Losing the WAR, Part Two

The fall came fast for Brad Lidge (via Darrin).

The fall came fast for Brad Lidge (via Darrin).

Last week, I published an article detailing players who have had the most seasons performing below replacement level in their careers. There are various reasons why teams would keep giving them chances. Sometimes the players set ceilings of performance high enough to lure clubs into dreaming they’ll reach those heights again. Sometimes the clubs misjudge players’ ability, for instance, thinking someone’s defensive skill offsets limp offense much more than it really does.

Those are the chronic cases. Today, I’ll be looking at the acute ones, but from a specific angle. An individual’s nightmare season usually will drag down his entire team or bury it even deeper in the loss column than it otherwise would have been. There’s a certain naughty pleasure to be gained from witnessing such ineptitude. Otherwise, teams like the 1899 Spiders and 1962 Mets wouldn’t have the places they hold in the lore of the game.

Sometimes, however, a team can ride out such localized disasters. It can drag itself to the top of the division, win a pennant or even the World Series itself. Today, I’ll be looking at examples of players whose failures did not leave their teams as also-rans.

Running the October Handicap

For both pitchers and non-pitchers, I tracked down the players who accumulated the worst WAR stats while their teams managed to reach the playoffs, for each level of accomplishment. (Again, I use Baseball-Reference’s bWAR stat as opposed to FanGraphs’ fWAR due to greater ease of searching at B-R.) I did omit the Wild-Card stage for the pitchers, as a divisional champion came in with a bigger boat-anchor than any of the Wild Cards. There are two players who spent only partial seasons with their playoff-bound teams. They are marked with asterisks and explained in my comments.

Everyone on the lists comes from the divisional era, though my search went back to 1901. Why did no old pennant winner crack the lists? Possibly because with today’s playoff structure, you don’t have to be the best team in the league to capture the flag, so a drag-chain player can be tolerated as long as he isn’t dragging you down in October. Or perhaps it’s luck: four teams isn’t a big sample.

We will begin with the position players.

Worst Position Player for a Wild Card Team

Nate McLouth, 2010 Atlanta Braves, -2.7 bWAR

In 2009, McLouth seemed like a good pickup for a Braves squad climbing back into contention. He’d had an All-Star 2008 with Pittsburgh, posting a 125 OPS+ with 26 dingers and a league-leading 46 doubles. He even picked up a Gold Glove, though both Defensive Runs Saved (DRS) and Ultimate Zone Rating (UZR) had him well below average in center field that year. He came over in a trade in June of ’09, and while there was an offensive drop-off, his defense simultaneously improved to average, so there was little front-office nausea going into 2010.

That didn’t last. Two months into the 2010 campaign, McLouth’s batting average was mired in the .170s, and while he was drawing walks, his power was kaput. His defensive metrics, meanwhile, had reverted to 2008 levels. He spent two stints with the Triple-A Gwinnett Braves, sandwiched around a brief call-up during which his numbers skidded even lower. Called back up for September, he started producing , but his fine-looking .273/.359/.527 numbers for that stretch could only drag his season stats up to a dismal .190/.298/.322.

One could at least comfort oneself that McLouth’s hot September helped pull his Braves back into the playoffs, except … Atlanta had a three-game lead on the Phillies when he came back up, but despite his strong month, they played under .500 the rest of the way. The Braves did edge the Padres by a game for the Wild Card spot, so one might argue that McLouth’s September tear got them through.

But the larger picture is still there: -2.7 bWAR in 288 plate appearances, effectively half a season. And considering that, minus his September, it would have been more like -3.0 bWAR or worse in a little over two months, McLouth may have put up the worst sustained performance we are likely ever to see a legitimate contender in the modern game tolerate.

Worst Position Player for a Division Winner

George Bell, 1993 Chicago White Sox, -2.5 bWAR

People like homers. Teams like homers, and they like players who hit homers. This can lead to those teams liking players who hit homers even after they have stopped doing other things to help the team (or, in cases like Dave Kingman and Steve Balboni, never started doing those other things).

George Bell hit homers, with a peak of 47 playing for Toronto in 1987 that, along with his league-leading 134 RBIs, netted him the AL MVP trophy. He would barely ever get past half that homer total in a season again, which, combined with his poor walk numbers and middling defense at a position well down the defensive spectrum, made him less valuable to the Blue Jays than he appeared. They’d let him go in free agency, with the Cubs and then the White Sox giving him chances.

Bell’s game faded in 1992 with the Pale Hose to a -0.2 bWAR, but his 25 dingers got him a smidgen of down-ballot MVP support—and since he had a two-year deal, he wasn’t going anywhere yet. Where he’d go in 1993 was straight down. His average, which had gotten over .300 twice in his career, crashed and burned at .213, even as what little batting eye he had deserted him. He still would have been on pace for a 20-homer season, but a July knee injury held him to a mere 13 in 436 plate appearances. As a full-time DH, his defensive contribution was exactly nil.

Luckily, the White Sox had some stout starting pitching and big offensive threats like Frank Thomas and Tim Raines making up for what Bell no longer could do. Chicago took the AL West with room to spare, but the ChiSox succumbed in a six-game ALCS to Bell’s old squad, the defending champion Blue Jays. Bell sat for the whole series, and with the proverbial writing covering the wall, retired after the season.

Worst (full-year) Position Player for a Pennant Winner

Pete Rose, 1983 Philadelphia Phillies, -2.1 bWAR

The Phillies had every chance to escape this fate. Rose’s four-year, $3 million-plus free agent deal (back when that was virtually tops in baseball) expired at the end of 1982, on the discordant note of a -1.1 bWAR campaign. Two of Rose’s four years in Philadelphia had come in at sub-replacement level. He would turn 42 in the opening weeks of the ’83 campaign. It was a great time to wish him well and look for their new first baseman.

They couldn’t do it. Rose was a clubhouse leader, earning extravagant praises from players all the way up to Mike Schmidt for what he taught them by example. With him, the Phillies had won their third pennant and first championship in their 98 years of existence, breaking a skein of futility only Cubs fans today can truly understand. Besides, Rose was just one season removed from leading the league in hits, and if his performance had been zig-zagging up and down, 1983 was in line to turn upward.

So the Phillies brought him back for the zig but got the zag. Rose, with 15 years batting better than .300 under his belt, plunged to .245 with a mere .602 OPS and a 69 OPS+. His defense at first base was roughly what one would expect from a 42-year-old. Bad offense at an offense-first defensive position, played poorly, added up to a gaping wound. In September, the guy who famously despised taking even a single game off was reduced to the insult of part-time play.

Remarkably, the Phillies didn’t pay for their pholly. In the weakest National League East since 1973 (see below), Philadelphia managed 90 wins and a comfortable six-game margin in the divisional race. A take-off on the 1950 Phillies’ “Whiz Kids” nickname tarred them as the “Wheeze Kids”—Rose’s fellow ex-Reds Joe Morgan and Tony Perez were 39 and 41; Steve Carlton was 38; only one starting position player was under 30*—but they were spry enough to dump Tommy Lasorda‘s Dodgers for the pennant. The Baltimore Orioles were a wheeze too far, and Philly dropped the World Series in five.

* They sound just like last season’s New York Yankees. Except for winning.

This time, even with another flag to fly, and even with Rose 10 hits from 4000, the Phillies let Charlie Hustle hustle up his next gig elsewhere. As you’ve likely heard, he did manage to catch on.

Worst Position Player for a World Series Winner

Deron Johnson*, 1974 Oakland A’s, Milwaukee Brewers, and Boston Red Sox, -2.2 bWAR

Deron Johnson just missed the -0.5 list above, with two of his six apparently qualifying seasons rejected due to rounding. (The tables read “-0.5,” but it’s really -0.48 or -0.51 or such, and the search engine rejects the former.) He’s exactly the right type for both that list and this one, with strong seasons like his 1965 with Cincinnati and 1971 with Philadelphia giving teams reason to hope he’d snap out of a drought and start playing well again.

His 1973 campaign with the Oakland A’s hadn’t been that great, a modest 1.5 bWAR, but when you’re winning multiple World Series, plenty of things look brighter than they are. (See Rose above.) Johnson’s 1974 didn’t count as one of those things. In 50 games with Oakland, he batted .198/.239/.345 (71 OPS+), which is rough if you’re a middle infielder. Johnson was a first baseman and DH. (Again, see Rose above.)

The A’s had many more options than this rolling calamity. During that season, eight men played first base, and 13 served as designated hitters for Oakland. For comparison, and to blow youngsters’ minds, they had 11 pitchers that year, nine of them handling all but 10.1 innings. Almost as mind-boggling, Ray Fosse would log the most games on the team at first base in 1974 while logging the most games at catcher, as well.

The Athletics had enough by late June and swung a waiver deal with the Milwaukee Brewers. Johnson did even worse with the Brew Crew and got shunted to Boston, where he really cratered in very limited time. Only -0.8 of his -2.2 WAR accrued with the A’s, so while he is technically the worst position player with the proverbial ring, there is a better candidate for the role.

Worst (full-year) Position Player for a World Series Winner

Todd Benzinger, 1990 Cincinnati Reds, -1.6 bWAR

Benzinger helps to fill out the template for single-season position player disasters. The not-so-secret formula is to field a position with little–or, in the DH’s case, no–defensive value, and field it poorly while mustering batting numbers that are awful against expectations for an offense-leaning position, but not quite so terrible that everybody can see you should be benched, demoted, traded, or waived, thus stopping you from burrowing deeper into negative numbers.

It also helps to have good past performance that justifies a team’s sticking with you. Benzinger had that in a solid 1.2 WAR half-year debut with the Red Sox in 1987. He stumbled the next year in Boston, and in 1989 with the Reds, but his home run numbers creeping upward (8/13/17) may have mesmerized incoming manager Lou Piniella into staying with him, hoping for a breakout into true power-hitter territory.

What really happened is by now dismayingly familiar, save perhaps in how little Benzinger’s performance fell off. His .253/.291/.340 line gave him an on-base average just two points below his 1989, though his OPS+ did nose-dive from 89 to 70. His first base defense, already uninspiring, also slid. The one element in the equation where he faltered was that he couldn’t stay a regular, as rookie Hal Morris swooped in to start taking the majority of first base at-bats by July.

The Reds survived and surmounted the damage he did. Almost the entire front line of the team was in the age-25-to-29 sweet spot, the peak years for the average major leaguer. This may have helped a team of not extraordinary talents pull everything together to take the division, beat the Pirates for the pennant (starting Pittsburgh’s string of three straight NLCS defeats), then blitz the defending champion A’s in a four-game sweep.

(The bygone sports paper The National listed odds of a Reds’ sweep at 40-1. I have not forgotten those odds in 24 years, they looked that good. Had young me been in Nevada, and not been broke, young me could have made a bundle. Which young me probably would have blown buying a Nolan Ryan rookie card or something, so shed no tears for young me.)

Benzinger started 1991 doing even worse before being bundled off to Kansas City. He had enough flashes of ability to stay in the majors until 1995, with nobody who would earn him a second ring. But he did get one. And a 3-for-9 with two walks in the NLCS means he did actually contribute in October. (We’ll gloss over the World Series 2-for-11.)

With that, we switch over to the pitchers, starting with a short detour.

Dishonorable Mention

Steve Blass, 1973 Pittsburgh Pirates, -4.0 bWAR

Blass had, by the measure of bWAR, the worst season a major-league pitcher ever fashioned. Over 88.2 innings, he coughed up a horrifying 98 runs (all but one earned) for an ERA of 9.85 and an ERA+ of 36. He chalked up 84 walks against 27 strikeouts, along with a WHIP close to 2.2. His 3-9 record highly exaggerated his worth to the Pirates.

He stuck on the pitching staff as long as he did in 1973 because of 1972. With a 19-8 record and a 2.49 ERA (135 ERA+), Blass earned his first and only All-Star berth, finished second in the NL Cy Young voting, and got a little down-ballot MVP support. Manager Bill Virdon, recalling that Blass had started slowly in earlier years, gave him half a season to turn himself around. Blass did not. Virdon gave him just three appearances after the end of June, and early in August said Blass would not be starting again for Pittsburgh that year.

The reason I recapitulate this horror story—to this day, some old-timers will still refer to a bout of wildness as “Steve Blass Disease”—is not because the Pirates made the playoffs, but because they nearly did. Nobody, it seemed, wanted to win the National League East in 1973. After the games of Sept. 1, the last-place Phillies were a mere six games out of first, with the teams tied at the top just one game over .500. And the division leaders that day, by slivers of percentage points over St. Louis, were the Pittsburgh Pirates.

The Bucs would slip off the pace in early September, and Virdon was fired in favor of Danny Murtaugh. Virdon’s promise regarding Blass lapsed with his departure, Murtaugh giving him three starts in the September stretch drive. It nearly worked out: Blass took a hard-luck 2-0 defeat against the Cubs on Sept. 11, then eked out a quality start on the 16th against the Cardinals, getting a no-decision though his Pirates would lose. His Sept. 21 start against the Mets, though, was a train wreck, over in the first.

Pittsburgh would finish in third at 80-82, a scant 2.5 games behind the New York Mets. Had someone other than Blass filled out the Pirates’ rotation, a replacement-level performance would have given them the division crown. Even if someone other than Blass had started that last game versus the Mets, it all could have been different. Win that one, and Pittsburgh would have been tied with St. Louis at 81-81, with the Mets pulled down to 81-80. New York would have to play out that final rained-out game against the Cubs, and if Chicago got its revenge for 1969, there would have been a three-way pile-up at the top of the NL East.

This is not a back-door entry to my Alternate Baseball series, but the near-miss is something I had to acknowledge. It also provides a handy bookend, for reasons that will become clear.

Worst Pitcher for a Division Winner

Chad Qualls*, 2010 Arizona D-Backs and Tampa Bay Rays, -2.8 bWAR

Qualls gets the asterisk you see above because he spent less than half his nightmare season with the playoff-bound Rays, 21 out of his total 59 innings pitched. This quibble is decisively overcome by Tampa Bay having traded for him. Right at the trading deadline, the Rays actually gave up a player (to be named later) to acquire a pitcher who had posted a -2.2 bWAR for the season so far.

Andrew Friedman and the Rays front office are supposed to be geniuses. This might be Exhibit A against that proposition (the Rays’ performance in 2014 contending for status as Exhibit B). Then again, they still did win the division. And Qualls would return to form as an effective reliever … three years and five teams later.

In 2010, though, he had trouble, and plenty of it. For 38 innings in Arizona, Qualls posted a dire 8.29 ERA (51 ERA+). His change of scenery to Tropicana Field helped a little, and no more. He struggled to a 5.27 ERA and a 71 ERA+, giving him a -0.7 bWAR*, though a 2-0 record there salved the pain.

* Yes, -2.2 plus -0.7 can equal -2.8 with the right kind of rounding.

Yet just under the surface, one can see what the Rays were thinking. Overall for the season, Qualls had a 49:21 K:BB ratio, better than the MLB average. The seven home runs he surrendered in 59 innings produced a rate above league average, but not terribly so (1.07 HR/9 versus 0.96). His FIP (Fielding Independent Pitching) for 2010 was 4.13 against an MLB mean of 4.08.

By those lights, Tampa Bay was buying low on two months of a roughly league-average pitcher who had been getting luck-whipped. Too bad for them that he kept getting whipped, at a reduced rate, while a Ray. If you believe in the numbers beneath the numbers, you’re inclined to forgive Tampa Bay for a savvy process that didn’t translate into on-field results.

It still looks pretty ghastly, though.

Worst Pitcher for a Pennant Winner

Brad Lidge, 2009 Philadelphia Phillies, -2.6 bWAR

Brad Lidge was a shutdown closer in 2008 for the Phillies, converting all 41 of his save opportunities and posting an ERA below 2.00 (with a FIP below 2.50). He copped his second All-Star appearance, serious NL Cy Young consideration, and a top-10 MVP finish, not to mention a World Series win. The Phillies were sure to stick with him as their closer for 2009.

This, of course, was the trap. The burnout rate for closers is even worse than the current Tommy John surgery rate for starting pitchers: a painful statement, but true. Lidge himself was already an example. He’d had two great years closing for Houston before two rough years saw him bundled off to Philadelphia. Philly thought they had gotten the real Brad Lidge. They had, but they’d gotten both halves of him, and the other half was about to show itself.

Lidge’s performance ballooned to a 7.21 ERA and a 59 ERA+ for 2009, though his FIP was less grotesque at 5.45. His strikeout rate fell below 10 per nine innings for the first time in his career, and his walk rate, always a little elevated, soared to 5.2 per nine. Worst of all, his 58.2 innings saw 11 balls depart the premises, a 1.7/9 IP homer rate. His record, 0-8, portrayed total futility.

Lidge did record 31 saves, but he blew 11 opportunities. FanGraphs’ Shutdown/Meltdown metric (measuring when a reliever adds or subtracts at least 0.06 Win Expectancy with his pitching) credits Lidge with a 19/15 ratio, a mark of 1.27 where 2.5 is considered average. In only half of his 42 save opportunities did Lidge record a clean inning (or less: he had some two-out opportunities, but never more than one inning). On average, more than 70 percent of innings produce zeroes. Lidge managed 50 percent during his save attempts.

Philadelphia still won the division, and Lidge almost redeemed himself in October. He saved three games and won one, allowing no runs over four innings, in the two series that got the Phils back to the World Series. His lone appearance against the Yankees, though, was in the ninth inning of Game Four. After Johnny Damon exploited an overshift against Mark Teixeira with a two-out first-to-third steal, the tie game disintegrated for the stunned Phillies. Lidge plunked Teixeira, then gave up hits to Alex Rodriguez and Jorge Posada for three runs and a Yankees win that put them in the driver’s seat for the Series.

Elbow surgery in the offseason presaged hard times ahead. Though he had a partial rebound in 2010, further injury woes sunk his career. He can take pride in being a member of the only back-to-back pennant winners for the Phillies, but it is likely a bittersweet pride.

Worst Pitcher for a World Series Winner

Bob Veale, 1971 Pittsburgh Pirates, -2.1 bWAR

It sure looks like the Pirates of the early 1970s had their pitching problems, doesn’t it?

Looks may be deceiving. Bob Veale’s 1971 season wasn’t merely a study in contrasts, it was a parade of contradictions. A difficult 1970, combined with back troubles and conditioning worries, got this former two-time All-Star starter moved to the bullpen, so one could well have expected struggles. What ensued was a disaster that, from some angles, didn’t look bad at all.

In 37 games and 46.1 innings, Veale produced an ugly 6.99 ERA (50 ERA+) and 1.791 WHIP. His record, bizarrely, was spotless at 6-0 with two saves, but we know how deceptive those stats can be for a pitcher. More probative is his FIP number: at 3.68, it wasn’t exactly good and was the worst of his whole 13-year career, but it wasn’t anybody’s idea of a horror story. His strikeout and walk rates both had been worse before, but he did give up five homers in his 46.1 innings, where he had thrice before led NL pitchers with the lowest home run rate.

To untangle the discordant numbers, I looked at another number, the Win Probability Added he produced in all his games. Take all his 1971 appearances together, and he posted a WPA of -0.002. That means he cost his Pirates two-thousandths of a win. By that measure, he was dead average … which, you will recall, is well above replacement level.

On the surface, Veale was chopped liver, but in the deeper numbers he was pulling a Chad Qualls. It’s therefore less of a wonder that his Pirates took the East without a huge strain and knocked off the Giants in the NLCS. Beating the three-time pennant-winning Baltimore Orioles in the World Series was a little more of a wonder, and Veale didn’t help much. In his lone appearance, he got just two outs while allowing one run and permitting two inherited runners to score. But this was in the middle of an 11-3 rout, so the harm to his team was minimal.

At each level of postseason progress, the pitchers have low-balled the position players. Adding the uncertainty of defensive metrics, I was ready to state categorically that teams can survive a black hole on the mound better than one on the field. Veale’s deceptive 1971, though, makes me rethink that, which is probably fortunate. My record in making confident statements has taken some hits lately.

In Conclusion

The reasons for keeping a failing player on one’s team range all over the map, but the core cause beneath each one is misinterpretation. A catcher’s defense seems to be enough to compensate for a woeful bat; a first baseman’s offense seems to be enough to justify gritting your teeth over his immobility; a good campaign a few years back seems reason enough to hope that last year’s toxic waste spill of a season was the anomaly.

If you’re right, you don’t make the lists above. If you’re wrong, well, the immediate pain that causes is usually much worse than anything some online writer can inflict on you. And occasionally, say if your under-performer is named Benzinger or Veale or Rose, being wrong ends up not hurting at all. This fact makes it that much more likely you’ll be wrong again. And again.

We have a bunch of new statistics and measurements, but don’t be fooled. Nobody has figured baseball out yet … which is good. Wouldn’t it be dull if everyone did crack the code?

References and Resources

Print Friendly
 Share on Facebook0Tweet about this on Twitter7Share on Google+0Share on Reddit0Email this to someone
« Previous: Searching for Racial Earnings Differentials in Major League Baseball
Next: The Screwball: Fan Diaries in the Age of Television »

Comments

  1. AC of DC said...

    Although the season in question did not meet the criteria, these articles did bring to mind an Honorable Mention from the Recent Past; and while he became perhaps too favored a whipping-boy and badly abused at the end of an otherwise decent career (some of which may have been brought on by his own comments), and served as a focus for much anti-Ron Washington sentiment, the Michael Young of 2012 was, in plain facts, notably bad.

    Coming off a good-hitting performance and an All-Star selection in 2011, in the midst of a hefty (by that point) $16M/year contract, and with a solid track record, Young could be expected to be given a good stretch in which to turn it around and return to form. In the end, he was given 651 PA and 735 Innings all over the infield of miserable, end-of-career futility, measuring up as the second least-valuable player in MLB.

    In all, he compiled -1.6/-1.7 (f/b)WAR. You may recall that the Texas Rangers lost the division by 1 game to the Oakland A’s in 2012, forcing Texas into the Wild Card Play-In, which they lost to Baltimore. With any other replacement scrub in those appearances, what could have been?

  2. Cliff Blau said...

    It was Gene Tenace who played the most games at first for the 1974 A’s. Fosse didn’t play a single game there.

  3. Shane Tourtellotte said...

    Cliff, you are correct of course. It was Tenace who played the most at catcher and first: 106 and 79 games respectively. (Obviously he was being switched around plenty in-game.) Everyone kindly ignore my Fosse brain-cramp.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>