July 18, 2008
Braking and Breaking at the Break
Here's the concept: the All-Star break disrupts the progress of teams riding high, and at the same time allows reeling teams to regroup. I heard something to that effect put forward during a game telecast on Sunday. I honestly don't remember who it was that said it, since I was surfing back and forth between several games at the time, and writing it down would have required actually rising up off the couch to find a pen and then sitting back down again, only to remember that I also needed paper and having to get back up to find that as well. Getting up that often in such a short period of time would set a dangerous precedent for activity; you see how it is. Take my word for it, though: something approximating the following exchange took place during a telecast on Sunday.
Announcer: The Mets have won eight in a row. I'll bet they're sorry to see the All-Star break arrive.
New York went on to win their ninth straight later that night, giving us two teams with opposing extended trajectories heading into the annual mid-season respite. Naturally, I was skeptical of this gentleman's thesis. Can three days off (or four in some cases) really change a streaking team's fortunes in ways that having to go back out the next night can't? One can certainly understand the inclination to believe this is so, but that can also be said about a large percentage of the baseball myths that have been busted in this modern era of reality checking.
The Mets' nine-game streak is tied for second-longest in the run up to the All-Star Game, going back to 1957. The longest was by the 1975 Cincinnati Reds, who reached 10 wins in a row before knocking off for the duration, and they then went on to win the World Championship. Tied with the Mets at nine are the 1995 Atlanta Braves, a team that also went on to win the over-sized novelty trophy. In the interests of full disclosure, however, it should be pointed out that the All-Star break in 1995 was only two days long, owing to compression from the strike-caused late start. Two teams do not a fair comparison make, however, so I extended the pre-break streak threshold to seven consecutive wins. This grew the field to 13 teams since 1957. Here they are along with what fate had in store for them upon their return to action after the All-Star Game:
1975 Reds (10 in a row): lost 1995 Braves (9): lost 1968 Red Sox (8): lost 1974 Pirates (8): lost 1998 Braves (8): won 1978 Twins (8): lost 1963 Senators (7): lost 1968 Cardinals (7): lost 1975 Red Sox (7): won 1984 White Sox (7): lost 1988 Expos (7): won 1991 Mets (7): won 1997 Pirates (7): lost
There are some good and very good teams on this list. The '68 Red Sox were defending league champions and went on to win 86 games. The '74 Pirates won their division and the '98 Braves won 106 games, while both the '68 Cards and '75 Red Sox went to the World Series. There are also some sub-par teams; the '78 Twins, '84 White Sox, '91 Mets and '97 Pirates were all under .500, while the '88 Expos broke even. (The Twins' loss coming back from the break started a nine-game losing streak.) The real outlier is the '63 Senators, who won just 56 games, enduring losing streaks of six, eight, and 10 games that year.
Together, this group went 4-9 upon their return from the break. Does that prove the theory that the All-Star holiday derails the speeding locomotive of victory? Aside from the fact that it's a small sample size, we should also consider how all streaking teams do when their run of victories gets that high. Over the past 50 years, teams with winning streaks of seven to 10 games have gone 1,013-869-2 for a winning percentage of .538. The first-game-back results certainly run counter to that, although we saw the Mets succeed last night where most of their predecessors had failed.
One theory that bears at least a cursory glance is that, since teams can, in many cases, restart their pitching rotations when they come back, opponents are bound to run into an ace on their first day back from recess. Could this negatively impact a streaking team's chances of continuing their winning ways? A problem with this theory is that the streaking team can also load its ace back to the top on the first day back. Another is that the opponent's ace might have pitched in the All-Star Game, and needs another day or two before starting again.
With this in mind, I looked at the win-streak teams that lost on their first day back to see who they faced. The '75 Reds were blanked for six innings by Steve Rogers, the Expos' best pitcher that year, and didn't score off of reliever Dale Murray, either. The '95 Braves, on the other hand, were met by Steve Parris, the Pirates' 27-year-old rookie, who outdueled John Smoltz. The '68 Red Sox were beaten by George Brunet, a league-average pitcher who had not really landed a full-time big-league gig until the age of 30. The '74 Pirates were beaten by the Expos' bats more than starter Steve Renko, as Montreal chased Jim Rooker early. The '68 Cardinals met the very talented Larry Dierker, but played him to a tie before losing in extra innings (and then winning their next six games). This is looking like a blind alley.
On the other side of the ledger, the Rays head into tonight retracing the steps that 13 other teams have taken since 1957: coming out of the break riding a losing streak of seven or more games. The longest such streak was endured by the 2004 Seattle Mariners, who ended their nine-game slide immediately upon their return. The other pre-break sliders are:
Teams that lose seven games in a row tend to skew south of the quality line and this group is no exception. Collectively, they boast a .426 winning percentage and include only three .500 teams; the '03 Twins, '67 Cubs and '00 Rockies. Both the '61 Phils and '76 Expos lost 107 games (the Phillies on a 154-game schedule) and the '89 Tigers and '62 Senators also lost over 100. The '04 Mariners lost 99 and the '01 Royals 97. This group combined to go 5-8 in the first game back. Over the same period, the record for all other teams who were on losing streaks of seven to nine games is 802-978, a winning percentage of .451.
Based on the tiny bit of evidence available, how did that announcer's contention hold up? In one case fairly well, and in the other not at all. The truth is that there are just never going to be enough examples to create a database worthy of producing tangible results to test the layoff theory. Well, I shouldn't say that. One hundred years from now, each of these lists will have more than tripled in size and, by then, we will have also added the 1933-1956 period to the database. We can revisit this topic then. I 'll put it at the top of the to-do list I plan on bringing with me into the cryogenic chamber.
Thanks to Bil Burke for his research.