April 2, 2003
The Great Leap Forward
Which Teams Will Gain 20 Wins this Season?In a recent article about the 1967 Boston Red Sox, I wrote that the team's 20-win improvement was not particularly unusual. I had spent a few minutes convincing myself that there were a few other teams in neighboring seasons that accomplished the feat, but made no attempt to determine how common it was, or whether the 1960s were particularly unique in this regard. This article delves into the topic quite a bit further, presenting an historical survey of the phenomenon, while contemplating patterns that might help us figure out who is most likely to leap forward this year. An upcoming article will look at the reverse-teams that have declined by 20 wins in a season.
First, we need to explain the methodology. To account for different season lengths, I adjusted all team seasons to 162 games. For illustration, lets look at the largest improvement by a team in baseball history:
BASE PRIOR Team Lg Year W-L EQW W-L EQW Gain Louisville AA 1890 88-44 108 27-111 32 +76
The 1890 Louisville Colonels finished 88-44, good for first place in the American Association, after having finished 27-111 in 1889. Their 61-win improvement is impressive enough, but when adjusted for a modern season, their Equivalent Wins (EQW, shown above) become 108 and 32, a full 76-game improvement. As is often the case, especially in the 19th century, there is a Mo Vaughn-sized asterisk associated with this accomplishment. In 1890 most of the best players fled the two established leagues (the NL and AA) in favor of the rebel Players' Association. Louisville happened to keep a few of their good players and won a fluky pennant. When order was restored in 1891, the Colonels reverted to 55-84.
This study is looking for all teams that improved by 20 EQW in a single season. It might have been easier to use a team's winning percentage, and just look for clubs that improved by.123, which is 20 games in a 162 game season; such a method would yield the exact same list of teams. But this is less satisfying, as most of us do not naturally think in these terms. Devil Ray fans are not pining for a .432 winning percentage--they are longing for 70 wins. A 20-win improvement is something we can all relate to.
Cutting to the chase, there have been 161 such teams in major league history, a little more than one per season. Since you are likely reading this at your job, and I would not want to cause you any additional problems with your boss, we won't be exploring all of these teams' stories. We will instead survey the phenomenon chronologically, briefly pointing out some of the more interesting stories along the way.
Baseball in the 19th century saw new major league teams being created and dissolved nearly every year, leagues coming and going, and players jumping from team to team. As romantic as the baseball may have been, the era is a total mess when it comes to a study such as this. In 1877, which was the second year of the National League and therefore the first year that a team could improve their record, there were only six league teams, three of which improved by at least 20 EQW and three others that declined by 20 or more. Things got better as the century moved along, but with a vast difference between the best and worst teams and frequent movement of the star players, massive leaps up and down the league standings were not uncommon.
The standard deviation of winning percentage in the 19th century was .138, which is about 22 modern wins. Standard deviation is a measure of how dispersed a set of numbers are. The statistic might not mean much by itself, but when applied across history, a steadily shrinking standard deviation indicates that team records have become less dispersed, more tightly packed. It stands to reason that as there is less variance between teams, it becomes more difficult to advance (or decline) by 20 EQW.
Without a lot of explanation, here are the biggest advancers of the 19th century:
Louisville AA 1890 88-44 108 27-111 32 +76 Cincinnati NL 1878 37-23 100 15-42 43 +57 Detroit NL 1886 87-36 115 41-67 62 +53 Cleveland AA 1884 69-39 104 32-65 53 +50 Brooklyn NL 1899 101-47 111 54-91 60 +50
(Note: because the EQW numbers are rounded to the nearest integer, it may occasionally appear as if the last column is incorrectly calculated. It is not.)
The most interesting teams here are the 1886 Detroit Wolverines, who purchased the dissolving Buffalo franchise and took all of its best players (Dan Brouthers, Hardy Richardson, Jack Rowe and Deacon White), and the 1899 Brooklyn Superbas, whose ownership combined with the Baltimore Orioles and moved manager Ned Hanlon and several of the best Oriole players to the Brooklyn club. Both teams waltzed to pennants. Again, these stories are all interesting in their own way, but their lessons are irrelevant to the modern game.
The American League was the latest challenger to the dominant Nationals, becoming a major league in 1901. Unlike the fragile and poorly managed usurpers of the previous century, the AL quickly gained a stronghold, leading to franchise stability in both leagues for 50 years. The standard deviation of winning percentage continued to decline, from .120 in the 1890s, to .110 in the 1900s, and to .096 (about 15.5 wins in a modern season) in the 1910s. The SD remained fairly steady until it began dropping again in the 1960s.
There were 30 teams that advanced by 20 EQW in these two decades, many of them at least indirectly the result of the battles with the Federal League in 1914-15, and the resultant sell-off of many of the best players in the game. The breakup of the Philadelphia Athletics after the 1914 season, for example, distributed Eddie Collins and other stars throughout the league, and the team's subsequent collapse tossed 56 more wins into the pile for other clubs to claim.
The largest gainers in this period were:
New York NL 1903 84-55 98 48-88 57 +41 St. Louis AL 1902 78-58 93 48-89 57 +36 Philadelphia NL 1905 83-69 88 52-100 55 +33 St. Louis NL 1914 81-72 86 51-99 55 +31 Pittsburgh NL 1918 65-60 84 51-103 54 +31 Washington AL 1912 91-61 97 64-90 67 +30
John McGraw took over a lousy Giants ballclub in the middle of the 1902 season, and turned the team around the next year with the single greatest advancement in EQW of the 20th century. This type of improvement, more or less sustained for 30 years, is what fans of all terrible ballclubs aspire to in their dreams.
The other teams on this list enjoyed only temporary advancements. The 1902 St. Louis Browns were on their maiden voyage (they had played in Milwaukee in 1901) and would not field a team as good for twenty years. Clark Griffith, a man who would come to be the very symbol of the Washington Senators, first managed the team in 1912 and promptly hoisted them to a season that was miles better than any they had had before. The Senators played well in 1913, then was not really competitive again until their surprise 1924 pennant.
The next two decades saw the fewest 20 EQW advancers (14) of any similar period in this study. In fact, there were no such teams for six years (1920-25), the longest drought ever. When a team finally sprang forth, the New York Yankees did it in two straight seasons:
New York AL 1926 91-63 96 69-85 73 +23 New York AL 1927 110-44 116 91-63 96 +20
This remains the only repeater since 1900. The 1926 pennant winners simply recovered from a one-year downturn, and the 1927 team found another gear, becoming one of history's great teams. This is also the only league champion to ever pull it off, though they were not the best team to do it. The 1906 Cubs leapt from 97 to 124 EQW.
The best improvements of this period:
Boston NL 1936 71-83 75 38-115 40 +34 Cincinnati NL 1938 82-68 89 56-98 59 +30 Philadelphia NL 1929 71-82 75 43-109 46 +29 Detroit AL 1934 101-53 106 75-79 79 +27 St. Louis AL 1928 82-72 86 59-94 62 +24
The 1936 Braves, one of the centuries' great leapers, illustrate the benefits of starting so low. Their great achievement was akin to the 1962 Mets transforming into the 2002 Mets, which might give you some appreciation for the quality of Casey's amazin' juggernaught.
The 1938 Reds are a better story, rising above .500 for the first time in a decade and following that up with two straight pennants and a World Series title in 1940. The 1934 Tigers acquired Mickey Cochrane to catch and manage, and enjoyed the breakout of Hank Greenberg, propelling the team to two pennants of its own.
The 1940s witnessed twelve 20 EQW advancers, the most for any decade between the 1910s and 1980s. Six of these took place during World War II, when a shortage of players created an annual scramble to fill rosters.
The largest improvements were these:
Boston AL 1946 104-50 109 71-83 75 +35 Philadelphia AL 1947 78-76 82 49-105 52 +31 St. Louis AL 1940 67-87 70 43-111 45 +25 Brooklyn NL 1945 87-67 92 63-91 66 +25
With all of their real players (Ted Williams, Bobby Doerr, Johnny Pesky, Dom DiMaggio, etc.) returning from the military, the 1946 Red Sox bore only a passing resemblance to the 1945 squad. Although they technically pulled off one of the greatest improvements of the twentieth century, the uniqueness of the situation was obvious. The 1947 A's are much more interesting: they had finished dead last for nine of the past thirteen seasons when they unexpectedly gelled for three years. The Browns were a similar story, rising from ineptitude to temporary respectability.
In keeping pace with the times, the 1950s was among the more stable periods in baseball history. There were but five teams who improved by 20 EQW. For all succeeding decades, we will list all of the teams that pulled it off, and will do so chronologically.
Chicago AL 1951 81-73 85 60-94 63 +22 Milwaukee NL 1953 92-62 97 64-89 68 +29 Cleveland AL 1954 111-43 117 92-62 97 +20 New York NL 1954 97-57 102 70-84 74 +28 Pittsburgh NL 1958 84-70 88 62-92 65 +23
Three of these teams are good stories because they were able to maintain their new level for several years. The 1951 White Sox, led by rookie skipper Paul Richards, and bolstered by a great rookie season from the wonderful Minnie Minoso, began a string of 17 consecutive winning seasons, the second longest such streak in league history.
The Braves improvement coincided with their move to Milwaukee, the gelling of a deep pitching staff, and the breakout season of Eddie Mathews. All of this improvement was sustained, as the Braves added Henry Aaron the next year, contended for pennants for the rest of the decade, and remained over .500 for every one of their 13 seasons in Wisconsin.
The 1958 Pirates' improvement was largely due to its league-leading pitching staff. After a year of consolidation, a young core of hitters burst forth to lead the league in runs in 1960, and the team won the World Series.
There were three renowned "miracle" pennant winners in 1960s (the 1961 Reds, the 1967 Red Sox, and especially the 1969 Mets), about whom dozens of book have been written. There were a few other big advancements in the decade, which saw each league expand from 8 to 12 teams.
Cincinnati NL 1961 93-61 98 67-87 70 +27 Detroit AL 1961 101-61 101 71-83 75 +26 Minnesota AL 1962 91-71 91 70-90 71 +20 Philadelphia NL 1962 81-80 82 47-107 49 +32 Chicago NL 1963 82-80 82 59-103 59 +23 Minnesota AL 1965 102-60 102 79-83 79 +23 Boston AL 1967 92-70 92 72-90 72 +20 Chicago NL 1967 87-74 88 59-103 59 +29 Oakland AL 1968 82-80 82 62-99 62 +20 New York NL 1969 100-62 100 73-89 73 +27 Washington AL 1969 86-76 86 65-96 65 +21
Expansion can have a dramatic effect on the records of the other teams in the league. In 1962, the Mets and Astros joined the National League and finished a combined 104-216-creating an opportunity for large-scale improvements for the other eight teams. The Phillies, a hapless outfit in 1961, took the most advantage, parlaying a 31-5 record against the two new kids and a few breakout years (especially from Johnny Callison, Tony Gonzalez and Don Demeter) into a decade-best 32-game EQW improvement.
After 13 desultory years in Kansas City, the Athletics moved to Oakland in 1968 and (with the prominent additions of Reggie Jackson and Sal Bando) promptly jumped over .500. When the American League broke into divisions in 1969, the top five 1968 finishers plus the Senators joined the East, while the West got finishers six through nine and both expansion teams. This fortuitous split made the A's instant contenders, but within a few years the Athletics were a legitimately great team.
A lot of the Mets' improvement was likely an illusion, as their run differentials suggests that they were very unlucky in 1968 (72 wins instead of an expected 77) and extremely lucky in 1969 (100 wins rather than 92). The team's stellar pitching, and great years from Cleon Jones and Tommie Agee, were enough to carry an otherwise mediocre offense to their "miracle" season.
The 1969 Senators were a huge story at the time, as they were piloted by rookie skipper Ted Williams, who copped the Manager of the Year award for his efforts. Williams ought to have quit right then, since his subsequent three years with the club were beset with player disgruntlement, losing, and a steady erosion of his reputation as a manager. He never tried it again.
Things settled down a bit in the 1970s. Although the new divisions complicated matters somewhat, there were no real surprise pennant winners as there had been in the 1960s.
Montreal NL 1970 73-89 73 52-110 52 +21 Chicago AL 1971 79-83 79 56-106 56 +23 Kansas City AL 1971 85-76 86 65-97 65 +21 Cincinnati NL 1972 95-59 100 79-83 79 +21 Texas AL 1974 84-76 85 57-105 57 +28 Chicago AL 1977 90-72 90 64-97 64 +26 Montreal NL 1977 75-87 75 55-107 55 +20 Milwaukee AL 1978 93-69 93 67-95 67 +26 Montreal NL 1979 95-65 96 76-86 76 +20
The 1972 Reds were recovering from an abrupt tumble from their great 101-win breakthrough in 1970. The team reacted to a large decline in their offense by making a ten-team deal with the Astros that cost them slugger Lee May, but in turn netted them second sacker Joe Morgan. The team actually wanted Morgan because they felt they needed to diversify their offense, specifically by … stealing more bases. Fortunately, the team got the rest of Morgan's devastating offensive game in the bargain, and was soon one of history's great teams.
The biggest leapers of the decade, the 1974 Rangers, were skippered by Billy Martin, who made a habit of this sort of thing. His two previous managerial debuts, with the 1969 Twins (+12) and the 1972 Tigers (+15) had led to division titles. His delightful personality could not keep him employed very long, and the Rangers could not sustain its success with or without Martin.
The other managerial miracle worker of the era was Dick Williams. He had led the 1967 Red Sox (+20) to their dream season, and had also piloted the 1971 Oakland A's (+12) to a division title (and subsequently two World Series crowns). What he accomplished in Montreal was equally impressive. He took over the Expos in 1977 and led them to a 20 EQW improvement. After a year of consolidation (+1), the club improved by 20 more games in 1979, leading to the team's best ever era. The team's success was due mainly to Williams continual commitment to an impressive pipeline of young players, including, amongst many who enjoyed long careers, Gary Carter, Andre Dawson, and Tim Raines.
Defying many prognosticators, who foresaw free agency as a boon to only the wealthy teams, the 1980s were the least stratified decade in history. The following table lists the standard deviation of won-loss percentages over the last several decades, and the number of equivalent wins in a 162-game season.
1950s .090 14.6 1960s .080 13.0 1970s .073 11.8 1980s .066 10.7 1990s .067 10.9 2000s .078 12.6
On the other hand, the 1980s did see a rise in the number of big improvements, with 13 teams jumping by 20 EQW.
Oakland AL 1980 83-79 83 54-108 54 +29 St. Louis NL 1981 59-43 94 74-88 74 +20 Toronto AL 1982 78-84 78 37-69 57 +21 San Diego NL 1982 81-81 81 41-69 60 +21 Chicago NL 1984 96-65 97 71-91 71 +26 New York NL 1984 90-72 90 68-94 68 +22 Cincinnati NL 1985 89-72 90 70-92 70 +20 Cleveland AL 1986 84-78 84 60-102 60 +24 Texas AL 1986 87-75 87 62-99 62 +25 San Francisco NL 1986 83-79 83 62-100 62 +21 Oakland AL 1988 104-58 104 81-81 81 +23 Los Angeles NL 1988 94-67 95 73-89 73 +22 Baltimore AL 1989 87-75 87 54-107 54 +33
The three largest advances were all big stories. After a couple of fun-filled stints managing the Yankees, Billy Martin took over the 1980 Athletics, fresh off of a car wreck of a season. The team popularized something called "Billy Ball," but Martin's genius was mainly just getting rid of some dead wood and finding a few good players to ride. He found three young outfielders, Tony Armas, Dwayne Murphy and Rickey Henderson, played them every day and allowed them to become, briefly, one of histories' best groups of flychasers. He cobbled together five promising pitchers (Mike Norris, Steve McCatty, Rick Langford, Matt Keough, and Brian Kingman) gave them 159 starts, allowed them to complete 84 of them, and, not wanting to pamper them too much, brought them into relieve in 8 games as well. (Speaking of which, what exactly did the relievers do in the bullpen? Only three relievers on the club threw as many as 20 innings all year. It must have gotten awful boring out there.) The rest of the roster was nothing special, but Martin horsewhipped them to 83 wins in 1980, and a division crown the next year.
Many people consider Billy Martin to have been a great manager, perhaps even a Hall of Fame manager, based on his uncanny ability to turn bad teams around. Unfortunately, Martin was also a self-destructive drunk whose ballclubs generally imploded after a year or two. When Billy left Oakland after a desultory 1982 campaign (68-94) his team was a broken down mess. This was all part of the entertaining Martin package.
The 1984 Cubs, the best team seen on the north side in several decades, were also the first in what has been an unusual series of Cubbie one-year-wonders. The 1984 club formed the midpoint of the following three-year run of EQW: 71, 97, 77. The 1989 division winners (77, 93, 77), the 1998 wild card club (68, 89, 67) and the 2001 contender (65, 88, 67) were all unable to maintain any of their supposed progress. It is logical to expect some backsliding after a large improvement, but the Cubbies consistently give back the whole enchilada. They have not finished over .500 two straight years since the reign of Spiro Agnew.
The third big breakout team of the decade was the 1989 Orioles. The 1988 squad was a national story for losing their first 21 contests, eventually rolling to a dismal 54-win campaign. Of all of the great leapers in modern times, this club offers the best short-term hope for fans of the Devil Rays, Brewers, Royals, and their ilk. Unfortunately, the team was a total fluke, relying on one-year breakouts from non-prospects like Jeff Ballard and Craig Worthington. The main lesson of this team was how much ground you can gain by replacing terrible sink-holes with average players in the lineup and on the mound.
The 1990s saw 17 teams improve by 20 EQW, the most in any decade since the 1910s, understanding that there are many more teams now. There were a few dramatic stories.
Chicago AL 1990 94-68 94 69-92 69 +25 Detroit AL 1990 79-83 79 59-103 59 +20 Pittsburgh NL 1990 95-67 95 74-88 74 +21 Minnesota AL 1991 95-67 95 74-88 74 +21 Atlanta NL 1991 94-68 94 65-97 65 +29 Baltimore AL 1992 89-73 89 67-95 67 +22 Philadelphia NL 1993 97-65 97 70-92 70 +27 San Francisco NL 1993 103-59 103 72-90 72 +31 Cincinnati NL 1994 66-48 94 73-89 73 +21 New York NL 1994 55-58 79 59-103 59 +20 Boston AL 1995 86-58 97 54-61 76 +21 California AL 1995 78-67 87 47-68 66 +21 Detroit AL 1997 79-83 79 53-109 53 +26 San Francisco NL 1997 90-72 90 68-94 68 +22 Chicago NL 1998 90-73 89 68-94 68 +21 San Diego NL 1998 98-64 98 76-86 76 +22 Arizona NL 1999 100-62 100 65-97 65 +35
The 1991 season saw both league champions ascend from last place. The Twins won the World Series, but the Braves accomplishment was the more impressive because (a) they were a much worse team in 1990, and (b) they remained a great team for twelve years. The Braves got to the top on the strength of the simultaneous breakthrough from three young starting pitchers (Tom Glavine, Steve Avery and John Smoltz), and the influx of a few veteran free agents, one of whom became the league's MVP (Terry Pendleton).
The 1990 Pirates, propelled by Barry Bonds's first great year, won their first of three straight division titles. After the 1992 season, Pittsburgh elected not to sign Bonds when he attained free agency (having made the decision a year earlier to instead throw huge jack at Andy Van Slyke). Bonds fled to the Giants, who promptly gained 31 games while the Pirates regressed by 21. This was all likely a coincidence.
The biggest one-year improvement of the past 50 years, and arguably the most impressive advance by any team in history, was accomplished by the 1999 Diamondbacks. To find another team that gained 35 EQW in a single season you have to go back to the 1946 Red Sox, whose improvement was completely related to an historically unique season, and then to the 1903 Giants and 1902 Browns, who played in the midst of the turmoil caused by the creation of the American League.
After playing like the first-year expansion team that they were in 1998, Arizona's improvement came largely through incredible seasons from several aging veterans. Randy Johnson signed as a free agent and pitched like Randy Johnson, but the bulk of the team's advance was due to its offense, which improved from 665 runs to a league-leading 908. Matt Williams, Jay Bell, Luis Gonzalez, and Steve Finley combined for 77 home runs for various teams in 1998, but 133 for the D-Backs in 1999. As great as the team was, it was also very old. Accordingly, their story did not seem like a recipe for success, and the future for this ballclub did not look very bright. Umm, except that they won the World Series in 2001. With all the above players save Bell still on the team, the club is quite a bit older now. Nonetheless, no one should be too surprised if they win the World Series again. Their great players are old, but, then again, their old players are great.
2000-02 and Beyond
Baseball may be entering a period of lesser competitive balance. Last year's standard deviation in winning percentage (.092, or 14.8 wins), was the highest since 1969, and the largest in a non-expansion season since 1954. What this means, essentially, is that there has been an increase of teams with either high win or high loss totals. There were four 100-loss teams in 2002, the most ever, and three 100-win teams. In the entire 1980s, there were only 18 teams who did either.
The frequency of great advancements appears to be accelerating as well, with seven teams improving by 20 EQA since 2000, and twelve in the past six years.
Chicago AL 2000 95-67 95 75-86 75 +20 St. Louis NL 2000 95-67 95 75-86 75 +20 Seattle AL 2001 116-46 116 91-71 91 +25 Chicago NL 2001 88-74 88 65-97 65 +23 Houston NL 2001 93-69 93 72-90 72 +21 Philadelphia NL 2001 86-76 86 65-97 65 +21 Anaheim AL 2002 99-63 99 75-87 75 +24
If this trend holds, we can expect to see approximately two teams per year improving by 20 or more wins. Who is most likely to pull of this trick in 2003?
I asked several Baseball Prospectus authors to provide one or two teams that they thought were most likely to do it. All eight respondents chose the Cubs, and five of the eight picked the Rangers as a second team. No other club was mentioned more than once. I would say that this qualifies as consensus.
The Cubs have several things going for them. They are coming off a 21-game decline, from 88 wins to 67, which is often followed by some rebound the next year. The Cubs seem to have been quite unlucky, as their run differential (706-759) is typically associated with a team that would win around 76 games. They have a new manager, Dusty Baker, who was very successful in San Francisco. Most importantly, the Cubs are loaded with good young players with high upside-Hee Choi, Bobby Hill, Corey Patterson, Mark Prior, and Kerry Wood (who is still just 25).
Texas, which won 72 games last year, was likely quite a bit better than that. Their run differential, like the Cubs, indicates a team that ought to have won several more contests, about 78 in the Rangers case. A lot of the blame for this discrepancy likely lies with the bullpen, which has been bolstered in the off-season. Texas has a couple of superstars already, Alex Rodriguez and Rafael Palmeiro, plus two of the better young breakout candidates in the game, Hank Blalock and Mark Teixeira. Playing in the best hitters park in the league, the offense ought to score 900 runs, and might score 1000. The team hired Buck Showalter, one of the best managers of the past 20 years. If the pitching improves a little (they allowed 882 runs last year), it is not hard to envision a 20-win improvement.
That having been said, if baseball analysts had been polled a year ago on this same question, would anyone have suggested the Angels? Likely not, since all 14 Baseball Prospectus writers, and most of the rest of us for that matter, picked Anaheim to finish last. The darling of the analysts last year at this time was San Diego, who subsequently declined from 79 to 66 wins.
Every baseball season brings wonderful surprises. It is likely that one or more of those surprises will involve a team advancing by 20 wins.
Mark Armour is the co-author (with Dan Levitt) of Paths To Glory, the stories of the building of several interesting baseball teams, to be published this Spring by Brasseys, Inc, and the director of SABR's Biography Project. You can contact him at firstname.lastname@example.org.