Losing the WAR, Part One

Aramis Ramirez had some tumultuous seasons before he landed with the Cubs (via Sean Benham).

Aramis Ramirez had some tumultuous seasons before he landed with the Cubs (via Sean Benham).

The concept of Wins Above Replacement, or WAR, has wrought a far-reaching change in how analytical fans view baseball. It’s created a baseline by which to judge both historical and contemporary performance. It’s provided a holistic view of players’ worth balanced between the offensive and the defensive (even if the defensive side may still be a work in progress). It’s given fans an objective standard with which they can gauge when a struggling player on their favorite team may be doing more harm than good and should clear the way for an eager rookie or an unattached veteran looking for another chance.

That latter species of second-guessing has always been the fans’ prerogative, but having figures to put behind it gives it a sharper edge. If there’s this objective, pitiless number saying some player ought to get off the field, and the fans can see what it says, then surely the front office can, too. If they’re so smart, they’ll act, soon.

They usually do, but sometimes it takes a while—and sometimes much longer than a while. The existence of WAR as a statistic has made it more obvious, but there have always been players who have hurt their teams and kept getting the opportunity to hurt them. I decided to learn who those escapees from their own under-performance were.

I’ve gathered together the players with the most accumulated seasons at levels below replacement, pitchers and position players separated, going back to 1901 and the emergence of the American League. In a future installment, I will track down players who had the worst seasons for teams that survived those individual struggles and succeeded, from wild cards all the way to winning the World Series. Between the two, we may learn what can lead a team to over-estimate a player’s worth, or put up with someone they know is a hole in the lineup.

Second Chances, and Third, and Fourth …

Despite being able to produce numbers with multiple decimal points, WAR is still not a precise statistic. Defensive metrics remain iffy enough that they can miss a player’s actual value, perhaps substantially so. Therefore, a safety factor in judging a player to be performing below replacement level is a sensible caution.

I’ve provided two such cushions. I ran the search for player seasons with a maximum of -0.5 WAR, and a maximum of -1.0 WAR.

(Note: I used Baseball-Reference for my study, rather than FanGraphs. This does not imply superiority of bWAR as a measure over fWAR. It springs from the easier searchability of B-R’s historical data. It also provided an added reason for using a buffer: What comes out as mildly negative with one method could be zero or positive with the other.)

I’ll begin with the position players and the cutoff of -1.0 bWAR. Listed are the number of seasons the leading players (misleading term, I know) had at or below that level, and what their first and last qualifying years were. As you will see, the phenomenon of fooling the front office has not yet vanished.

Position Players with Most < -1 WAR Seasons
Name Seasons First/Last Year
Bill Bergen 7 1902/1911
Johnnie LeMaster 5 1977/1985
Doug Flynn 5 1977/1984
Juan Castro 4 2001/2010
Jose Guillen 4 1997/2009
Andres Thomas 4 1987/1990
Dalton Jones 4 1964/1972
Ted Kazanski 4 1953/1958
Billy Hitchcock 4 1942/1949

Two players out of nine stretching their chronic under-achievement into the 21st century isn’t dominant, not when one takes expanded leagues and longer schedules permitting more time to pile up numbers into account. It does show, though, that a sabermetrically-armed front office is not immunized from hanging on too long with some players.

I will defer comment on the top/bottom man on this list for a while, going instead to runners-up, plus a few who fell a bit short.

Johnnie LeMaster spent most of his 12-year career exasperating the San Francisco Giants. His OPS+ never reached 80 in a season, and despite playing shortstop virtually his entire career, with its inherent defensive worth, his total defensive value comes in just below replacement level. Presumably his position led the Giants to overestimate his contribution—for more than 11 years.

His final sub-minus-one season was 1985, when he was bounced from the Giants to the Indians to the Pirates, digging down to -1.2 bWAR in just 45 games. He actually got another chance with Oakland in 1987, but they gave up on his .203 OPS before he could get below -0.4.

Doug Flynn opened his career on high notes, as a backup on the 1975-76 Big Red Machine. His 85 and 83 OPS+ scores those years would be career bests, and it was straight into negative WAR territory from there. He’d surface above zero once, in 1980 as the Mets’ regular second baseman, a Gold-Glove year redeeming his accustomed weak bat. His worst year would be a short two seasons later, his combined -2.5 bWAR for the Rangers and Expos worst in the majors—just beating out LeMaster.

Both Juan Castro and Jose Guillen churned out sub-negative-one years deep into the 2000s, not exactly a ringing endorsement of front offices in the Saber Era. Castro’s anti-accomplishment is perhaps more noteworthy in that he was a permanent part-time player: He never topped 348 plate appearances in a season. His offense never was strong, and his utility infielding was up-and-down. When both took a dip in the same year, he climbed up the above list.

Guillen was a better player over his career than Castro, but he started with absolutely terrible seasons in 1997 and ’99. He improved, stumbled hard, then caught fire, posting a 4.7 in two-thirds of a season with Cincinnati. He went flat after a deadline trade to Oakland, but his next two seasons with two different clubs both went over 3.0 bWAR before he had another crash that just missed the -1.0 cutoff. A good bounce-back 2007 in Seattle netted him three years and $36 million with Kansas City, where he crashed and burned, a -2.0 in 2009 the nadir.

There are some interesting names among those with three seasons below -1.0 bWAR. Recent ones include Yuniesky Betancourt, Jeff Francoeur, and Aramis Ramirez. Francoeur was a fan-favorite whipping boy in 2012 and ’13, but kept getting chances after a 3.0-WAR 2011. Weirdly, this repeated what happened in Atlanta: a good 2007, an atrocious 2008. Betancourt is a little more inexplicable, peaking at just 2.0 bWAR in 2007 before six straight sub-replacement years that seem to have ended his MLB career. (He’s now with the Orix Buffaloes in Japan.)

Ramirez is a much different animal. Two of his terrible seasons came at the start of his career in Pittsburgh. He then had a 4.0-WAR breakout and a -1.3 WAR crash-landing the next year. Pittsburgh pulled the ripcord and traded him to the Cubs, where he figured things out. In Chicago, he’s made three All-Star teams and had three top-10 finishes in MVP voting. He’s been a genuinely good player—despite a 2010 that just missed being his fourth season of -1 .0 or worse (at -0.8).

Back in time a little more, Zoilo Versalles, AL MVP in 1965, suffered three subsequent seasons below -1.0 bWAR, as three separate teams went down hoping that one season wasn’t a fluke. (Those hopes may well have cost the 1967 Twins the pennant in their fabled four-way race.) Billy Martin also rings up three clunkers (with four teams: he split 1961), all after the Yankees traded him away in 1957. Look at how his numbers, offensive and defensive, start taking a header in that season, and you begin to suspect that the notorious Copacabana incident was a lucky excuse more than the cause for his being sent out of New York.

Now for a look at the list using -0.5 bWAR as its cutoff. There will be carryovers, and a few other familiar folks.

Position Players with Most < -0.5 WAR Seasons
Name Seasons First/Last Year
Bill Bergen 11 1901/1911
Lenny Harris 7 1989/2004
Bob Kennedy 7 1940/1957
Jose Guillen 6 1997/2009
Juan Castro 6 1996/2010
Ed Romero 6 1981/1989
Bill Buckner 6 1977/1990
Ed Kranepool 6 1963/1979
Bob Lillis 6 1959/1967
Don Heffner 6 1937/1943
Tuck Stainback 6 1936/1946

The superior analytical discernment of the modern era takes another blow. This time, three of the bottom 11 in bad seasons played into the 21st century, with Lenny Harris joining Guillen and Castro. Harris lasted 18 years in the bigs, half of them below replacement level. A good 1990 season (3.1 bWAR, 102 OPS+) was the lone qualitative inducement for teams to keep giving him chances. The stronger factor was versatility: Harris played every position but catcher in his career, including a one-inning pitching stint for the Reds in 1998. (He twirled a perfect frame, with a strikeout. Perhaps he missed his calling.)

Poor Bill Buckner. First 1986 happens, and just when people are starting to forgive him for that, I come along and deliver this blow to his reputation. Fact is, though, that after the awful injury in 1975 that wrecked his left ankle, robbing him of his speed and defensive range, he became a good candidate for sub-replacement seasons. He hit for high but empty averages, with few walks and not very much power, and was stuck mostly at first base, eroding his value.

Any time Buckner had a sub-par year at the plate, he was primed for negative WAR. This happened in 1977 with the Cubs, and in 1984 when the Cubs traded him, mid-season and mid-slump, to Boston. It then happened in all four of his post-Mookie years, the final four of his career. It may be that a succession of managers and GMs didn’t have the heart to tell him he was finished. It may also be that this is how cliches like “being cruel to be kind” get started, with frustrated fans.

Whichever measure one uses, though, the undisputed champion of lasting the longest with the least was Bill Bergen. His career lasted 11 seasons, and the best bWAR he ever mustered was a -0.5 in 1908. He played as few as 51 games in a season (in 1907), and this was still enough time to dive to his accustomed depths.

The obvious reason for this appalling performance was his batting. Bergen had a lifetime batting average of .170, the worst ever for a career of any meaningful length, with few walks and no power. He played in the Deadball Era, but this was horrible even by those standards. His career OPS+ was 21, with a peak of 41 in 1903.

Cincinnati and Brooklyn endured this offensive offense because of Bergen’s reputation as a defensive catcher. His strong arm, knack for running down pop fouls, and studies of opposing batters were lauded, possibly too much so. He did have two years leading the National League in catching opposing stealers, in the only two seasons he got to play more than 100 games, but his caught-stealing percentage never led the circuit. Today’s defensive metrics aren’t bowled over, rating him as just slightly above average.

It’s certainly possible that modern defensive metrics cannot pierce the fog of more than a century, but consider that Bergen checks in at -13.5 bWAR for his 11-year career. How wrong would Total Zone Rating have to be, year after year, for Bergen to have been even a replacement-level player? The reputation he’s gotten from his abysmal batting average is probably the right one.

Throw, Duck, Repeat

Teams do not seem to tolerate as many awful seasons from their pitchers, but there are still plenty of multi-stinker careers to go around. We’ll start again with those at -1.0 bWAR or worse.

Pitchers with Most < -1 WAR Years
Name Seasons First/Last Year
Kevin Jarvis 4 1995/2003
Danny Jackson 4 1989/1997
Dan Schatzeder 4 1981/1989
Mike LaCoss 4 1980/1991
John D’Acquisto 4 1975/1981
Ray Burris 4 1974/1986
Blue Moon Odom 4 1967/1975
Les Sweetland 4 1927/1931
Ferdie Schupp 4 1915/1922

There’s a hint here that today’s saber-inclined executives have figured out how to stop giving bad pitchers too many extra chances. With seven of the nine men on this list coming after expansion, there’s also a hint that old-time teams had not picked up that particular weakness.

There’s no single leader on the -1.0 pitchers’ table, so I would use the -0.5 table to filter out the most outstanding mediocrities. This is complicated by that table consisting of one player with six such seasons and a clump of 21 pitchers with five. I’m not going to inflict that eyesore on my readers, so instead I’ll just begin with the top pitcher on that list: Kevin Jarvis.

(Before that, I will observe that eight of the 22 pitchers with at least five seasons at -0.5 bWAR or under had at least one of those season in 2000 or later. That’s a counter-balancing hint that today’s execs haven’t figured out how to cut the bad pitchers loose after all.)

Jarvis’s first four years in the majors were below replacement level, the first one fortuitously cut off by the 1994 strike before it could qualify for these lists. After three straight years of -1.0 or worse, he went to Japan for most of 1998. Oakland gave him four games in 1999, and Jarvis gave them -0.6 bWAR, his already bad homer-proneness exploding into six dingers over 14 innings.

Jarvis’ career appeared finished. Who could want a pitcher who struggled this badly? The answer was almost like a punchline: the Colorado Rockies. And it sorta worked. Despite an ERA and FIP of almost 6.00 and a month-long demotion to the minors, he managed his first positive contribution in the bigs, a 0.6 WAR over 115 frames. Remember, this was pre-humidor Coors Field. An ERA around 6.00 was not that bad for someone pitching half his games there. His ERA+ was a pleasantly middling career-best 98.

The Rox let Jarvis go in free agency, and he fetched up on San Diego’s shores. As a regular starter in 2001, he produced 1.5 WAR for the Padres despite leading the league in home runs surrendered. This earned him an Opening Day start in 2002, if you call facing Randy Johnson earnings. He took a hard-luck, 2-0 defeat against Johnson’s D-Backs, pitched three more games pretty well, then hurt his elbow. After two DL stints and two brief returns to the mound, Jarvis had Tommy John surgery. His emergence as a decent pitcher was over. He did terribly in late 2003 with the Padres, put up three sub-zero seasons bumping around the league, and that was his career.

A player’s failure is always a sad story when considered personally, but Jarvis brings an extra pang. After years of struggles, one could see an Aramis Ramirez-like emergence as, if not actually a good player, certainly a useful one. With the taste of redemption and vindication in his mouth, his body betrayed him, in a fashion much too familiar to us these days.

Three other pitchers on the minus-one table also had five seasons of -0.5 WAR or worse. The most recent is Dan Schatzeder, who bounced around with nine teams over 15 seasons. Three-WAR-plus seasons with Montreal in 1979 and Detroit in ’80 set the bar of expectations, which he immediately submarined with disaster years in ’81 and ’82. Four subsequent years of competence in Montreal preceded his next plunge with several different squads. Mourners at his retirement consisted mainly of Expos fans and a lot of juvenile boys who liked to snigger at his name.

John D’Acquisto and Blue Moon Odom join Schatzeder in this club. (Not the one about making fun of their names.) D’Acquisto had good seasons just often enough (i.e., twice) to tempt teams to hang on during his wrecks. Odom was similar, though with more good years and more stability, lasting with the A’s for the first 11-plus of his 13 years in the bigs. Odom’s specialty was outperforming his Fielding Independent Pitching (FIP) in good years so that they looked even better. His ERA came in below 3.00 four times, but his FIP never did.

I’ll note two other examples of pitchers whose disasters didn’t outweigh their successes. Danny Jackson had four seasons below -1.0 bWAR, all of which came after a four-year string always above 3.0 WAR, topped by a 5.0 with the 1988 Reds. He had a 4.7 as late as the 1994 Phillies (and that was a shortened season), with two crashes down and two to go. That’s a lot of variance in what shook out to be a long but average career (100 ERA+).

Then there’s Jim Kaat. Kitty’s one of those guys who keeps getting boosted for a Veterans’ Committee ticket to the Hall of Fame, with 283 wins and 16 Gold Gloves. Not helping his case is the five seasons he had at -0.5 bWAR or worse. Two came early with the Senators/Twins; three came late with the Phillies and Cardinals. I’ll leave sifting Kaat’s qualifications to other debaters, but those rough campaigns, one with the only World Series winning team he was on (the ’82 Cards), cannot help his argument.

Until Next Time

Plenty of teams have put up with a lot of retrograde performance from their players. Often this will be its own punishment for the front offices involved, as their clubs won’t go anywhere. In Part Two of this series, though, we’ll learn about teams that managed to fly despite a particular player’s dead weight. Nobody in the lists above had that ambivalent fortune, but there will be people you know, including one of the biggest names in the sport.

References and Resources

  • Baseball-Reference
  • Baseball Prospectus
  • SABR Baseball Biography Project
  • Wikipedia, for tracking down some player injuries
  • Bruce Markusen’s Bob Veale installment of his Card Corner at THT

  • A writer for The Hardball Times, Shane has been writing about baseball and science fiction since 1997. His stories have been translated into French, Russian and Japanese, and he was nominated for the 2002 Hugo Award.
    Newest Most Voted
    Inline Feedbacks
    View all comments
    Jim S.
    8 years ago

    You should really mention that Lenny Harris holds the record for most career pinch hits with 212. Hitting off the bench is rare attribute.

    8 years ago
    Reply to  Jim S.

    very good point, but 212 ab over the course of a pretty long career is not going to have that big of a difference. It does raise an interesting hole in WAR hitting calculations. We currently compare guys against the replacement for the position they play (or if it is multiple positions, then the weighted average of those positions), but i do not think that pinch hitting is calculated. it may be hard to figure out a replacement level for pinch hitters, a league average is likely already calculated somewhere. Considering those ab normally go to bench players, it may be easy to just say those guys already are replacement level pinch hitters.

    8 years ago
    Reply to  james

    There’s no hole in the WAR framework. There *is* a PH (and DH and PR and defensive sub) adjustment. (And similarly for relief pitchers.)

    Whether Fangraphs and/or Baseball Reference applies any or all of that, I don’t know.

    8 years ago
    Reply to  james

    Bench players are a little bit better than replacement level, though.

    Paul G.
    8 years ago
    Reply to  Jim S.

    Pinch hitters are going to have limited playing time, so they are going to suffer from small sample sizes which leads to big value swings. Even elite hitters have poor stretches from time to time, perhaps even sub-replacement, but they get a full season to smooth those numbers out. Lenny Harris did not have that luxury. Not that there was any real indication that he deserved said luxury.

    There is also the matter that players who are valued as pinch hitters tend to have a lower bar of value. For example look at Lenny’s 2000 season when he scored splits of 260/317/381 for a 77 OPS+. That is not good offense, especially for a guy who was playing the field primarily at third, the corner outfield, and first. However, who would he be pinch hitting for? Pitchers! The Mets pitchers batted 148/197/158. In that role Lenny Harris is quite valuable. Now whether he is worth a roster spot, especially in these days of 13 man pitching staffs cutting the bench to the bare minimum, is another question.

    Also keep in mind that relatively extreme but flawed hitters can be leveraged to some extent as pinch hitters. A guy who hits for a very empty 300 average can be sent to the plate in situations where getting any hit is quite valuable, negating the lack of walks and power. And if that is not possible, most likely he’s still better than the guy he’s batting for.

    8 years ago

    I bet the 2013 Atlanta Braves will be featured in the next article. Two players with sub -1 WAR is impressive.

    8 years ago

    How can Yuni not be listed here? He’s had 5 sub-1 seasons just in the last two years:-)

    Danny C
    8 years ago

    Lemaster changed the name on the back of his jersey to
    Talk about no love!!

    Jason Lukehart
    8 years ago

    This is a fun read, well done.

    One minor point, the article makes it sound like Aramis Ramirez has been in Chicago ever since leaving Pittsburgh, but he’s now in his third season with Milwaukee, which is where one of his All-Star appearances and one of his top-10 MVP finishes have taken place.

    John C
    8 years ago

    I thought for sure I’d see Todd Van Poppel’s name turn up on the list of pitchers, but it looks like he only had three sub-1 WAR seasons. Of course, he did manage to somehow rack up -2.6 bWAR in just 99.1 innings in 1996, so that should count for something.

    Van Poppel, to me, has always been a poster boy for why pitchers in this day and age can get too many chances even though they’re generally awful. GM’s and scouts see a guy who can throw a snowball through a blast furnace without it melting, and they can’t resist, even if he’s clueless on how to actually pitch. I know enough about John D’Acquisto to know that’s what kept getting him opportunities, and it continued until he finally blew out his arm. He and Van Poppel had the habit of being decent once every few years, which, coupled with the 98 mph heat, kept getting them chances no matter how bad they pitched.

    Mr Punch
    8 years ago

    Very nice. But Matt Young is still the worst player of all time.

    victoria secret uk
    8 years ago

    I don’t know what the start times were, but if they were before 8, the Vampires would have missed the first pitch :).vdvcvc

    8 years ago

    Seeing so many names of Giants players from the 70’s and 80’s, when I grew up with them, this article makes the mediocrity of that period all the more stark, particularly LeMaster. Apparently the fans saw something that the front office couldn’t see, it was their reaction to his performances that inspired him to wear that uni linked above. I wonder what Johnnie Lee is doing today, maybe selling t-shirts?

    BW Radley
    8 years ago

    Actually obsessivegiantscompulsive your guess regarding Johnnie “The Disaster” LeMaster is correct! He owns a sporting goods store in his hometown of Paintsville, KY.