Baseball for the Thinking Fan

Login | Register | Feedback

You are here > Home > Primate Studies > Discussion
Primate Studies
— Where BTF's Members Investigate the Grand Old Game

Sunday, July 07, 2002

Pennants, Won-Loss Records and July 4th: Before and After

Don Malcolm studies how well that old adage about the baseball standings holds up.



Among other things, baseball is a game of old adages. Space and
time do not permit me to start listing all of them here, but one of
‘em that you don’t hear as much since the advent of divisional play is
about to get plucked out of the mothballs:

The team that is leading the league (division) on July 4th is
the team that wins the pennant (division).

Yes, yes, it’s coming back to you. It’s not exactly au
in this Age of the Wild Card, but it still has a certain
ring to it. This old adage, like the majority of such sayings, seeks
to encapsulate and embrace the idea of order and stability. Most of
the time, it claims, good teams are good all year long.

So what does the data tell us about this old adage? Thanks to


Retrosheet, we have game log
data for all of twentieth century baseball. First, let’s look at the
results from the years prior to divisional play (1901-1968). In those
136 pennant races, the team leading the league after play on July 4th
went on to win the pennant 88 times.

That’s a little under 65% of the time.

As adages go, that’s not too bad. It’s not as if it isn’t true at
least half the time or something.

Since the invention of divisions, the old adage has actually
become more robust. Looking at results since 1990 (and ignoring the
wild card for these purposes), we find that of the fifty-eight
divisional races occurring in that time frame (we are also not
counting 1994 . . .), forty-one of them were won by the team in the
lead after play on July 4th.

That’s a little over 70% of the time.

But let’s not stop here by filing this into “the more things
change” cubbyhole. There’s another way to look at this data that might
be a lot more interesting.

Instead of looking at whether the team in the lead after July 4th
goes on to win the pennant (division), let’s ask a different question.

How many times does the team in a league (division) with the
best won-loss record up to July 4th also have the best won-loss record
in a league (division) after July 4th?

Ah, I’m glad you asked that question. It turns out that there is a
big, big difference in the results when we segregate the seasons into
two rough halves.

From 1901-1968, teams with the best record in “the first half” (up
through July 4th, as we’re defining it here) had the best record in
“the second half” (July 5th on) only 52 times out of 136 pennant

You can this, if you want, the “second half repeater rate.” And
that rate is only 38%.

If the Lords of Baseball had decided to implement a “first
half/second half” split in 1901, they would have had a natural “first
round playoff” scenario in 62% of all seasons up through 1968,

Looking at those same figures for our recent sample (1990 on), we
see that the results are similar. Out of those fifty-eight divisional
races, only twenty-five featured teams that had the best record in
both “halves” of the season. That “second half repeater rate” is also
up a bit (43%), as we saw earlier with the “lead on July 4th and win
the division” rate.

Just think about all those great playoff series in the past that
never occurred because revenue streams hadn’t evolved past the
babbling brook stage.

For example, let’s take the 1950s as a way of looking at
this. Right off, we run into a roiling anomaly that would disrupt this
notion of an “organic” first half/second half playoff.

In the NL, the team with the best record for the entire year, of
course, was the Whiz Kid Phillies. But those Whizzes were coy about
it, as you’ll see.

Tm   W   L   WPct     Tm     W   L   WPct
STL   41   27   .603     NYG   52   33   .612
PHI   40   28   .588     BRO   52   35   .598
BSN   39   30   .565     PHI   51   35   .593
BRO   37   30   .552     BSN   44   41   .518
NYG   34   35   .493     CIN   41   45   .477
CHC   32   33   .492     STL   37   48   .435
CIN   25   42   .373     PIT   34   50   .405
PIT   23   46   .333     CHC   32   56   .364


The data on the left is the NL standings up through July 4th,
1950. The data on the right is the NL standings from July 5th
on. Notice that the Whiz Kids aren’t at the top in either list, but
have the most wins overall.

What to do? Do you let the Cardinals play the Giants-two teams who
had only one good half? Do you add the Phillies in as a third wheel

I’m not going to answer those questions here. You can find a more
lengthy examination of this matter over at

HREF=""> Suffice
it to say that there are several different ways to create an
interesting playoff here, even with a “messy” scenario such as the one
above (which is, thankfully, rare).

Over in the AL that year, the Detroit Tigers snarled their way to
a sizable first-half lead (five games) over the New York Yankees, who
turned things up in the second half. There’s a good playoff series
(still) waiting to be played there.

The next year, 1951, already had the great playoff between the NL
teams (Dodgers and Giants). The “perfection” of that series is due
mostly to the dramatic moment that concluded it, but could it have
been even more memorable if it had been a best-of-five or
best-of-seven series? I’ll let you argue about that.

In the AL, you have the first-half White Sox vs. the second-half
Yankees, who just barely edged out the Indians to qualify.

In 1952, its first-half Brooklyn vs. the second-half Phillies in
the NL, and the Yanks (first half) vs. the second-half Indians in the

In 1953, the Dodgers and Yankees led in both halves, so we skip
the playoffs and go right to the World Series. (That sound you hear is
a nation of TV executives dropping their jaws all the way to the

In 1954, the Indians led wire-to-wire, having the best second-half
record despite the valiant efforts of the Yankees (who are one of only
five teams to play .700+ ball from July 5th on and not win a pennant
or division). In the NL, however, Durocher’s Giants, who led in the
first half, would have to get by the hard-charging Milwaukee Braves
(53-28 in the second half). Who knows, there may never have been
Mays’s catch. . .

In 1955, the Yanks and the Indians square off in the AL, while the
Dodgers and Braves go at it in the NL. The second half standings in
the NL are especially close, by the way, with the Phillies only a game
back of Milwaukee.

In 1956, it’s the first-half Yanks vs. the second-half Tigers in
the AL, and another showdown between the Braves and Dodgers in the NL.

In 1957, the Yanks are wire-to-wire, so no AL playoff. Over in the
NL, it’s the first-half Cincinnati Reds (who just missed in the second
half of ‘56) vs. the second-half Braves.

In 1958, the Braves led wire-to-wire, so the NL gets a bye. In the
AL, the Yankees streaked to a first-half win, but the White Sox were
the best in the second half, so we get a very interesting playoff
series there.

In 1959, there was a pennant playoff in the NL, as you may
remember. In our scenario here, however, the identities of the two
teams are a bit different. The Giants had the best record up through
July 4th, and the Dodgers were the best from July 5th on. However, the
Braves and the Dodgers wound up tied at the top for the overall
season. We’re back to where we came in with the anomaly in the 1950
NL. (As noted, I discuss scenarios for resolving this in the “expanded
coverage” of this topic over at

The 1959 AL gives us a playoff between the first-half White Sox
and the second-half Indians.

That’s just a taste of what things could have been like in the
pre-divisional days. Looking since the three-division setup and
creation of the Wild Card in 1995 to the present, there are fourteen
additional teams that could have qualified for post-season play due to
having the best record in either the “first half” or the “second

It’s quite likely that such a scenario will seem too chaotic to
many of you, and that’s fine. Keep in mind, however, that our
perception of such matters might be entirely different if the “split
schedule experiment” of 1892 had been deemed a success. The two
different ways of looking at “before and after July 4th” should
indicate that our perceptions are much more malleable than might at
first seem to be the case.

Let’s close by honoring the teams with the best “before and after
July 4th” performances. These are the teams who played .700 ball or
better in either half. The lists are almost identical in length; 23
first-half teams at .700 or better, vs. 24 second-half teams. First,
here are the top “first half” performances:

Team Year   W-L   WPct
SEA   2001   61-22   .735
NYY   1998   60-20   .750
NYM   1986   54-21   .720
CIN   1970   56-23   .709
BKN   1955   55-22   .714
CLE   1954   52-22   .703
BKN   1952   49-21   .700
STL   1944   47-19   .712
BKN   1942   51-20   .718
NYY   1939   52-17   .754
NYG   1935   47-19   .712
PHA   1931   51-20   .718
PHA   1929   52-17   .757
NYY   1928   54-18   .750
NYY   1927   53-21   .716
CHC   1918   47-19   .712
NYG   1912   54-13   .806
PIT   1909   45-18   .714
CHC   1907   52-16   .765
CHC   1906   49-21   .700
NYG   1905   50-20   .714
NYG   1904   48-16   .750
PIT   1902   44-12   .786


I’ve ordered these not by WPct but in reverse chronological
order. That shows us a couple of things that we might not see
otherwise. First, the number of games before July 4 th has steadily
increased, even before television dictated that the season begin by
April Fools’ Day. Second, we can see how rare blazing first half
performances have been since the mid-fifties-only four in the past
forty-seven years.

Now let’s look at the best “second-half” performances:

Team Year   W-L   WPct
OAK   2001   61-18   .772
ATL   1993   53-23   .712
KC   1977   61-24   .718
BAL   1970   59-24   .711
NYY   1961   61-25   .709
CLE   1954   59-21   .738
NYY   1954   54-23   .701
BKN   1953   60-22   .732
BOS   1949   61-22   .735
BOS   1948   63-27   .700
STL   1943   64-26   .711
NYY   1943   63-27   .700
STL   1942   65-19   .774
CHC   1935   62-23   .729
DET   1934   57-24   .704
NYY   1932   58-24   .707
NYY   1927   57-23   .712
BOS   1915   64-26   .711
BSN   1914   68-19   .782
BOS   1912   56-24   .700
PIT   1909   65-24   .730
CHC   1909   64-25   .719
CHC   1906   66-15   .815
PIT   1902   58-24   .707


For all you A’s lovers out there, you can put their
post-Independence Day performance of 2001 into historical perspective
at last. Their .772 clip was the fourth best post-July 4th record
since 1901, behind the 1906 Cubs, the 1914 Braves, and the 1942

As noted above, there have been five teams to play .700+ ball from
July 5th on and not win a pennant or division. The 2001 A’s
join the 1954 Yankees, the 1948 and 1949 Red Sox, and the 1909
Cubs. The other four teams, of course, had to go home.

The only teams common to both lists: 1902 Pirates, 1906 Cubs, 1909
Pirates, 1927 Yankees, 1954 Indians.

Don Malcolm Posted: July 07, 2002 at 06:00 AM | 2 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. GMR Posted: July 08, 2002 at 12:34 AM (#605482)
Dan Schmidt:

According to Retrosheet (, the Tigers were 55-25 for a .687 percentage at July 4. My math says that they played .500 ball (20-20) immediately after the hot start.
   2. GMR Posted: July 08, 2002 at 12:34 AM (#605492)
Very interesting article. It got me thinking about the flip side of the coin (stats from Retrosheet,, and this article):

1) What teams have had the worst first/second half of the season? Some quick guesses (no exhaustive research here):
The 1962 Mets were 21-57 (.269) at the midpoint, 19-63 (.232) afterwards.
The 0-fer Orioles of 1988 were 24-57 (.296) at the midpoint but "rebounded" to go 30-50 (.375) in the second half.
The 1952 Pirates were 21-56 (.272) in both halves of the season.
The 1942 Phillies were 19-54 (.260) at the midpoint, 23-55 (.295) aftewards. The year before, they were 20-53 (.273) in the first half and 23-58 (.284) in the second.
The 1935 Braves were 20-50 (.285) at the midpoint, 18-65 (.217) afterwards.
The 1919 Athletics were 15-43 (.258) at the midpoint, 21-61 (.256) afterwards.
The 1916 Athletics were 17-47 (.265) at the midpoint, 19-70 (.213) afterwards.

2) What teams have the biggest differential of winning percentage between first and second halves?
The 2001 Athletics had, as the article mentioned, a 61-18 (.772) second half. This followed a 41-42 (.493) first half, a difference of about 279 percentage points.
Slow starts by the Red Sox in 1948 (33-32, .507) and 1949 (35-36, .492) kept them out of the postseason despite second halves of 63-27 (.700) and 61-22 (.735), respectively. The differentials are .193 for 1948 and .243 for 1949.
The 1942 Cardinals, 63-27 (.774) for the second half, were only 41-29 (.585) at the midpoint for a differential of .189.
The 1914 Braves were 26-40 (.393) on July 4 and dead last in the N.L. when they embarked on their incredible 68-19 (.782) second half, a differential of .389! Are there any bigger gaps than that?
None of the great first halves mentioned in the article jumped out at me as coming before a second-half collapse.

You must be Registered and Logged In to post comments.



<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF


Thanks to
Randy Jones
for his generous support.


You must be logged in to view your Bookmarks.


Page rendered in 0.2901 seconds
58 querie(s) executed