Winning double headers
Lou Bershad
harder to win both games of a double header than to win two games in two
days." I wonder if that statement is true.
I'd be interested in anyone's analysis.
What percentage of double headers are split? If there is no
double-header-effect, I would suspect that sweeps slightly outnumber splits.
If the better team wins 51% of the first games, then the chance of a split
is .51*.51+.49*.49=.5002.
Another way to look at this would be to actually compare the sweep/split
ratio of double headers to the sweep/split ratio of two consecutive games
against the same team.
Thanks in advance,
Lou
Gerry Myerson
"Lou Bershad" <loube...@yahoo.com> wrote:
Here are two messages that were posted to the SABR mailing list
9 years ago (to the day, pretty much):
Date: Thu, 14 Sep 2000 14:20:48 EDT
From: "Thomas J. Ruane" <tru...@VNET.IBM.COM>
Subject: Joe Morgan on Doubleheaders
Neil Cohen was wondering whether or not Joe Morgan was correct when
he wrote: "It's a little known fact that 80% of doubleheaders are
split." Well it may be little known, but it certainly isn't a fact.
Here are the results of all double-headers since 1980. The legend:
# -- total DHs
H -- the home team swept
V -- the visiting team swept
S -- the teams split
O -- other (one of the games was a tie)
# H V S O
1980 99 27 23 48 1
1981 47 16 9 21 1
1982 79 22 17 40 0
1983 94 31 15 48 0
1984 76 26 22 28 0
1985 53 15 11 27 0
1986 47 20 7 20 0
1987 33 13 7 13 0
1988 41 12 9 20 0
1989 47 16 6 25 0
1990 46 13 13 20 0
1991 29 8 5 16 0
1992 36 10 8 18 0
1993 27 9 3 15 0
1994 20 4 1 15 0
1995 20 8 4 8 0
1996 36 12 7 17 0
1997 45 7 9 29 0
1998 37 13 4 20 0
1999 31 11 7 13 0
943 293 187 461 2
So 80% of DHs are NOT splits. As a matter of fact, fewer than half
are. Which, after all, is what you'd expect.
Tom Ruane
************************************************************************
Date: Fri, 15 Sep 2000 00:35:08 -0700
From: Marc Alan Jones <marc...@EARTHLINK.NET>
Subject: Re: Joe Morgan on Doubleheaders
On 9/14 Neil Cohen informed us:
On the ESPN broadcast of the Indians - Red Sox game last night, the
announcers were talking about the large number of doubleheaders that
both
teams will be playing in the next week. Joe Morgan then said "It's a
little known fact that 80% of doubleheaders are split."
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Between 1901 and 1998 (I'll update the spreadsheet someday) there were
14,641 doubleheaders played in the majors. The results break down as
follows:
Split - 6,905 (47.2 %)
Home Sweep - 4,586 (31.3%)
Visitor Sweep - 3,150 (21.5%)
Joe was a bit wide of the mark on this one. I wonder if he was quoting
from some off-base source material, like the Drudge Report, or if he was
just making a WAG (wild ass guess). He might have been thinking about
that fun filled season of 1994, when there were only 20 double-headers,
and 15 of them were split (75%). Except for that year and 1997 (63%
splits), there has never been a season where even 56% of the double dips
were split.
One more trip down memory lane. These are the average number of
doubleheaders played per team per season, collated by decade:
1901-10: 9.2
1911-20: 10.0
1921-30: 9.8
1931-40: 12.8
1941-50: 14.2
1951-60: 10.5
1961-70: 8.8
1971-80: 5.7
1981-90: 2.2
1991-98: 1.1
Two games in one day is definitely going the way of the dodo.
Marc Alan Jones---
---Who managed a split once, and needed months of physical therapy to
recover (Those of you old enough should picture Harmon Killebrew at the
1968 All-Star game. It was sort of like that.).
--
Gerry Myerson (ge...@maths.mq.edi.ai) (i -> u for email)