Competitive Balance

Vintage BaseballIn sports law, the concept of “competitive balance” has played an important role as a justification for practices that would otherwise run afoul of the antitrust laws and the common law of unfair competition.  The conventional logic has it that unless teams in an individual sports league are reasonably competitive, fans of the weaker teams will lose interest, which will cause those teams to fold, which will in turn threaten the survivorship of the league itself.  Consequently, restraints on player mobility, like the amateur player drafts, salary caps, and reserve clauses, and territorial restrictions on franchise location and broadcast rights have been justified in the past (at least by their advocates) as necessary to maintain “competitive balance” in sports leagues.

Exactly what constitutes “competitive balance” and whether or not it has existed at particular times has been a matter of some controversy.  Should competitive balance be measured by the closeness of competition during an individual season, or should it be measured by the frequency with which different teams win championships?  Does competitive balance mean the same thing with regard to each sport?

Although the National Football League is often lauded for its competitive balance, within any given NFL season there is always great disparity between the winning percentages of the most and least successful teams.  For example, during the 2009 season, the Indianapolis Colts, the team with the best record (14-2) won 87.5% of its games while the St. Louis Rams, the team with the worst record (1-15) won only 6.3% of its games.  Overall, the Colts winning percentage exceeded that of the Rams by .812.

In contrast, in Major League Baseball in 2009, the winning percentage of the most successful team in the regular season, the New York Yankees, exceeded that of the worst team, the Washington Nationals, by only .272 (.636 to .364).  The 2009 gap in baseball was actually somewhat higher than usual; in 2000, admittedly a very balanced year, the gap between the best team, the San Francisco Giants, and the worst teams, the Chicago Cubs and Philadelphia Phillies, was only .196 (.597-.401).

The disparity between baseball and football is often attributed to the very short (in terms of number of games) NFL season.  While this may be part of the explanation, it is not the whole answer.  If one aggregates individual team records for the past ten NFL seasons (160 regular season games), the best team, the New England Patriots compiled a record of 112-48, for a winning percentage of .700, a mark exceeded in a single season by only three major league baseball teams over the past 70 years.  The worst team, the Detroit Lions, has a combined record of 42-188, for a winning percentage of only .262, a mark exceeded by only one baseball team (the 1962 New York Mets) since 1935.

Basketball season records also show much greater variation between the best and worst teams than is found in baseball.  If the NBA’s current best (Cleveland Cavaliers) and worst (New Jersey Nets) teams finish the season with their current winning percentages, they will have accomplished the equivalent of a major league baseball team finishing its season with a record of 126-36 and 16-146, respectively.  In the 134 year history of major league baseball, no team has ever done that well, or that poorly.  In the NHL, the performance range is more narrow, but still much broader than in Major League Baseball.

The following text, which I originally submitted to the Listserv of the Society of American Baseball Research, explores the status of the competitive balance questions in 1960, which was just before the modern explosion of the number of major league sports teams in the United States and Canada.  In 1960, there were a total of only 42 teams in Major League Baseball, the NFL, the NBA, and the NHL.  By 2004, that number had grown to 122.

Now that we are almost 50 years removed from the expansion of major league baseball that occurred in 1961, it makes more and more sense to use 1960/1961 as a dividing point in the “periodization” of baseball history.  Although relatively little changed in 1961 itself, it ushered in an era of significant changes in the way in which major league baseball was organized and, with the adoption of the designated hitter rule and the construction of multi-purpose stadiums, played.

This leaves us with three basic divisions—the Formative Era (1871-1900), the Golden Age (1901-1960), and the Modern Era (1961-present).  I will acknowledge that this choice of terminology has “Baby Boomer” written all over it.

At the end of the Golden Age, Major League baseball had achieved a measure of competitive balance missing in the highest leagues of other major sports.  The simplest measure of competitive balance is the index of how frequently lower performing teams defeat their higher performing counterparts.  If we rank the major league baseball teams of 1960 in regard to their final regular season records, we find that in almost 42% of games, the team that would finish lower in the standings at the end of the year nevertheless won the game.  In 1960, the lower ranked team came out victorious in 515 of 1232 games, for a winning percentage of .418  The NL was slightly more competitive than the AL, with its lower ranked teams having a winning percentage of .429, compared to .407 in the AL.

In the other sports, the teams that finished higher in the standings won much more frequently.  In the NHL season of 1959-60, higher ranked teams finished 113-61-36 in games with lower ranked opponents, which translates into a .376 winning percentage for the lower ranked teams (using the NHL’s system of counting a tie as the equivalent of one-half of a win).  In the NBA, the overall record of higher ranked teams was 211-89, which meant that the lower ranked teams won only 29.7% of the time.  The greatest disparity was in the NFL (in which teams played only 12 regular season games).  Not counting two games between teams that ended the season with identical records, the higher ranked NFL teams finished the season, 57-14-5, which translated to a .803 winning percentage by the method used at the time and to .783 by the current standards.  Either way, lower ranked teams won only about 20% of the time.

There were, of course, major league teams in 1960 that baseball fans considered “terrible,” particularly the Philadelphia Phillies (59-95) and the Kansas City Athletics (58-96), the last place finishers in their respective leagues.  However, the last place teams in the other leagues did much worse in 1960.  In the NFL, the Dallas Cowboys were 0-11-1, and the Washington Redskins 1-9-2 (defeating only the Cowboys).  The worst team in the NBA, the Cincinnati Royals, finished 19-56, which was the equivalent of 41-112 in baseball, while the last place finisher in the six-team NHL was the New York Rangers, whose record of 17-38-15 translates to 54-100 in 1960 baseball terms.

I am not arguing that this level of balance was a feature of major league baseball that developed only at the end of the Golden Age.  It may well have been the pattern throughout the history of major league baseball.  It is also my impression that the disparity between baseball and other major league team sports that can be seen in 1960 has continued into the modern age. However, I am suggesting that the factors that have led to greater competitive balance in baseball than in other sports do pre-date the modern era.

Continue ReadingCompetitive Balance

President for a Day

Atchison_David_Rice_-_Plattsburg_MO_3Today marks the 161st anniversary of the one-day Presidency of the United States of David Rice Atchison. Atchison’s story, once well-known, has been reduced to an item of trivia for American history buffs.

By common agreement, the Presidential term of James Knox Polk ended at noon on March 4, 1849. March 4 was a Sunday that year, and the decision was made to delay the inauguration of Zachary Taylor, the hero of the recently concluded Mexican-American War and the newly elected President, until Monday, March 5. Polk completed his final presidential duties on Sunday morning and then departed from Washington that afternoon.

Who then was President between noon on March 4, when Polk departed, and noon of March 5 when Taylor finally took the oath of office?

Under the presidential succession statute in effect in 1849, the President Pro Tempore of the Senate was to assume the office of President of the United States if neither the President nor the Vice-President were able to do so. During the 30th Congress, David Rice Atchison, a senator from Missouri, had been the President Pro Tem of the Senate, and on the morning of March 5, 1849, he was reelected to the same post for the 31st Congress. Legend has it that Atchison spent the day of March 4 in bed, recovering from his participation in several parties held on the evening of March 3.

At the time, no one seemed to have noticed that the office of President had fallen vacant, and there is no contemporary evidence that Atchison himself suggested to anyone that he was the President. However, later in life — Atchison was only 32 on March 4, 1849, and he lived until 1886 — he frequently claimed that he had been President of the United States for one day.

The question of whether or not Atchison had actually been President was raised by a handful of commentators in January 2009, when Chief Justice Roberts botched the oath of office taken by President Obama. Obama retook the oath, correctly, the following day, but if a proper recitation of the oath is a prerequisite to becoming President of the United States, then who was President on January 20, and who was president on January 21? (Since Vice-President Biden had been sworn in at that point he presumably would have been next in line.)

There are, however, a number of problems with the claim that Atchison was President of the United States on March 4, 1849. First of all, he failed to take the oath of office, and the Constitution specifically states that the President must either swear or affirm the prescribed oath.

Even more problematic is the fact that Atchison was technically not the President Pro Tempore of the Senate on March 4, 1849. Under the practices of the day, the office of President Pro Tempore of the Senate was understood to expire when the current Congress recessed for the final time. The first order of business for the new Congress was to elect a new President Pro Tem, although it was not unusual for the Senate to reelect the previous holder of the position.

The Thirtieth Congress recessed on March 3, 1849, so Atchison’s term ended the day before Polk’s presidential term expired at noon. Moreover, his new term as President Pro Tem of the Senate during the 31st Congress did not begin until the morning of March 5. Consequently, if Atchison was eligible to serve as President, it was only during the time between his swearing in on the morning of March 5, and Taylor’s taking the oath of office at noon.

Actually, by this logic, Atchison’s already truncated presidency would have ended with the swearing in of Vice-President Millard Fillmore shortly before noon, and Fillmore’s presidency would have ended a few minutes later when Supreme Court Chief Justice Roger Taney swore in President Taylor. Ironically, Atchison terminated his own presidency (under this theory) when he, as the newly elected President Pro Tem of the Senate, swore in Vice-President Fillmore. (Fillmore, of course, became President “again” in July 1850, when Taylor died in office.)

As it turns out, the Sunday presidential inauguration problem has occurred a number of times in American history. Logically, it should occur once every 28 years or so, since there are seven days to the week and inaugurations occur every four years.

The problem first occurred on March 4, 1821, when James Monroe was President. His inauguration for his second term was delayed for one day, and no one at the time appears to have noted any gap in the presidency. John Gaillard of South Carolina, who served as President Pro Tem continuously from January 25, 1820, to December 2, 1821, never claimed to have been President on March 4, 1821.

The second incident was the one involving Taylor and Atchison. The third came in 1877, following the Hayes-Tilden race, the most contested presidential election in American history. Although the results of the election were not finally determined until Friday, March 2, Hayes, for religious reasons, refused to be sworn in on Sunday, March 4. Several Republicans, including outgoing President Ulysses S. Grant, feared that disgruntled Democrats might attempt to disrupt the inauguration ceremonies, so they encouraged Hayes to take the oath of office as soon as possible. Consequently, Hayes took the oath of office on Saturday, March 3, the day before the end of Grant’s term. He also took the oath a second time (without incident) at a public ceremony on Monday, March 5.

In 1917, Woodrow Wilson delayed his second inauguration until March 5, but as had been the case with Monroe a century earlier, there was no evidence of contemporary claims that the presidency had fallen vacant.

In 1933, the 20th Amendment to the Constitution moved the presidential inauguration date to January 20. The first conflict with the new date came on Sunday, January 20, 1957. President Eisenhower, who had been reelected the previous November, took the oath of office privately on Sunday, and then was publicly inaugurated the following day. Ronald Reagan took the same approach when the date of his second inaugural fell on Sunday, January 20, 1985. Either President Obama or his replacement will have to deal with this issue in three years, as January 20, 2013, will fall on a Sunday.

The logical answer here is that the newly elected President becomes President on the appointed day, whether or not he (or she) has an inaugural ceremony or takes the required oath. The supposed problem of a gap in the presidency in 1849 should remind us that for all of their wisdom, the Founding Fathers did not think of everything, and there are times when the Constitution requires us to use common sense to implement its central provisions.

A picture of the Plattsburg, Missouri tombstone of David Rice Atchison—which proclaims him “President of the United States for One Day”—can be found here.

I am indebted to Sean Samis for calling this anniversary to my attention earlier today.

Continue ReadingPresident for a Day

A Different Way to Run the Electoral College

800px-Simple2008PresElections-USA-statesIn an earlier posting, Rick Esenberg expressed his opposition to recent George Soros-sponsored efforts to devise a plan to circumvent the operation of the Constitution’s venerable Electoral College.

The “problems” with the Electoral College are well-known.  Its “winner-take-all” feature supposedly distorts the electoral process, and on four occasions (1824, 1876, 1888, and 2000), it has chosen a president who received fewer popular votes than one of his opponents.

Debates over the future of the Electoral College often assume that there are only two options:  scrap the institution altogether or else accept that it will continue to operate as it has in the past.  Scraping the Electoral College is usually assumed to require a constitutional amendment, although the Soros plan would actually leave the Constitution unchanged but seek to bind electors to cast their votes for the candidate with the largest national popular vote regardless of the results in their particular state.

There is an alternative, however, that would make the results of the Electoral College more democratic but would leave the Constitution unchanged.

Although it has long been the practice that individual states award their entire electoral vote allotment to the candidate receiving the largest number of popular votes in the state, there is no constitutional requirement that states follow this approach.  In the first part of the 19th century, many states split their electoral votes, and today two states, Maine and Nebraska, have abandoned the “winner take all” rule.  (In 2008, Nebraska split its electoral votes between McCain and Obama.)

In the first third of the 19th century the selection by a single state of electors supporting different candidates was a frequent feature of American presidential elections.  As late as 1824, five of 24 states did just that.  In 1828, three states did so, including New York, which cast 20 votes for Andrew Jackson and 16 for the incumbent John Quincy Adams.  By the mid-1830’s, every state appears to have embraced the winner take all system, although in 1860, New Jersey reverted to the older system and ended up dividing its seven electors between the Republican Abraham Lincoln (4) and the Northern Democrat Stephen Douglas (3)

Article II of the United States Constitution makes it clear that each state has the power to adopt a different system of choosing electors, should it choose to do so.  There are at least three ways that states could chose to apportion their electors that would produce more equitable allocations of electoral votes.

A state could simply apportion its electors on the basis of the popular vote.  For example, in 2008, Wisconsin’s 10 electoral votes could have divided six for Obama and four for McCain based upon Obama’s 56% to 44% margin of victory.  On the other hand, such a system can run into problem because of the presence of “third” parties.  Like Wisconsin, Minnesota currently has 10 electoral votes.  In 2008, Obama and McCain received 54% and 44% of the popular vote, respectively, with 2% going to third parties.  The Constitution requires the appointment of actual electors, so it is not obvious as to whom the tenth Minnesota elector would be assigned.

A second possibility would be for a state legislature to divide the state into “electoral districts” with each district choosing one elector.  Because each state has two more electors than it has representatives in Congress, congressional districts could not be used for this purpose.  Instead, states would have to create special districts.  Wisconsin, for example, would have to be divided into ten districts whose sole function would be the selection of presidential electors.  Obviously, this could be done, but the process would involve many of the same difficulties that arise when the state has to redraw the lines of congressional districts.

The third possibility is the approach currently taken in Maine and Nebraska.  The presidential candidate receiving the largest number of popular votes in each congressional district receives one elector, and the winner of the statewide popular vote is awarded an additional two electors.  In Wisconsin in 2008, Obama received the largest number of popular votes in seven of the state’s eight Congressional Districts.  Consequently, under the Maine/Nebraska system he would have received nine of the state’s ten electoral votes with the other vote going to McCain.

While this approach would have had only a slightly different effect in Wisconsin (where Obama received all ten electoral votes), there are other states where the impact would have been greater.  In Ohio, for example, this system would have divided the state’s twenty electoral votes evenly between Obama and McCain instead of having all twenty go to Obama.  Even in large states won easily by one candidate or the other, the losing candidate would have collected electoral votes.  Under this system, McCain would have received eleven electoral votes in California and four in New York, while Obama would have received eleven votes in Texas and five in Georgia.

Obviously, the great advantage of this system is that it does not require the creation of additional electoral districts.  It would also create an incentive for candidates to campaign in states even if they perceived that they had no real chance of winning the overall popular vote in that jurisdiction.  On the other hand, it is also possible that this system might make third party candidacies more popular since it would not be necessary to carry entire states to have a presence in the Electoral College.

Had the Maine/Nebraska system been in effect in every state in 2008, the final result would have been a 301-237 victory for Barack Obama, compared to the actual margin of 365-173.  Very few states under this system would have awarded all of their electoral votes to a single candidate.  In fact, if we remove the eight jurisdictions with three electoral votes (where a divided vote was impossible), 32 of 43 jurisdictions would have divided their electoral votes.  Of the eleven that would have voted unanimously, only four had more than five electoral votes, and only one (Massachusetts) had more than ten.

Although the allocation of electoral votes would have looked much different, the Maine/Nebraska system would not have produced a different result in any of the last three presidential elections (which are the only ones for which I have recalculated the vote).  Obama would have won in 2008, although not by as large a margin.  Earlier in the decade, the geographic dispersion of his supporters would have meant that George W. Bush would have twice been elected president under the district system.  In fact, he would have won by even larger margins under this system than he did in 2000 and 2004, even though he received fewer popular votes than Al Gore in 2000.  Under the Maine/Nebraska system, Bush would have defeated Gore in 2000 by a margin of 288-250 (rather than 271-266), and John Kerry in 2004 by 317-221 (instead of 286-252).

Even though the Maine/Nebraska system would not eliminate the possibility of a “minority” president, it would “decentralize” the electoral process and would better protect the rights of those who reside in less populous parts of the country to participate in the presidential election process in a meaningful way.

The challenge of course is to convince the other 48 states that it would be in their and the nation’s best interest to adopt the Maine/ Nebraska approach.  Unless all states adopted the “district” approach, states retaining the winner-take-all system might actually become even more important.  As critics of this approach have pointed out in the past, the original system of district elections disappeared in the early 19th century because of a perception that the winner-take-all states were able to exercise greater influence.

Of course, states could be required to adopt a district system, but that would require a constitutional amendment.

What follows is a state by state breakdown of electoral votes for the 2008 presidential election, had the Maine/Nebraska system been in effect in each state. The numbers following the name of each state are the electoral votes for Obama and McCain, respectively.  In states marked with an * Barack Obama was the recipient of the largest number of popular votes.

2008  Results, Alternative Approach to Choosing Electors.

Alabama  1-8
Alaska  0-3
Arizona  2-8
Arkansas  0-6
*California  44-11
*Colorado  5-4
*Connecticut  7-0
*Delaware  3-0
*Florida  12-15
Georgia  5-10
*Hawaii  4-0
Idaho  0-4
*Illinois  18-3
*Indiana  5-6
*Iowa  6-1
Kansas  1-5
Kentucky  1-7
Louisiana  1-8
*Maine  4-0
*Maryland 8-2
*Massachusetts  12-0
*Michigan  14-3
*Minnesota  7-3
Mississippi  1-5
Missouri  3-8
Montana  0-3
Nebraska   1-4
*Nevada  4-1
*New Hampshire  4-0
*New Jersey  12-3
*New Mexico  4-1
*New York  27-4
*North Carolina  8-7
North Dakota  0-3
*Ohio  10-10
Oklahoma  0-7
*Oregon  6-1
*Pennsylvania.. 11-10
*Rhode Island  4-0
South Carolina  1-7
South Dakota  0-3
Tennessee  2-9
Texas  11-23
Utah  0-5
*Vermont  3-0
*Virginia  8-5
*Washington  9-2
West Virginia  0-5
*Wisconsin  9-1
Wyoming  0-3
District of Columbia  3-0

Presumably the DC election would be on a winner-take-all basis, as it is in other states with three electoral votes.

Continue ReadingA Different Way to Run the Electoral College