005 would be highly unlikely. Again, let's go back to the empirical distribution and its meaning. By looking at the empirical distribution of ERA+ for the 300 wins club members at the end of their careers, ex-ante a performance like Clemens' 2005 had no chance of happening. Also, a performance like Clemens' 2006 had less than a 3.5% chance of happening. Based on this analysis, if 2007 is Clemens' retirment year, a 20% improvement over his career ERA+ would be unheard of. From this historical perspective, the NY Yankees are very likely making a costly -in million dollars- mistake.

Woody Williams and the Astros: Why bring a flyball pitcher?

Date: May 3, 2007.

Before trying to understand the rationale behind the Astros decision to offer a 2-year $12.5 million deal to Woody Williams, I want to briefly present some facts about the value of different pitching styles.

Dave Studeman wrote a piece in the 2006 Hardball Times Annual, where he discusses the run value of every outcome from an at-bat. These run values are calculated from real life play-by-play data, and, thus, present real outcomes:

Line Drive: .356 (an average line drive is worth 35% of one run)
HBP: .342
Non-Intentional Walk: .315
Intentional Walk: .176
Outfield Fly: .035
Groundball: -.101
Bunts: -.103
Infield Fly: -.243
Strikeout: -.287

The difference between a flyball out and a groundball out is small, however, over the course of the season this difference can mean 5 to 10 extra runs. This is why baseball experts value more a groundball pitcher.

Now, back to Woody Williams. Woody Williams has had a slightly better than average career. His ERA+ (ERA measured against the league average, and adjusted for ballpark factors. 100 is average) is around 105. His last four years show ERA+ of 107, 100, 79, and 116. The weak side of Woody Williams is that he is a flyball pitcher. Below, I show some stats for Woody Williams' past four years. (Note: GO: groundball outs, AO: flyball outs.)

Season	IP	BB/9	K/9	HR/9	GO	AO	GO/AO
2003 (STL)	220.2	2.24	6.24	0.82	225	275	0.84
2004 (STL)	189.2	2.75	6.22	0.94	192	238	0.83
2005 (SDP)	159.2	2.87	5.97	1.35	163	207	0.83
2006 (SDP)	145.1	2.17	4.46	1.30	159	205	0.84

During the past years, especially in 2006, Williams' K/9 have gone down, and, thus, flyball outs have become more relevant. Note that Woody Williams has always been prone to the HR. During his last two seasons at PETCO Park, surprisingly, he returned to his pre-Cardinals ways (Busch is also a below average HR park). His career HR/9 is 1.22.

The Houston Astros have in the past years had a rotation with high GO/AO. Let's look at the 2006 Astros starting pitchers' relevant stats.

Pitcher (ERA+)	IP	BB/9	K/9	HR/9	GO	AO	GO/AO
R. Oswalt (152)	220.2	1.55	6.77	0.73	289	195	1.60
A. Pettitte (108)	214.1	2.94	7.47	1.13	292	168	1.89
W. Rodriguez (81)	135.2	4.18	6.50	1.13	174	129	1.50
R. Clemens (197)	113.1	2.30	8.10	0.56	139	96	1.56
T. Buchholz (77)	113.0	2.71	6.13	1.68	137	124	1.18
F. Nieves (108)	96.1	3.83	6.54	1.68	111	107	1.12

The signing of Woody Williams is a departure from the Astros' traditional (maybe Mr. Hunsicker's) emphasis on groundball pitchers. Keep in mind that Minute Maid Park is consistently in the top 10 hitter's parks in MLB. In light of this fact and the traditional makeup of the Astros starting pitchers, the signing of Woody Williams is a move difficult to justify.

Notes:
(1) What kills Wandy Rodriguez is the walks. Throw strikes, Wandy!
(2) What happened to Fernando Nieves? Ezequiel Astacio was brought back in 2006, after an uninspired 2005 season (in 81 innings, ERA+ 73. Ouch!).