The Worst of the Best: Holes on World Series Teams

The end of the season is rapidly approaching. We’ll soon be down to 10 teams playing baseball, and a few weeks after that it’ll be the final two in the World Series. They will represent the best that baseball has to offer—theoretically.

In practice, they probably won’t be across-the-board powerhouses. Each team will have its weak spots, and perhaps worse. They may have a position on the diamond so weak, you might wonder why it doesn’t drag the entire team down.

Plenty of teams have these holes, and for many it does drag them down, finishing out of the playoffs or at least getting bounced before they can snag the pennant. More interesting, though, are the teams that succeed despite these holes. So today, I’m looking at the worst patches on the teams that have played for all the marbles.

I have done something a little like this before. In a pair of articles last year, I looked both at players with the most sub-sub-replacement seasons, and players who did the worst with various levels of playoff achievers. This is different in that it includes all players who took a specific position that year. One player’s disaster can be redeemed by a fill-in who did well in limited duty or exacerbated by subs who did as badly or worse.

How Deep Is a Hole?

So what constitutes a hole? An easy definition would be a position that has production at replacement level or worse. We have WAR numbers for every big-league player, so superficially this looks simple.

The sticking point is that very few positions are handled by just one player for a whole season. This is especially true of weak spots, where managers may cycle through several players trying to find one who can perform adequately. Running the numbers for all the players at one spot would be a trial. Fortunately, someone’s done that work for us.

Baseball-Reference’s season pages now include calculations for Wins Above Average (WAA) for every team in the World Series era at every position, measured against the average for a position rather than for all players. Collective ratings are done for pitching and pinch-hitting. This isn’t exactly what I sought, but I can adapt it to this purpose.

With this information in hand, I collected performance data for every field position on almost all the teams that have played in the World Series*. This excludes 1904 and 1994, when no World Series was played. I also gathered WAR data for all relevant starting pitchers for separate figuring.

* I did exclude three World Series years from my research: 1943 to 1945. The manpower demands of World War II stripped the majors to such an extent that it becomes unfair to judge teams on the holes in their lineups. Enough regular players were around in 1942 to act as if it was a normal season, but the Yankees and Cardinals won their pennants without any holes, so I could have treated it as a war year without a problem.

I left out relievers as a position for two reasons. First, the nature of the bullpen allows for “hiding” an underperformer much more than other positions do, especially for getting a failing reliever out of the closer role. Second, unlike position players and starting pitchers, a game does not have to involve a reliever. (This also led me to exclude pinch-hitters, which B-R’s tables would have let me include.)

I will first take a look at the biggest gaps World Series teams had to work around before going on to the broader picture of how many teams actually suffered these gaps.

Holes in the Field

Starting off are the position players. Out of the 214 teams included, 15 suffered a position that produced -3.0 Wins Above Average or worse. That works out roughly to a collective -1 WAR, or worse (a translation I’ll explain later).

Of those, I will list and give sketches for the bottom 10. They cover every position but shortstop—though two shortstop holes fall into the 11-15 finishers, so there’s no immunity—and range from single players stubbornly stuck with to revolving doors that never spun the right way.

T9: 1941 Yankees, First Base, and 1987 Twins, Catcher. -3.3 WAA

Two years after losing Lou Gehrig, the Yankees gave up on replacement Babe Dahlgren (a hole for the ’39 champs) and tried Johnny Sturm instead. Sturm batted .239/.293/.300, which could have cost a shortstop his job. Fill-in duty by regular second baseman Joe Gordon kept the first-base hole from being bigger than it was. Sturm would go off to war and never return to the majors.

An 85-77 team, even if World Series-bound, is likelier to have holes than your average pennant winner. Minnesota didn’t beat the odds, at least not at catcher. Of their four semi-regulars there, only one had better than a 62 wRC+. That was Mark Salas, who began the season on a batting tear. So naturally he was traded to the Yankees for Joe Niekro, who pitched like the wrung-out 42-year-old he was. Add mediocre defense by all concerned, and the Twins had a lot to overcome—which they did.

T7: 1936 Giants, Third Base, and 1939 Reds, Left Field. -3.4 WAA

Travis Jackson has a probably-undeserved plaque at Cooperstown, proving the voters completely forgot this, his final season. A .230/.260/.297 batting line over 126 games says it all (since his defense was about scratch). His relief, Eddie Mayo and Joe Martin, served only to dig the hole deeper.

Cincy’s front-line left fielder, Wally Berger, was not the problem, as he hit decently. Old injuries limited his playing time, however, and his replacements were horrible. Lee Gamble batted for a 64 wRC+, Frenchy Bordagaray managed just 21, and nine games by legit Hall-of-Famer Al Simmons produced an awful .143/.217/.143 line. Sometimes, half a hole can be plenty deep by itself.

T5: 1970 Reds, Second Base, and 1973 Mets, Center Field. -3.6 WAA

Cincinnati went with Tommy Helms at second, and despite all evidence stayed with him; no other Red got more than 29 plate appearances at the keystone. Helms’s defense was actually a tick above average (winning him a Gold Glove he likely didn’t merit), which at second base produces plenty of value. So imagine how much his .237/.262/.282 batting dragged down his worth. Helms actually held onto his job, produced his career-best WAR in ’71—and then got traded away for Joe Morgan. Great swap, if weird timing.

Center field wasn’t the 82-79 Mets’ only hole, but it was the most painful. Eight players cycled through center, Don Hahn and his .229/.285/.290 bat leading in PA there. Most of the others hit worse and didn’t exactly dazzle with the glove. The best option on hand was Willie Mays, with his 41 games started in center. Unfortunately, he was playing like yet another wrung-out 42-year-old. If you thought watching Derek Jeter hobble through his final season was rough, ask some older fans about Willie in ’73.

4: 1983 Orioles, Third Base. -3.8 WAA

The O’s started off with Leo Hernandez at third, who posted an 82 wRC+ and awful fielding numbers. Midway into the season, they bought Todd Cruz from the Mariners, converting him from shortstop. He fielded much better than Fernandez but hit much worse, with a .208/.259/.299 line for his Baltimore stint. Aurelio Rodriguez and Glenn Gulliver picked up the rest of the starts. Rodriguez’s slash numbers all started with ones. I will say no more.

3: 1967 Red Sox, Catcher. -4.1 WAA

Boston’s Impossible Dream wasn’t winning the pennant, it was finding a backstop. Four different players got at least 18 starts behind the dish, led by Mike Ryan’s 74. The 52 wRC+ he put up was actually better than his career average but still awful. Russ Gibson’s wRC+ was a little worse; Elston Howard’s was much worse. At least they could field, which Bob Tillman couldn’t. Boston sold him in August to the Yankees, presumably thinking it was just what they deserved. (But New York had traded them Howard five days before, so they were even.)

T1: 1961 Reds, Catcher, and 2012 Tigers, Right Field. -4.2 WAA

Three times on this list? What was it about the Reds, anyway? They traded starting catcher Ed Bailey away in the opening weeks, thinking rookie Jerry Zimmerman plus the catcher they received in the deal, Bob Schmidt, could hold things together in ’61. Not with OPSes of .483 and .389, respectively, they couldn’t. They rushed up another rookie, Johnny Edwards, to stem the tide. Edwards would be a capable catcher in his time, but 1961 wasn’t his time. They found a good stopgap when they bought Darrell Johnson from Philly in mid-August, but his 17 starts weren’t enough to fill up what the others had excavated.

Standing with the Reds at the nadir is easily the most modern team on the list. The lion’s (tiger’s?) share of blame for this hole goes to Brennan Boesch. After a promisingly average 2011, his batting and fielding both collapsed. (72 wRC+, -18 UZR/150.) With roster options like Jeff Baker and Don Kelly only spreading the blaze, Detroit called up prospect Avisail Garcia for the stretch drive and postseason. Avisail hit a little better, fielded right field rather worse, and didn’t exactly compensate in October, either. Neither Boesch nor Garcia has provided a positive fWAR season since.

Taken together, these bottom-10 teams went 3-7 in the Fall Classic, which would be 1-7 for the bottom eight. This points toward a deleterious effect from having a big hole on one’s team, but we’ll want a bigger sample than that before committing ourselves.

Holes in the Rotation

Determining holes in the ranks of starting pitchers was a trickier matter. B-R’s handy boxes measure all pitching in most cases, and in many years break that down to starters and relievers. But looking at the rotation as a whole (or hole) does not really equate to a weakness at one of the other eight positions. I needed to look at individual starters.

This raised a fresh problem. I could get bWAR numbers for those pitchers but could not directly measure them against a league average as I did for position players. I instead could have gone with whoever was at replacement level or worse, but this butted up against the WAR calculations of FanGraphs. FG rates the poor starters I combed out at B-R significantly higher in almost all cases.

I split the difference, literally, by averaging starters’ bWAR and fWAR scores. I then counted only those pitchers who actually started one or more games in the World Series (with an exception I’ll explain), and were at least within range of replacement level with an averaged WAR of 0.5 or worse. This represented teams that could not, or would not, “hide” the holes in the ranks of their starters.

With those criteria, I came up with nine starting pitchers who represented holes on their World Series teams. I’ll give synopses for them, with their WAR averages in parentheses.

Phil Douglas (0.4), 1921 Giants: Douglas started three games in a best-of-nine Series, going 2-1 as his Giants toppled the upstart Yankees.

Bump Hadley (0.3), 1937 Yankees: Hadley started Game Four as the Yanks went for a sweep of the Giants. He departed in the second, five runs hung around his neck. The Bombers had to content themselves with winning the Series in five.

Jack Billingham (0.2) and Fred Norman (0.45), 1975 Reds: This two-fer is emblematic of the spotty starting pitching on the Big Red Machine. Billingham started Game Two and, while he pitched pretty well, had to be bailed out by a ninth-inning Cincy rally that won the game. Norman started Game Four, gave up four runs in Boston’s fourth, and took the L. Both men would fill out the Series with scoreless relief duty.

Jack Billingham (0.2), 1976 Reds: Billingham is the only repeat offender, but it’s a special case. He had no starts in Cincinnati’s sweep of the Yankees but got a high-leverage relief assignment in Game Two (coming in, ironically, for Fred Norman). He pitched 2.2 hitless innings, though he let the tying run cross on a groundout, and got the win on Tony Perez’s walkoff single.

Al Nipper (0.0), 1986 Red Sox: Nipper drew the Game Four assignment. While he managed the quality start minimum of three runs in six innings, it wasn’t enough as the Mets cruised to a 6-2 win.

Mike Moore (-0.1), 1990 A’s: Moore was routed out of his Game Three start, giving up six runs, though just two were earned. Oakland was swept by the Reds.

Brian Anderson (-0.35), 2001 D-Backs: Anderson started Game Three, yielding two runs in five and a third innings. He took the somewhat hard-luck loss, though Arizona would win the war.

Barry Zito (0.5), 2012 Giants: Of all the pitching holes, only Zito and Douglas got Game One starts. Douglas lost his; Zito, with one run allowed in five and two-thirds, carded the win, getting the G-Men off on the right foot in their sweep of Detroit.

If these nine examples are any indication, having a hole in your World Series starting rotation is not a problem. Six of the eight starters, plus the special case, played on the winning side, despite the pitchers combining for a 4-6 record. Perhaps managers can tolerate sending out weaker arms when they trust the offense to put up runs (the 1937 Yankees and ’75-’76 Reds being outstanding examples).

Quantity and Quality

How often does a hole crop up on a World Series team? Or more than one? And how destructive is a hole to a team’s chances of winning it all?

It’s at this point I finally need to lay out the criteria for a “hole.” The limit I chose for field positions is -2.0 WAA. Average performance for a full-time position player is considered to be about 2.0 WAR; subtracting two wins from that brings one to replacement level, which feels right. It’s not quite as neat as that, since total PAs for a position will be somewhat over what is considered a full player season, but it gets us about to the right place. For pitchers, I’m going with the nine examples I produced above.

Using those standards, of the 214 teams that have played in the World Series (outside my wartime exclusions), 72 have had at least one hole, or just about one-third of them all. Multiple holes afflicted 15 of those, and there were three teams that endured three holes apiece.

First were the 1907 Tigers, lacking at catcher (manned primarily by Boss Schmidt), second base (Germany Schaefer), and third base (Bill Coughlin). They lost the Series to the Cubs, 4-0 with one tie. Next were the 1986 Red Sox, with holes at first base (Bill Buckner) and shortstop (various) along with Nipper in the rotation. They almost overcame that. Lastly were the 2012 Tigers, the right field debacle joined by holes at second base (several) and DH (Delmon Young).

Going by their examples—3-12-1 in World Series games—one might rapidly conclude holes on a team are fatal in the Series. Well, having that many holes certainly has been, but the more common one or even two may not be. Besides, your holey team might be facing an opponent as holey, or even more.

To properly figure out records for holey teams in World Series play requires not just counting up holes but measuring their severity. With position players, I judge anything from -2.0 to -2.9 WAA to be a Category-1 hole. From -3.0 to -3.9 is Category-2, and -4.0 on is Category-3. Teams accumulate one “hole point” for the category assignment of each hole it has.

As an example, a team that’s -3.0 WAA at second base and -4.2 WAA at catcher has five hole points. That was the 1961 Reds’ situation. The only other team to carry five hole points into the World Series was Detroit in 2012, with three in right and one at both second and DH. They combined to go 1-8 in their World Series, which is probably not coincidental without being predictive, either.

For the pitchers, 0.5 to 0.1 WAR is Category-1, while zero and below is Category-2. I am less sanguine about these rankings, but since it affects the balance of holes in only three World Series, and two of those cancel each other out, I will stick with them, with a short explanation of the remaining case.

In the 56 World Series played where at least one team had at least one hole, five were evenly matched in categories. In the other 51, the team with more hole points posted a 24-27 record. The one odd case was 1975, where the Reds with two Cat-1 pitching holes faced the Red Sox with one Cat-1 positional hole (at third base). If you dispute the holeyness of Billingham and Norman, you might call it even or give Boston the holeyness edge.

Even doing that, though, being the holeyer team is not that destructive to one’s chances of winning the Series. Holes, naturally, are not the only measure of a team’s quality. Would you take two average players, or a lousy shortstop plus Babe Ruth? The 1923 Yankees wisely were content with the latter.

Teams with holes and nothing to counter-balance them don’t make the World Series, though with the current Wild Card system that’s somewhat less true than it used to be. In the era with just the World Series, there were 35 holes over 124 WS teams, or 0.28 holes per team. In the LCS era, it was 19 holes for 50 teams, 0.38 holes/team, and the Wild Card era has seen 18 holes for 40 clubs, for a 0.45 holes/team average. This does reflect a drop in average winning percentage for teams in the Series, though this is partly from rising parity along with less superior teams getting their chances in the October crapshoot.

So glaring weaknesses have become more common for teams making the World Series, but it does not make them easy prey for their opponents. Unless, perhaps, they have as many as the 1961 Reds and the 2012 Tigers.

Don’t Raise the Roof, Raise the Floor

If having holes is only mildly inconveniencing for a World Series team, could the opposite state be more decisive for one’s chances? I do not mean lacking holes. I mean lacking even mild weaknesses, positions where the team is “Eh, okay, I guess.” And can a team even be constructed that has no such weaknesses?

Not often, but it does happen. I looked for World Series participants that were above average at every position and did find some.

I counted all field positions, plus both starting and relief pitching during the times when B-R broke down the numbers between the two. (For early periods, with little established relieving, I counted pitching as a whole.) This division cost 10 teams a spot on the list, eight for relief weakness, two for starting pitching. One of the latter examples was the 1953 Brooklyn Dodgers, often thought of as an all-time great club. The other I’ll identify soon.

With all this taken into account, we come up with seven-and-a-half teams that meet the criteria. I’ll explain the fraction once I’ve listed the first six teams with an uncomplicated claim to being better than average everywhere. With World Series results in parentheses, they are:

• 1905 New York Giants (won)
• 1910 Chicago Cubs (lost)
• 1922 New York Giants (won)
• 1948 Boston Braves (lost)
• 1969 Baltimore Orioles (lost)
• 1978 Los Angeles Dodgers (lost)

The seventh club is the 1984 Detroit Tigers. All their position players finished above average, and their total pitching beat the mean by 1.7 WAR. The numbers aren’t broken down into starters and relievers for this year (weirdly, the site stopped doing that for AL teams when the designated hitter was implemented), so one category or the other could be low enough to disqualify the Tigers.

A dip into the season splits puts Detroit starting pitching at an sOPS+ of 97, seven percentage points better than league average of 104. (The AL/NL split causes this odd number.) This may not translate perfectly to the WAR scale, since Detroit starters pitched 67 percent of team innings as opposed to 71 percent for the full AL. I’m surmising, though, that the seven-point gap isn’t closed by losing around one out per game. Given that, and the great bullpen numbers, Detroit makes the list and lifts the World Series record of all-solid clubs to a still-uninspiring 3-4.

The half-club on the list is one of the teams you might expect: the 1998 New York Yankees. This team achieved one of the all-time great seasons with at most one superstar-level performance but strength everywhere—except designated hitter. The tables put the Bombers right on 0.0 at DH. Sorry, Straw, you didn’t stir the drink quite enough. (Okay, Straw and Rock and Chili.)

One other team deserves notice, if only to lead into what another team accomplished. The 1975 Reds probably have the most famous lineup in baseball history, and by the numbers they deserve it. Of their eight regular positions, the worst performance was 1.2 wins above average by the right fielders (meaning primarily Ken Griffey, Sr.). It’s an incredible standard, and it’s only the weakness of their starting rotation (with examples you saw above) that keeps the team off the all-solid list.

But amazingly, another club had a higher floor everywhere. That distinction belongs to the 1969 Baltimore Orioles. Their weakest position was second base, Davey Johnson’s bailiwick, at 1.3 wins above average. They combined this with good pitching at both ends to make the list easily.

Truth to tell, their fourth and fifth starters that year, Tom Phoebus and Jim Hardin, did scuffle. This was more than balanced out by the front three of Jim Palmer, Mike Cuellar, and Dave McNally. Since this was still the time when a three-man rotation would take you through the postseason, this trimmed away Baltimore’s lone soft spot and made them perhaps the most solid team ever to play in the World Series.

But as the rest of the list shows, rock-solid does not mean a lock to win the Series. Maybe it didn’t take a miracle to beat the Orioles after all.

Looking Ahead

With this year’s playoffs rushing toward us, it’s not too early to think about which potential World Series participants may have holes of their own. Through the end of August, these teams are on pace for at least a chance to have one or more holes. WAA values and the prime starter at that position (if he’s over 50 percent of the PAs) are listed for position players; mean bWAR/fWAR is given for starting pitchers.

• New York AL: 2B, -2.0 WAA (Stephen Drew)
• Kansas City: 2B, -2.2 WAA (Omar Infante); RF, -1.8 WAA (Alex Rios)
• Minnesota: SS, -2.3 WAA (several); RF, -2.3 WAA (Torii Hunter); Ervin Santana, SP, 0.0 WAR
• Houston: 1B, -2.2 WAA (Chris Carter); DH, -1.8 WAA (Evan Gattis)
• Los Angeles AL: 2B, -2.0 WAA (Johnny Giavotella); LF, -2.3 WAA (many)
• Washington: 1B, -1.8 WAA (Ryan Zimmerman); LF, -3.1 WAA (several); Doug Fister, SP, -0.15 WAR
• Pittsburgh: 1B, -2.6 WAA (Pedro Alvarez)
• Chicago NL: SS, -1.8 WAA (Starlin Castro); RF, -1.9 WAA (Jorge Soler)
• Los Angeles NL: SS, -1.8 WAA (Jimmy Rollins)

All but four of the potential playoff teams have at least one hole. Those without are Toronto, Texas, the Mets, and St. Louis, which currently include three of the six division leaders. Every position but third is represented. The biggest chasm on a team that would play on if the season ended today is first base for the Pirates.

The status of the pitchers depends on whether they actually get a postseason start. Fister in Washington and Santana in Minnesota probably won’t be in the playoff rotation if their teams make it.

So we have a good chance of seeing a World Series team this year with a positional hole. I’m tempted to make a forecast, but I’ll hold back. Too many teams have managed to overcome their holes for me to feel confident in consigning anyone to defeat on that basis. We will just have to watch the games.

That’s the whole point, isn’t it?

References and Resources

• Baseball-Reference
• FanGraphs
• SABR Baseball Biography Project
• The New Bill James Historical Baseball Abstract
• Roster Resource for current rotations