Tuesday, August 31, 2004

If I See "Catch-22" One More Time...

I know that Cleveland 22, New York 0 is a fabulous non-story story, but it would be nice if the mainstream baseball analysis would do some actual baseball analysis. I spent hours looking through the minutia of folks like Joe Randa and Eric Munson in preparation for my big Adrian Beltre entry and felt guilty when I realized I wouldn't be able to keep up with the schedule I'd set for myself to have 4-5,000 words out a day because I just looked up to much darn stuff to form a complete opinion (but not as guilty as when I don't check out someone's PECOTA card for more insight or look up their in-season splits or find out what happened to the player they were traded to in 1998 and so forth). Meanwhile, Buster Olney... gosh, I really shouldn't finish this thought. Anyway, I'll have something good up tomorrow night and by the end of the weekend at the very latest I'll have finished my magnum opus on what will/should happen with Adrian Beltre and all the other major league third basemen in the offseason. I doubt anyone will actually refuse to check this blog henceforth on the basis of me not having 4,000 words up on my third day, but it sure feels like it, so please check back in the next few days.

I haven't seen anybody else make this comment yet, but someone probably has and I just haven't seen it, so apologies to whoever deserves credit for saying it first: 22-0 losses really hurt New York in the Pythagorean standings, which is unfortunate because if the Red Sox finish one or two games back with an 8-game lead in the Pythagoreans (they lead by 5 right now), we may hear about it another 1700 times or so. The Indians, on the other hand, needed that 22-0 victory for their Pythag. to come within a game of their actual record, and they still trail Detroit.

Brief Jason Kendall Update

Archi took a little longer than I was hoping, mainly because of how interesting it was to look up the career numbers of people like Tim Teufel, Randy Ready, Bip Roberts and Randy Velarde, none of whom even made the final cut. On top of that, Monday was my first day of classes, which stole a chunk of my blogging time, and with the two-hour beginning of the year council meeting at the co-op I live at, I wasn't able to finish research for Jason Kendall, part 2. Since tomorrow is the last day for a Jason Kendall deal this year (at least for it to help the Dodgers in the playoffs), I thought I would try to provide a glimpse of Jason Kendall, part 2 because in part 1 I left that kind of open ended. Basically, I think at this point that if the Dodgers got Pittsburgh to pay about 40% of Kendall's salary in each remaining year of his contract (and to pay any deferred money that's still owed to him that I've yet to discover), a David Ross and Chuck Tiffany for Jason Kendall deal would benefit both the Dodgers and the Pirates substantially (basically because marginal wins are more expensive for the Dodgers than for the Pirates for the forseeable future). Whether those terms are agreeable to Frank McCourt and Kevin McClatchey is a different story. Whether Dave Littlefield and Paul DePodesta agree is also a different story, but I think, based on my cursory investgation, that they would unless DePo knows something about Dave Ross or other young catchers that I don't. But as I said, this is based on only cursory research of the payroll issues, Dave Ross' development curve, trends for 30-33 year-old catchers with high OBP and low SLG and inconsistent defensive records, etc. I hope to have a full Jason Kendall analysis up tomorrow night, and my gut feeling is that we won't see him with the Dodgers in 2004 because McCourt is squeemish financially and Pittsburgh is unreasonable in negotiations. The Pirates did get a pretty great return on the Giles deal, but I don't think DePo would trade away anything resembling Jason Bay and Olver Perez unless Pittsburgh paid all of Kendall's salary. And taking on Kendall's full contract by claiming him on waivers would be a terrible move unless McCourt paid for it out of his own pocket (and since he only has virtual money at this point, I don't know how that would happen).

Offshoot of Archi

When writing my paean to Archi (which I'm hoping someone somewhere reads eventually), I realized that both Rickey Henderson and Gary Sheffield have been traded four times in their respective careers. I'm trying to think of any other Hall of Fame caliber players to have been traded four times during the productive portion of their careers. Sheffield and Henderson have had tremendous years after their most recent trades and both had had excellent years before their first trade (although Sheffield's 1991 was pretty bad when he played before his injury, his 1990 was pretty excellent). Can anyone in my super-limited readership think of anyone else who meets these criteria?

Wow, my shortest post yet.

Monday, August 30, 2004

Archi Cianfrocco Revisited

Several teams had the day off today, so I thought I'd spend the day with a playful journey through the past.

In the cinema verite classic "This Is Spinal Tap," Tap's manager, Ian Faith, is asked if the band's shift from playing arena shows to playing small clubs indicated a decline in its popularity. He replies "Oh, no, no.....not at all. It's that their appeal is becoming more selective." In that vein, the following statement probably has a very selective appeal: Archi Cianfrocco is the second best name of all time, just behind Scatman Crothers.

You likely only remember Cianfrocco if you were a Padres fan in the mid-90's. I heard announcers pronounce his name as "Arkey Sin-frock-uh", which is probably correct. For me, however, he will always be "Ar-chee See-an-FRAW-ko". That's how me and my cousin Zack pronounced his name back in 1993 when we made the ultmate discovery: the 93 Fleer Ultra Archi Cianfrocco card. At the time, we didn't think he was a particularly significant player. We did, however, recognize greatness in names instantly, and Archi had it in spades, even more than Mike Piazza. Zack was a fairly decptive twelve-year-old, and he convinced me that we should try to trade 93 Ultra Archi Cianfrocco to my brother for 86 Topps Roger Clemens. We told my brother that he was the can't-miss prospect of the moment, and with no BaseballAmerica.com to check his minor league stats he took our word for it (I mean, he would have to-- this was a great name!). Eventually I told my brother what we had done, and everything was resolved in a remarkably peaceful fashion.

So we had sold Cianfrocco as being the next George Brett (except with more power), but what kind of career did he really have? First of all, Archi was born Angelo Dominic Cianfrocco on October 6, 1966 in Rome, New York, and he played collegiately at Purdue. In January 1986, the Pirates drafted Archi in the 11th round of the free agent draft. He didn't sign, so Pittsburgh tried again in the 4th round that June. Once again, Archi broke Pittsburgh's heart by not signing. I haven't located his college stats, but I did find out that one year he had 79 hits, which was second place on the all-time Purdue single-season list at the time, and he probably did that in 1987 after spurning the Pirates. In June 1987, the Expos took Archi in the fifth round, and he signed. He hit .247 in the short season at Jamestown that year and was promoted to high A Rockford the next year where he hit .253 with 15 HR. He was promoted to Double-A Jacksonville for 1989, and he hit .249 in '89 and .219 in '90 with an ISO of .130 or so each year, and his K/BB ratio was just under 4. When the Expos AA affiliate became Harrisburg in 1991, Archi improved his AVG to .316, but was repeating Double-A for a second year at 24-years old and didn't improve his K/BB or ISO much. On the basis of that he started 1992 with the big club in Montreal. At the time, the Expos might have mistakenly seen Cianfrocco as a potential long term replacement for an aging Tim Wallach at either first or third base, especially after Wallach's .225/.292/.334 line in 1991, on the basis of his strong Double-A batting average. In the short term he platooned with Bret Barberie at third base (Archi was a right-handed hitter and Barberie switch-hit but was terrible against lefties) while Wallach started the season as the regular first baseman; the Expos might have had some expectation that Barberie and Wallach would be their regular corner infielders for some time to come after Barberie hit .353/.435/.515 in 162 PA in his first cup of coffee in 1991. On May 20, the Expos ditched manager Tom Runnells after a 17-20 start, and hired Felipe Alou, who guided the Expos to a 70-55 record the rest of the season. When Alou was acquired, Archi was hitting .230/.269/.429, Barberie was hitting .244/.376/.289, and Wallach was hitting .200/.299/.304. Alou must have got it into his head that Wallach's struggles came from moving to first base, because on May 23 he started Wallach at third and Archi at first, leaving Barberie on the bench. A glib analyst might say that Alou preferred outs and occasional extra bases over getting runners on base, and such an analyst might point toward Neifi Perez hitting leadoff in 20 or so games and hitting second in the lineup another 40 or so games during Alou's tenure to support that statement, but this analyst does not have time for such glibness or to research the exact number of games. In any event, Alou probably believed that this move paid off, with Archi and Wallach both hitting home runs (off Charlie Liebrandt and Mike Stanton, respectively) that day in a 7-6 win over Atlanta with a combined line of 4 hits, 3 runs, and 4 RBI in 7 plate appearances. With that rousing success, Archi became the everyday first baseman until July 9 when, with Archi hitting .250/.274/.383, the Expos turned over the first base job to new call-up Greg Colbrunn. Archi only logged 46 more plate appearances before being sent down to AAA Indianapolis on August 17 where he hit .305 with 4 HR in 15 games. His final line at the major league level was .241/.276/.358, and he'd been an above average fielder at first base but below average while fielding at third base. His replacement, Colbrunn, finished with a .268/.294/.351 line. Wallach went .223/.296/.331, leading to an offseason trade for Tim Barker, and Barberie's walk-driven .231/.354/.281 and good glovework at third and while filling in at second for an injured Delino Deshields in the last four weeks of the season were underappreciated by the Expos, who left him exposed in the expansion draft to be taken by Florida (Barberie had two league average seasons of contribution with Florida before netting them Jay Powell in a trade with Baltimore; his production dropped a bit with the Orioles before his terrible 15-game stint with the Cubs after signing as a free agent forced him out of the major leagues). Both Barberie and Wallach were somewhat expendable in part because of Archi's disappointing season-- the Expos acquired B+ prospect Sean Berry on August 29 by giving spare parts Bill Sampen and Chris Haney to the Royals, hoping he could help them make up the 3.5 games they trailed the Pirates by. Berry had spent the year at Triple-A but made a smooth transition to the majors, hitting .333/.345/.404, but only started twelve times. The Expos reversion-to-the-mean 18-16 record after the trade left them 9 behind Pittsburgh. But Colbrunn's failure to capitalize on his opportunity meant that Archi may have had a slim chance to be Montreal's first baseman of the future.

Cianfrocco started 1993 with the Expos, and the starts at first base were divided between him and John Vander Wal. In game three of the first major league series played in Colorado, played on April 11, he started and hit 8th in the lineup. He had one of the best days of his career, slugging a solo home run and two singles to go along with two strikeouts in six plate appearances. That home run was his first since June 3, 1992 (probably the best day of his career- 4 for 4 with a solo shot), and it would be his last with Les Expos. After April 11, he played in only seven more games with the Expos, logging only two plate appearances. On April 21, the Expos decided to call up Gregg Colbrunn and send Archi back down. Archi went to Ottawa of the International League, where he hit .298 with 4 HR in 188 AB over 50 games with 33 strikeouts and only seven walks.

On June 23, the Padres were thinking one or two of these things, possibly all three:
1. Gary Sheffield will probably revert to "headcase" status, his great 1992 season was an aberration since he's "only" hitting .295/.344/.473, and he's making big time money ($3.1 million)
2. Trevor Hoffman will be the most valuable relief pitcher in the history of baseball, and we will need a replacement level starter (Jose Martinez) and reliever (Andres Berumen) in a couple years.
3. Archi Cianfrocco is hitting .298 at Triple-A!!! He's the next big thing and he'd play for the league minimum!!!
Obviously, that's an oversimplification. The Padres' new ownership apparently forced a Sheffield trade, and on June 24 the Sheffield to Florida for Martinez, Berumen, and Hoffman deal was completed, giving San Diego nothing close to equal value-- neither Martinez nor Berumen were ever considered high ceiling prospects. In preparation for the Sheffield deal, the Padres on June 23 traded Tim Scott to Montreal for Cianfrocco (a good deal for the Expos-- Scott had a 3.38 ERA over 197 innings with the Expos before being traded with Kirk Rueter for Mark Leiter in 1996). Archi took over as the Padres' everyday third baseman, and remained so until August 27. On that day, the Padres decided 25-year-old Guillermo Vasquez wasn't cutting it since taking over as the starting first baseman after San Diego traded Fred McGriff on July 18. The Padres decided to give Vasquez' at bats to 26-year-old Jarvis Brown, who they had signed as a free agent that winter after the Twins gave up on him. To facilitate the lineup shakeup, the Padres called up Jarvis Brown to play centerfield and had then-centerfielder Derek "Operation Shutdown" Bell (whenever I think of Bell, I recall the Rob Neyer column that referred to him as "one of the ugliest men in the major leagues") play his first and only 19 career games at third base (which, for anyone keeping track, meant the Padres had gone from Sheffield playing third to his Little League World Series teammate playing third). On September 17, Archi moved back to 3rd base so that 25-year old rookie Dave Staton could log playing time at first, forcing Derek Bell to fight Phil Clark and Bill Bean for playing time at the right field position vacated by Tony Gwynn's injury. Even with the regular playing time, though, Archi's bat didn't manage much improvement over 1991. Archi hit only .194/.264/.398 in his first 5 weeks with the Padres, improved those numbers to .272/.320/.455 over the next five weeks, and hit poorly over the final five weeks to finish at .244/.289/.412 for the season with San Diego.

In 1994, Archi Cianfrocco was San Diego's opening day starter at third base. He had started 27 of the Padres first 31 games when on May 10 San Diego probably decided that his .229/.259/.413 line was a pretty good representation of his ability since it so closely mirrored his .241/.276/.358 and .243/.287/.416 lines in '92 and '93. They filled third base with 30-year-old rookie Keith Lockhart for a few games while making the Gene Harris for Scott Livingstone and Jorge Velandia deal, which was actually a surprisingly good one for the Padres, as Livingstone was pretty bad in '94 (.272/.294/.383) but good in 1995 and serviceable in the 1996 playoff run while Harris was a terrible pitcher both before and after the trade and his salary was $400K higher than Livingstone's. Livingstone took over at third for the remainder of the season (his numbers in late August and September were particularly good), but Archi couldn't get playing time at first since, even though the Padres gave up on Dave (.189/.289/.394) Staton, San Diego still preferred Tim (.254/.307/.280) Hyers to Archi at 1B (Hyers was later replaced by Eddie Williams, who put up a .331/.392/.594 line to momentarily resurrect his career). Even on a team that, outside of its outfield, was composed entirely of castoff utility men and Brad Ausmus, Archi could only log occasional playing time as a defensive replacement at first or third or pinch-hitting and pinch-running.

With Ken Caminiti coming to the Padres in the massive 11-player offseason "Strikes Are Boring, Let's Do Something" swap, San Diego had a legit third baseman, and Eddie Williams was slotted to play first base after tearing the cover off the ball in 1994, Archi didn't start the season on the Padres roster. He was finally recalled to the big leagues in mid-July. 1995 is the first year that "UT" is listed as Archi's position at Retrosheet, and the Padres utilized him to the tune of 5 batting runs above replacement (.263/.333/.449 with a .268 EqA, his only year with an above average EqA) and 2 fielding runs above replacement while playing every position but catcher and centerfield. Over a third of his playing time came at shortstop, where he was terrible defensively but was playing in place of Andujar Cedeno who was nearly as bad defensively and had the added bonus of being a truly terrible hitter. All in all, this was probably ound for pound Archi's best year.

In 1996, the Padres had ditched Eddie Matthews and replaced him with Wally Joyner, which, along with Caminiti's steroid-enabled career year and the addition of Rickey Henderson, contributed to the Padres scoring a good deal more runs, propelling them to the playoffs. But the addition of left-handed Wally Joyner also meant more playing time for Archi, who frequently played at first base against left-handed hitters. Combined with occasional play at right field, shortstop, second base, and third base, Archi was on the field enough to come to bat 203 times. His offense and fielding were both just below average, and he contributed eight runs above replacement.

In October 1996, the moment finally came for Archi when he made his playoff debut in game one of the division series against the Cardinals. In the top of the 8th trailing by two with one out and Steve Finley on first base, Bruce Bochy sent Greg Vaughn to pinch hit for Wally joyner against legendary first LOOGy Rick Honeycutt (although Honeycutt wasn't a one-out guy in this game-- he got Gwynn in the seventh, was left in to hit and struck out with a runner on second to end the innng, gave up a single to Steve Finley and struck out switch-hitting Caminiti); in response, Tony La Russa put Dennis Eckersley into the game, who induced a pop-up to shortstop and then retired Chris Gomez to strand Finley. Archi came into the game at first in the bottom of the inning, and immediately busied himself trying to hold Royce Clayton on first after his lead-off walk. Clayton went on a two-strike count, and the Cardinals recorded a strike-em-out throw-em-out 2-6 double play. Ron Gant then walked and stole second before Brian Jordan was fanned to end the inning. Whether Archi would have hit for himself in the ninth is unknown to me, but the Padres came up a couple batters short of finding out when, after two-out singles from Chris Gwynn and Rickey, Tony Gwynn grounded out to Eckersley, giving St. Louis a 3-1 win.

In game 2, Steve Finley tied the game at 4 when he scored Scott Livingstone with a one-out groundout to first in the top of the eighth. After Caminiti was intentionally walked to load the bases, Wally Joyner popped up to end the Padres' rally chances. Doug Bochtler entered the game for the Padres and walked Brian Jordon. After Gary Gaetti grounded out, Bochtler intentionally walked Jon Mabry before throwing a wild pitch that put Jordan on third and Mabry on second. Bochy decided it was time for Hell's Bells and, with the pitcher spot due up fourth in the ninth, made the call for Archi to enter the game at first base as part of a double switch. The next batter, Tom Pagnozzi, hit a liner that ricocheted off Hoffman's glove to second baseman Jody Reed, who couldn't get the ball home in time to prevent Jordan from scoring the go-ahead run and had to settle for throwing the ball to Archi to retire Pagnozzi. Luis Alicea ended the inning by flying out to right. Had the Padres been able to put the tying run on base in the ninth, Archi would have his first playoff at bat. Sadly, all three Padres hitters were retired, sending the series back to San Diego with the Cardinals up two games to none.

With Archi's only positive influence in the playoffs (recording the put out on Tom Pagnozzi) coming on the play that lost Game 2 for the Padres, Bruce Bochy and the whole world knew one thing: Game Three would be Cianfrocco time. Well, either that or Bochy didn't want to have Joyner facing lefty Donovan Osborne. Archi started at first at number five in the order. He started the game off with a bang by recording a rare inning-ending 3-2 double play on Gary Gaetti's grounder, although Retrosheet notes that catcher Brian Johnson dropped the ball and the ump blew the call. In the second inning, Cianfrocco came up with one out and no one on and singled to right. Brian Johnson came up next and hit what appeared to be a sure inning-ending double play grounder to short, but a bad hop let the ball get past Royce Clayton and Archi moved to third. Chris Gomez then hit a grounder that Archi tried to score on; the throw home was not in time. After Johnson scored on a Jody Reed double, the Padres led 2-1.

In the third, Archi contributed more solid defense, retiring Donovan Osborne unassisted on a grounder and throwing Royce Clayton out at second on a Willie McGee grounder. In the bottom of the inning, he flied out with none on and two out. In the fifth, Archi came to the plate with the Padres leading 4-1, none on, and two out, and struck out. Then the Cardinals narrowed the gap in the top of the sixth on a Gant homer, a Jordan single, and a Mabry triple. Bochy knew it was time to pull Ashby, but decided upon a double switch, with Joyner playing first in the 9-slot (due up fourth the next inning). The next batter, Pagnozzi, singled in tying run off Tim Worrell. In the sixth, the Padres went 3-up, 3-down, meaning Joyner was due to lead off the bottom of the seventh. In the top of the seventh, the Cardinals tied the game, driven by a pinch-hit single by Mark Sweeney hitting for the pitcher. With the knowledge that Joyner was due up first in the bottom of the inning, La Russa had Honeycutt warming up in the pen (he likely would have entered the game in the seventh anyway, with Henderson, Gwynn, and Finley due up after Joyner). Honeycutt plowed through the Padres' four lefties, yielding only a single to Gwynn (the quintet would finish the day with two hits and 15 outs in 17 PA). Caminiti led off the bottom of the eighth with a homer off Honeycutt, bringing the pitcher's spot to the plate with a tie game. The Cardinals brought T.J. Matthews in to pitch and the Padres had Greg Vaughn hit; he struck out. The next batter, Johnson, hit a double but was stranded. In the ninth, Hoffman entered the game and gave up a walk to Gant and a homer to Jordan. Joyner led off the ninth by flying out to center against Eck, and Eck got Gwynn and Finley to end the series.

Second guessing Bruce Bochy eight years afterward seems silly, but looking at this with an Archi-enhanced perspective makes me wonder: should he really have gone for the double switch in the sixth? If we reconstruct the game had he left Archi in at first, Worrell can still pitch 1.2 innings since the pitcher's spot didn't come up in the bottom of the inning. Even if Bochy wants to use Worrell for longer, why not bring in someone else to get the two outs in the sixth and have Worrell pitch the seventh and eighth? Worrell ultimately did give up the tying run, not that Bochy could have known that. But had Archi stayed in the game, Bochy could have had Vaughn face Honeycutt in the seventh instead of handcuffing himself to an unfavorable Joyner-Honeycutt matchup. If Vaughn reaches base, then Caminiti's home run becomes a 3-run bomb. Alternately, LaRussa would have had a right hander face Vaughn, meaning he either has to waste a pitcher for one out (and he had to have a new pitcher for the inning since he had pinch-hit for the pitcher in the top of the inning) or he leaves a right-hander in to face Rickey-Gwynn-Finley. Even if Honeycutt retires Vaughn in the seventh, the Padres still have a better chance in the eighth since they still have Joyner on the bench, which means LaRussa might leave Honeycutt in the game to face Archi to deter a Joyner pinch-hit, which is a much better matchup for San Diego than Mathews-Vaughn, or he brings in Mathews and Bochy can put in Joyner, which is a much better matchup than Mathews-Vaughn. And if that batter reachers base, Johnson's double scores a run and gives the Padres a runner on second with none out. The only downside I see to not doing the double-switch, aside from the likelihood of Worrell (or whomever would go in if Bochy really has a jones for Worrell pitching more than an inning) can only go two thirds of an inning, is that if Vaughn hits in the seventh and Joyner hits for Archi then that at bat in the ninth slot the second time around goes to Chris Gwynn. This is all just speculation, and the Padres weren't too likely to win the series anyway, but I have to wonder if taking Archi out actually cost the Padres the game. How likely is it that anyone would ever make that statement?

The day after being pulled from Game 3, Archi turned thirty. That made 1997 the first season where Archi was officially on the wrong side of thirty (from a baseball standpoint, anyway). That wasn't enough to stop Archi. He logged the most playing time he'd had since his first year with the Padres and had his only season in San Diego where his total defensive contribution was at a level above league average. That combined with league-average hitting made him good for 2.1 wins above replacement (using WARP3), nearly twice as good as his next best season. He once again played first base frequently against LHP and played third base frequently behind an ailing Ken Caminiti, and also logged innings at second, short, and right field. His career year wasn't enough to keep the Padres out of the cellar in the NL West however, as poor seasons from Finley, Vaughn, and Andy Ashby combined with the losses of Bob Tewksbury and Scott Sanders to make the '97 Padres inferior to the '98 version.

Cut to 1998. The Padres replace Derrek Lee's far-off potential with Kevin Brown's career year. Andy Ashby, Sterling Hitchcock, and Greg Vaughn rebound from bad years with the best seasons of their respective careers. The Padres cruise to the playoffs and make the World Series. Unfortunately, Archi is left behind. After starting the year hitting a paltry .143/.213/.250, Archi went down with a strained oblique muscle in late May. He returned a month later but was underwhelming before being sent down to AAA Las Vegas. He rejoined the Padres in September, appearing in only four games and coming to the plate once, striking out against Terry Mulholland in a 6-3 loss to the Cubs on September 16. Archi's plays his final major league game in a 3-2 loss at Arizona, playing second base in the bottom of the eighth. No balls were hit to him.

Archi Cianfrocco was released by San Diego on November 17, 1998, and probably forgotten by most people who weren't Padres fans in the mid-90's or Purdue Boilermakers fans in the mid-80's. But he will never be forgotten by the few of us who saw his name in 1993 and knew instantly that his was our generation's best sports name.

Sunday, August 29, 2004

Notes on Player Evaluation

This blog has been online for less than 24 hours and I've already gotten a fair amount of hits. Based on the feedback so far (a ridiculously small sample size), I'm willing to believe that if I continue this quality of work some people will return regularly and perhaps gve a good deal of weight to my arguments. This prospect is fairly exciting for me (and maybe it is for you too-- just think, three years from now you could be bragging to your friends about how you'd been visiting The Fourth Outfielder since Week One). It should be noted, however, that my baseball expertise (if it can even be called that) is confined to a limited set of observational tools. I only watch 60-70 baseball games a year on TV, I've only been to one MLB game in person in the last fifteen years, and I haven't played baseball in over eight years. As such, I am a child of secondary sources when it comes to non-statistical player evaluation. I strongly believe that scouting and statistics are more or less equal in both their powers to deceive and their powers to predict, and any smart baseball decision almost certainly involves a careful consideration of both. That being said, I think I have a good feel for many areas of baseball statistics and I derive a bizarre level of enjoyment from hunting down obscure data (if you don't think I was loving the two or three minutes spent on Retrosheet.org finding the last time Robin Ventura faced Kris Benson, you're sadly mistaken). I honestly can't tell you anything about James Loney's swing other than that most people who've seen it speak of it as if it was the Venus di Milo with arms, but I can take a good look at how is Double-A stats compare to other 20-year old first basemen; more importantly, I'll try my best to show how the stats illuminate his raw talent and how the prevalent observations of his raw talent illuminate his stats. My analysis won't always be perfect, but I genuinely want to do what my favorite writers have done: offer perspectives on issues that many readers wouldn't have thought of on their own, offer analysis of different perspectives that synthesizes them effectively, and make readers enjoy reading.

Jason Kendall, Part One

In this morning’s Around the Majors over at The Hardball Times, Lee Sinnis noted that the Pittsburgh Post-Gazette reports the Dodgers are interested in acquiring Jason Kendall. Here’s the original (http://www.post-gazette.com/pg/04241/369318.stm, 8/28/04):

Another player who was placed on waivers is Jason Kendall, and the Dodgers have
expressed interest in acquiring him, just as they did before the non-waiver
deadline July 31. However, sources said a trade, if one happens, would more
likely occur in the offseason.
Kendall will earn about $1.6 million for the rest of this year. He’s on the hook for 10, 11, and 13 million dollars for 2005, 2006, and 2007, respectively. I have no good idea how much of that the Dodgers would have to take on in any of those years in a trade.

For the sake of argument, let’s say that they trade David Ross and Chuck Tiffany to the Pirates for Jason Kendall and the Pirates pay about a third of Kendall’s salary in each of those years so that the Dodgers owe Kendall $23 million for 2005-2007, plus $1 million for the rest of 2004. From a Pittsburgh point of view, this is a good deal, and I can’t imagine them turning something like that down.

Right now, the Dodgers catcher situation is slightly above replacement level. Using Wins Above Replacement (WARP1) from Baseball Prospectus, which measures both offensive and defensive contribution, Brent Mayne is 0.4 wins above replacement level for the year and Dave Ross .6 wins above replacement. That’s one win above replacement over 78 adjusted games at catcher. Kendall has 5.6 wins above replacement in 115 adjusted games. So if Kendall played 25 games over the rest of the regular season (the Dodgers have 32 left), we could expect a contribution of 1.22 WARP vs 0.39 WARP for Dave Ross in that same time, meaning we could expect the Dodgers to finish with one extra win in the standings. With the extra days off in the playoffs, Kendall would play every day and, if the Dodgers were to play the maximum number of playoff games, 19, could contribute an expected 0.92 WARP versus 0.24 WARP for the Mayne/Ross combo. Given the Dodgers’ 5-game lead and that the Padres and Giants are not great bets to outplay the Dodgers over the next month, the extra game advantage is probably not worth much. The real question is how much of a difference Kendall would make in the playoffs.

This is why I need to have a blog. To answer a somewhat simple question like this, I made an entire chart of the WARP for the playoff lineups of each of the likely NL playoff teams. I figured that all of the regular position players would play every game, so I used the Davenport cards at Baseball Prospectus to find each player’s WARP and divided it by AdjG (resulting, basically, in wins per 9 innings). I then totaled each team's WARP/AdjG and multiplied by 5 and by 7 for a 5-game series and a 7-game series (and also by 162 for contrast). Since pitchers don't play every day, I prorated pitching innings by choosng the four likely playoff starters, multiplying their WARP/AdjG by their IP/GS and dividing by 9 and then dividing by four (since there are four starters, and though starting distribution is uneven over the course of the series this is still a roughly accurate snapshot). That left several extra innings of pitching per game, and I arbitrarily assigned inning workloads to each relief pitcher based on their past workload and how I expected they'd be used. The innings were assigned so that a team would total nine innings of pitching per game-- that is, the sum of the coefficients of each players WARP/Gm would be 1. Yes, I probably made that more confusing just now, but if you have questions about my methodology email me and I'll send you the Excel file. Here are the numbers:

Ramirez0.048869C. Jones0.043478
5-game wins3.274595-game wins3.34444
7-game wins4.5844267-game wins4.682216
162-game wins106.0967162-game wins108.3598
CardinalsWins/GameDodgersWins/GameKendall wins/G
5-game wins3.4737685-game wins3.062443.24139
7-game wins4.8632767-game wins4.2874164.53795
162-game wins112.5501% improvement5.84348
162-game wins99.22306105.021

It should be fairly obvious that this is nothing more than an educated snapshot, but it's probably pretty useful. Remember, this is a projection of how successful teams would be in the playoffs using playoff lineups and rotations (I didn't work pinch hitting in, but I don't think there are any real standout pinch hitters). As it stands, the Cardinals are substantially better than the other teams, the Braves are second best, the Cubs just behind them, and the Dodgers are the worst team. With the projected value of Kendall, however, the Dodgers close most of the gap. Here are the estimated win percentages for the Dodgers against each opponent, with and without Kendall:

Without KendallWPCTWith KendallWPCT

Is getting just under 6% better worth taking on a million in payroll and giving up David Ross' potential? I feel much less sure about this than I did about the A's in my Jeff Kent post. The Dodgers stand to improve about 5.8% by increasing their payroll (in our hypothetical situation) by about a million from its current 90-100 million figure (I would get a more exact figure but I haven't seen the amounts of cash that changed hands in the D'backs and Marlins deals). So a 1% increase in payroll could yield a 5.8% increase in results. I think that, all else being equal, that is certainly a good deal (compare to the A's and Kent: a ~2% increase in results with a 3% increase in payroll). All else may not, however, be equal; in part 2 I'll look at the effects of acquiring Jason Kendall for 2005-2007.

More A's Second Base Speculation

Ken Arneson, who is, to my knowledge, the first person to have read an entry on this blog, pointed me in the direction of Ray Ratto's column on Kent-to-the-A's in today's Chronicle, which made a similar argument to mine. Ratto argued that a better choice would be to acquire Placido Polanco. Let's compare Polanco's splits to the McLemore/Scutaro combo (Scutaro against LHP, roughly 1:1 Scutaro to McLemore ratio against RHP):

Polanco: .315/.351/.490 vs LHP, .255/.321/.356 vs RHP
Scutamac: .292/.341/.513 vs LHP, .270/.325/.338 vs RHP

Looks like about the same run production. Polanco is, however, a better second baseman defensively. If you prorate his FRAA/AdjG over the next 30 games, he saves 4 runs over a 2:1 Scutaro to McLemore ratio, and about two and a half runs over nineteen playoff games. So a Polanco deal would probably provide about the same difference in production as a Jeff Kent deal. Would it be worth it for the $800K or so left on Polanco's contract? Possibly. Would it be worth it for the $800K or so left on Polanco's contract on top of a decent prospect? Given the margin for error with the A's budget, I doubt it.

Free Hee Seop Choi Part 1

Day two of the Fourth Outfielder Baseballl Blog, and Hee Seop is already on the bench. I guess this is because Robin Ventura is “hot” (.250/.368/.438 in 19 PA) in August while Choi is “cold” (.175/.314/.250 in 52 PA) and because Robin Ventura’s career numbers against Benson are better than Choi’s (.353/.421/.765 in 19 PA vs. .333/.333/.444 in 9 PA). Let’s revisit the last time Ventura hit against Kris Benson: June 25, 2000. Ventura hit .232/.338/.439 that year; going into that game, he was hitting .246/.360/.512 on the year and was a few weeks shy of his 33rd birthday. Benson was a 25-year-old with 304.2 major league IP under his belt at the time, brandishing a career 3.66 ERA. I think it’s fair to argue that the Ventura’s 19 plate appearances against Benson in 1999 and 2000 are not a great basis for starting Ventura over Choi.

When a player’s only hitting .175 in a month, benching him a few times might be a good idea. But when a player’s hitting .175 for the month of August while being repeatedly benched after putting up a .270/.388/.495 line in the first four months of the season, giving him more playing time is probably a good idea. I'll admit I haven't done any research nor seen any that talks about relative success of players based on the amount of playing time they're given while slumping, but my fan's intuition tells me that players will improve with more playing time, and I'd love to see Choi have more opportunity to develop his hitting.

Today, Robin Ventura came up with runners on 1st and 2nd and two out in the first and fouled out to first. In the fourth he came up with Green on first and none out and reached on Cliff Floyd’s error, scoring the second of the Dodgers 3 runs in the inning. Had Floyd caught it, the Dodgers wouldn’t have scored any runs. In the fifth inning, Ventura hit a grand slam. He did this by hitting Kris Benson’s 97th pitch of the game just over the right field wall after Benson had just walked two hitters and had yet to retire a batter in the inning. I appreciate the grand slam, but I’d have to say that I’d like Hee Seop’s chances in that situation. In the seventh Ventura flew out. So as of this moment, Ventura is 1 for 4 with a home run. Through Ventura’s first four plate appearances today, here is how Choi’s Dodger career compares to the players who have taken plate appearances from him while with the Dodgers:

As PH / Late Game Replacement40000000000
As starter528911400.34620.23080.21352.037
PH for Choi620000.33330.33330.23333
Lineup spot after PH501000.200.090
Non-Choi/Green Starters vs RHP255100110.240.360.1981.555

Now, that's certainly a tiny sample size, and Choi has been bad, but given that Choi has outperformed his replacements in both the small sample size of July 31-present and in the large sample size of the entire season, and given that he's the best defensive first baseman on the team, sitting Choi against RHP just doesn't make sense. I understand the desire to get Jayson Werth into the lineup, but Choi is a high-ceiling young player with a high-probability of reaching his ceiling. Werth, on the other hand, has been productive in the minors but is not a strong bet to be an above average major league starter. Almost all of his perceived value comes from his dominating Vegas this year and his .267/.343/.511 line in just 197 PA this year. He's likely best suited to being a (here it comes) Fourth Outfielder to platoon with a big-platoon split lefty (his 2004 GPA splits: .345 vs. LHP, .245 vs. RHP). Given the choice between developing Choi and developing Werth, there's no question the Dodgers should focus on developing Werth. Pinch-hitting for him against left-handed relievers in high-leverage at bats does make sense, but pinch-hitting for him whenever there's a southpaw on the mound (Choi has only 4 PA against LHP while with the Dodgers) doesn't seem beneficial since Olmedo Saenz and Jose Hernandez aren't exactly God's gift to hitting and Dodger hitters occupying Choi's lineup spot after he was pinch-hit for (the "Lineup Spot after PH" line above) aren't doing to well (most of these PA came from Robin Ventura).

Hee Seop Choi is the better Dodger for the future and the better Dodger for the present. The more playing time he receives, the happier I will be.

Jeff Kent to the A's?

I’m not sure whether I like Peter Gammons or not. It’s easy to dislike his analysis and it’s easy to dislike his writing, but does he writes at a site that prominently features Buster Olney and Joe Morgan and I do go to ESPN.com several times a day because I like their player cards and scoreboard better than the other corporate sportsites (and yes, I’m the kind of guy who finds reason to check player splits several times a day). And I do get a kick out of transactions gossip, and he’s probably the best in the biz in that regard (not to mention that he has his own unique color scheme and font). So today, when I saw the link to his newest column, I quickly read it. The following paragraph maybe tells you all you need to know about what to expect from Peter Gammons:

As the race for the final spots in the American League playoffs appears to be
coming down to the final weekend -- with an unusual importance placed on
Boston's upcoming consecutive series with Anaheim, Texas and (at) Oakland -- the
balance of the league is such that, heading toward the Labor Day turn, one can
make an argument for almost every one of the contenders going all the way to the
World Series. And that's with the added hope that Troy Glaus and Trot Nixon will
return for the final few weeks, and that somehow Oakland ownership will decide
the $1.8 million it would cost to have Jeff Kent
is worth what they will lose if they miss out on the playoffs after four
straight years in the postseason.

That is good writing, if you ask me, because of it’s emotional appeal, but it’s also not entirely logical, either in terms of analysis or syntax. Gammons can always capture an emotion of excitement, and it’s tough to enjoy baseball without an emotional attachment. At the same time, I have no idea what, if any, argument he’s making other than 1) the AL is exciting 2) any of the teams with a strong probability of making the playoffs could, upon making the playoffs, win seven games to reach the World Series. He points to the potential arrival of three players without really making an argument about what their potential impact would be. The one that piqued my interest most was the question of Oakland acquiring Jeff Kent. Later in the article, he clarifies that the A’s, in fact, need Kent—well, sort of:

The A's need Kent because of their vulnerability against lefties, but then
neither New York nor Boston has a left-handed starter or a left-handed reliever
that dominates left-handed batters.

It’s true that Kent is good against southpaws; this year he’s hitting .296/.348/.531 (a .290 GPA) against them, and from 2001-2003 he hit .340/.421/..577 against them. But exactly which A is vulnerable to left-handed pitching? Dye, Chavez, Hatteberg, Byrnes, Durazo, Kotsay, Scutaro, and Miller all have OBP’s of at least .340 and SLG’s of at least .400 against lefties (and Kielty has been decent against lefties when he’s played). Only Bobby Crosby has been bad against lefties, hitting .200/.313/.418 (a.245 GPA) which is not something you’d love to have in the lineup but which certainly isn’t terrible, and that probably has much more to do with sample size than anything else—Crosby is a right-handed hitter. I don’t have his minor-league splits available, but I can’t say I think his line in 130 PA is enough to justify adding $1.8 million in payroll. Of course, the A’s who are doing well against lefties are also doing so in small sample sizes. Here are the 2001-2003 lines for the A’s lefty regulars against LHP:

Kotsay: .269/.335/.400 (.297/.366/.432 in 2004)
Durazo: .247/.344/.398 (.295/.346/.475 in 2004)
Hatteberg: .234/.330/.357 (.295/.357/.403 in 2004)
Chavez: .229/.278/.395 (.326/.426/.558 in 2004)

All four are outperforming their past platoon splits, though it should be noted that all of them are outperforming their past lines against RHP by roughly the same amount except Chavez. Considering that Durazo, Hatteberg, and Chavez are all killing RHP this season while Kotsay has been as productive as Kent against RHP and is the A’s best defensive centerfielder, I’m not really sure who should be coming out of the lineup to add Kent’s bat. I guess Hatteberg is the most likely candidate, and Kent does have some experience at 1B. That would give the A’s maybe one extra base for every ten plate appearances vs. LHP that go to Kent instead of Hatteberg, so with 30 games Kent would contribute about 3 total bases, which means maybe a run or two. Given the data in my previous post about lefties in the playoffs, I don’t think the Kent over Hatteberg vs. LHP advantage is worth $1.8 million.

But the southpaw issue is really a form of beating around the bush: the real reason that Kent-to-the-A’s rumors and phone calls have happened is that Kent is a big-name second baseman and Marco Scutaro is not. If Mark Ellis was healthy, I doubt anyone would be talking about the A's picking up Kent.

Kent is an offensive upgrade over Scutaro, but not against lefties:

Kent, 2004: .296/.348/.531 vs. LHP, .280/.334/.486 vs. RHP
Scutaro, 2004: .292/.341/.513 vs. LHP, .274/.287/.348 vs. RHP

Given that Kent hits at the House that Orange Juice built (well, technically Andersen accounting should get the credit) and Scutaro hits at the stadium famous for “My Sharona,” those numbers against lefties are pretty much identical. The difference in between Kent vs. RHP and Scutaro vs. RHP, on the other hand, is quite substantial; Kent has a .272 to .216 edge in GPA. Scutaro doesn’t draw walks or hit for power against right-handed pitching and Kent does. This is somewhat mitigated because the A’s can play Mark McLemore against RHP, and he’s hit .266/.363/.328 (.245 GPA) against them this year, pretty much the same as Scutaro except McLemore draws walks. Since there’s nothing to suggest at this point that Kent is a significant defensive upgrade over either McLemore or Scutaro, that means the question is whether the A’s should add $1.8 million in payroll and presumable give up a mid-level prospect or two to boost their offense against RHP from .230 GPA (assuming McLemore takes about half the PA vs. RHP) to .270. The A’s as a team have had 3624 plate appearances against RHP this year in 129 games, an average of 28.1 per game. If Kent took a ninth of the A’s PA against RHP over the final 30 games, that would be 94 PA. If he approximated his current line against RHP then he would create 14 runs using the basic Bill James RC formula. A Scutaro/McLemore combination with a .270/.325/.338 line creates nine and a half runs. There are a maximum of 19 games in the playoffs, and let’s just say for the sake of argument that the A’s would only face RHP throughout the playoffs (if they played the Red Sox, Yankees, and then Cardinals then they would only face a few lefties out of the pen). That would mean about 760 PA if they average the same number of PA per game they currently are. That means we have 84 PA to give out to A’s second basemen, and Kent projects to creating 12 runs in that span while the Scutaro/McLemore combo projects to eight and a half. So a fair expectation for the difference between Kent and what the A’s have now is four and a half runs in the regular season and up to three and a half runs in the playoffs (remember, it’s likely that the A’s will have fewer PA/game in the playoffs and it’s likely they will play fewer than 19 games in the playoffs). That means the maximum expectation for additonal productivity from Kent is eight runs. It could be that Kent would get hot, of course, but it could be that he would get cold or that Scutaro/McLemore gets hot, so it doesn't make sense to make a decision expecting more than eight runs in September and October.

Eight runs is, in some senses, a lot. Over the course of a season, it’s about an entire game in the standings (check out Win Shares—Kent has a 3 win share lead on Scutaro, which equates to one win). When you’ve lost Game 5 in the division series four years in a row, you might want to do whatever it takes to get an extra run per week. So it’s certainly understandable that the A’s would have a desire to pick up Jeff Kent, it’s also unrealistic to expect them to want to do so. Paying $275,000 per expected run could make the difference between missing the playoffs entirely and winning the World Series, just as spending a first round draft pick on a high school pitcher with upside instead of a college pitcher who’s almost ready could make the A’s substantially better. The point is, not making the deal is probably the smart play since, unfortunately, money in baseball is finite and those 18,000 Ben Franklins could probably make more of a difference if used to get better players next year.

Gammons is, perhaps, absolutely right on an emotional level: as a baseball fan, I want to see the A’s get better (both because it makes the season more interesting and because, outside of the Dodgers, I’d rather see the A’s win than anyone else). And in looking at what Gammons wrote, it’s very clear why it’s so easy to want the A’s to pay $1.8 million for eight runs: because that money doesn’t belong to the fans, it belongs to the A’s ownership. We feel that they simply should spend more money. And the most painful but most necessary lesson that any baseball fan can learn is that, from a fan’s standpoint, that is our money, and when we think about what our team does we have to consider the financial implications because the ownership will do so without exception.

Platoon Splits For The Playoffs

One thing I spend entirely too much time looking up is platoon splits. For some reason, I have probably compiled this list in my head a dozen or so times in the last month, and have posted versions of it in the comments at Dodger Thoughts twice. Anyway, this is a list of all of the left-handed pitchers playing for contenders, with their lines this year against left-handed batters (avg/obp/slg).

St. Louis has no lefty starters and 2 effective lefty relievers, Steve Kline (.150/.258/.200) and Ray King(.153/.240/.165). Theoretically, Ankiel could see playing time in the playoffs, and he's been dominating in A and AA, but I'm not sure how necessary he'd be with Kline and King being even more dominant.

Chicago would not start any lefties in the playoffs and would likely have 3 effective lefties in the pen: Glendon Rusch(.227/.288/.268), Mike Remlinger( .261/.288/.435), and Kent Mercker(.247/.315/.358).

Atlanta would likely start Mike Hampton(.269/.307/.369), and would have Effective Left-Handed Setup Man Who Anchored Dodgers Bullpen Chemistry Tom Martin(.308/.380/.462) out of the pen. My guess is that Horacio Ramirez(.224/.316/.388) probably wouldn't see much action even if his injury clears up.

Los Angeles would start Odalis Perez(.262/.301/.475). Wilson Alvarez (.304/.340/.522) will be used in some capacity; hopefully Kaz Ishii (.265/.366/.419) will succeed in very low-leverage innings. The Mike Venafro(4 PA, .000/.500/.000, 0 BB, 2 HBP) and Scott Stewart(.351/.400/.459) parade could factor in somewhere.

The Giants will probably start Noah Lowry(.319/.333/.468 tiny sample size) and possibly Kirk Rueter(.287/.325/.387), and Wayne Franklin(.236/.300/.431), Jason Christianson(.234/.351/.313), and Scott Eyre (.227/.259/.467) will all log innings for the official bullpen of Rolaids (and not because they spell relief).

San Diego would likely start David Wells (.282/.292/.515), and a Sterling Hitchcock appearance (.300/.500/.700, 16 PA) is always possible if his injury situation "improves".

Oakland: Mark Mulder (.250/.323/.350) starts, one or possibly both of Barry Zito(.350/.451/.504 in 165 or so PA) and Mark Redman (.298/.361/.461) starts (whomever doesn't start probably takes a few innings away from one of the A's seven relievers). Chris Hammond(.275/.283/.333), Ricardo Rincon(.182/.214/.242), and Arthur Rhodes(.317/.388/.512) relieve.

Boston: Alan Embree (.269/.282/.397) and Mike Myers (.233/.329/.356) relieve.

New York: the greatest word in baseball-- Nitkowski. (.271/.375/.417)
Minnesota: Honus Santana (.191/.232/.331) starts, J.C. Romero (.250/.316/.295) relieves, and Terry Mulholland (.286/.331/.419) mans the torture chamber in Aaron Gleeman's nightmares (but might actually be the best option to start game 3 instead of Lohse/Silva) .

Texas: the Gambler (.307/.376/.420) starts; Ron Mahay(.209/.286/.267) and Brian Shouse(.169/.215/.324) would relieve; Erasmo Ramirez(.333/.385/.500) could appear.

Anaheim: "Ace" Jarrod Washburn(.225/.273/.396) will start if he's recovered from injury in time.

Not a particularly bad year to have a big-platoon-split-southpaw-stocked starting lineup, unless you're facing the A's.


Hello, and welcome to the Fourth Outfielder Baseball Blog. You have probably been directed here by a generous link in another blog, and I thank you for trying that link. I had no particular reason for picking "Fourth Outfielder" as a title; I thought it was catchy enough and didn't want to go for any of the "Above Average," "Above Replacement," or "VORP"-related puns that I had thought of. I'm a college student (at least, that's what I'm called-- I don't catch myself studying anything related to my classes too often) at UC Berkeley, and I hail from various Northern California locations. I'm a Dodgers fan, although I have loyalty to the A's as well. Both of those allegiances long predate the Beane and DePodesta eras, but each GM has substantially increased my enthusiasm. My current favorite Fourth Outfielder is, predictably, Jayson Werth, although it's sometimes hard not to channel my anger at Hee Seop Choi being left out of the Dodgers' lineup against RHP at Werth. My favorite blogger is probably Aarong Gleeman, and I hope my "Free Hee Seop Choi!" feelings will yield as much eventual success as his "Free Johan Santana!" feelings have. I have thought about starting my own blog for a long time, and very nearly did back in March. At the time, I was planning on calling it "Adrian Beltre Wins MVP!" or something along those lines to reflect five years of Adrian Beltre frustration that could never quite quell my enthusiasm for him. At the beginning of the year, I thought he could breakout to be a .290/.340/.520 hitter or so, but thought it more likely that he's hit .245/.285/.440 in the first half and then .290/.330/.480 in the second half, making the decision to bring him back even more frustrating. Now that the player I've had the most emotional attachment to as a fan has gone from chronic frustration to one of the top 5 players in the league (in terms of performance this season, anyway), I'm stuck sweating out the contract negotiations that surely will not Bore Us (and if you get that joke, you've come to the right place, though you might be regretting doing so) while poring through the career stats of Ron Santo of Mike Schmidt and wondering if I can actually believe what Buster Olney wrote about Tim Wallach working on Beltre's approach. Oh, that and wondering how Win Shares can value Rolen's offense at 24% more than Beltre's offense when VORP has them equal (post on this hopefully forthcoming). Oh, and did I mention that I don't live in the LA area, don't have DirecTV or a connection fast enough for MLB.tv? That's right, I've been a huge Dodger fan for twelve years, the majority of my life, and only get to watch them play the Giants and Braves and the occasional ESPN or Saturday afternoon game. It wasn't until the past year that I've heard Vin Scully call a game. So I hope this first entry provides a nice snapshot of what kind of perspective you can get from this blog, and hopefully you'll check back in a few times. And you can email me at meagher at berkeley dot edu with comments or suggestions or bizarre research assignments.

This page is powered by Blogger. Isn't yours?