This is our old blog. It hasn't been active since 2011. Please see the link above for our current blog or click the logo above to see all of the great data and content on this site.

More Random Division Series Notes

Posted by Neil Paine on October 6, 2010

Tacking on to Andy's playoff thoughts from earlier today...

Familiarity Breeds Contempt

The 2010 Yankees-Twins ALDS represents just the 8th time in the WS era (1903-2010) that the same two teams faced each other in a postseason series 4 times in 8 seasons... The Dodgers & Yankees played each other in 4 world series from 1946-53, 48-55, and five times (!!!) from 1949-56. The Yankees and Royals faced off in 4 series from 1973-80, and the Braves & Astros had two different 8-year stretches with 4 playoff showdowns in the late nineties/early 2000s (1997-04 and 1998-05). The Red Sox and Angels capped off a 4-series-in-8-years stretch last season, and now the Yankees and Twins have joined that group with ALDS faceoffs in 2003, 2004, 2009, and 2010.

Snapping Playoff Droughts

For those who complain about parity and the same teams always being in the playoffs, this is the 3rd time in the last 4 years that two playoff teams are snapping a 10+ year playoff drought in the same season (2010 Cincy - 14 yrs/2010 Texas - 10 yrs; 2008 Milwaukee - 25 yrs/2008 Tampa - 10 yrs; 2007 Philly - 13 yrs/2007 Colorado - 11 yrs). That hasn't happened since 1984 (DET/CHC/SDP), '86 (NYM/BOS), & '87 (MIN/SFG).

("Droughts" in this case being defined as consecutive seasons in which the team missed the playoffs.)

Wide-Open Race

As Andy noted, only 7 wins separate the winningest and losingest playoff teams. Since the Wild Card era began, that's the smallest spread between the best and worst playoff teams' records:

Year maxW minW Spread
1995 100 77 23
1996 99 88 11
1997 101 84 17
1998 114 88 26
1999 103 94 9
2000 97 87 10
2001 116 88 28
2002 103 94 9
2003 101 88 13
2004 105 92 13
2005 100 82 18
2006 97 83 14
2007 96 85 11
2008 100 84 16
2009 103 87 16
2010 97 90 7

You're Not Only As Good As Your #1 Starter

Most baseball fans would think the team with the superior #1 starter has the leg up in a playoff matchup, especially in a short series. However, it turns out that knowing which team has the better #1 starter (by regular-season WAR) tells you almost nothing about which team will go on to win the Division Series:

Year Round Winner Game1 SP WAR Loser Game1 SP WAR
1995 ALDS CLE Dennis Martinez 5.3 BOS Roger Clemens 1.8
1995 ALDS SEA Chris Bosio 1.8 NYY David Cone 6.2
1995 NLDS ATL Greg Maddux 8.8 COL Kevin Ritz 3.9
1995 NLDS CIN Pete Schourek 3.7 LAD Ramon Martinez 2.5
1996 ALDS BAL David Wells 3.3 CLE Charles Nagy 6.8
1996 ALDS NYY David Cone 2.6 TEX John Burkett 1.8
1996 NLDS ATL John Smoltz 6.1 LAD Ramon Martinez 2.4
1996 NLDS STL Todd Stottlemyre 2.6 SDP Joey Hamilton 2.6
1997 ALDS CLE Orel Hershiser 2.3 NYY David Cone 6.7
1997 ALDS BAL Mike Mussina 5.1 SEA Randy Johnson 7.7
1997 NLDS FLA Kevin Brown 6.4 SFG Kirk Rueter 2.6
1997 NLDS ATL Greg Maddux 7.3 HOU Darryl Kile 5.4
1998 ALDS CLE Jaret Wright 1.8 BOS Pedro Martinez 6.6
1998 ALDS NYY David Wells 4.3 TEX Todd Stottlemyre 3.0
1998 NLDS ATL John Smoltz 3.2 CHC Mark Clark 1.8
1998 NLDS SDP Kevin Brown 8.4 HOU Randy Johnson 5.5
1999 ALDS BOS Pedro Martinez 8.4 CLE Bartolo Colon 3.8
1999 ALDS NYY Orlando Hernandez 3.9 TEX Aaron Sele 3.2
1999 NLDS ATL Greg Maddux 2.9 HOU Shane Reynolds 3.2
1999 NLDS NYM Masato Yoshii 2.0 ARI Randy Johnson 8.2
2000 ALDS NYY Roger Clemens 4.7 OAK Gil Heredia 3.2
2000 ALDS SEA Freddy Garcia 1.8 CHW Jim Parque 2.6
2000 NLDS STL Rick Ankiel 2.8 ATL Greg Maddux 6.1
2000 NLDS NYM Mike Hampton 4.6 SFG Livan Hernandez 3.1
2001 ALDS SEA Freddy Garcia 4.0 CLE Bartolo Colon 4.4
2001 ALDS NYY Roger Clemens 5.4 OAK Mark Mulder 5.2
2001 NLDS ATL Greg Maddux 4.5 HOU Wade Miller 4.9
2001 NLDS ARI Curt Schilling 7.3 STL Matt Morris 3.7
2002 ALDS ANA Jarrod Washburn 4.3 NYY Roger Clemens 2.3
2002 ALDS MIN Brad Radke 0.3 OAK Tim Hudson 6.6
2002 NLDS SFG Russ Ortiz 2.7 ATL Tom Glavine 3.7
2002 NLDS STL Matt Morris 2.7 ARI Randy Johnson 8.8
2003 ALDS NYY Mike Mussina 6.2 MIN Johan Santana 4.0
2003 ALDS BOS Pedro Martinez 7.4 OAK Tim Hudson 6.7
2003 NLDS FLA Josh Beckett 3.4 SFG Jason Schmidt 5.9
2003 NLDS CHC Kerry Wood 5.3 ATL Russ Ortiz 2.2
2004 ALDS BOS Curt Schilling 6.4 ANA Jarrod Washburn 1.6
2004 ALDS NYY Mike Mussina 2.1 MIN Johan Santana 7.4
2004 NLDS STL Woody Williams 1.3 LAD Odalis Perez 3.4
2004 NLDS HOU Roger Clemens 5.2 ATL Jaret Wright 3.0
2005 ALDS CHW Jose Contreras 3.9 BOS Matt Clement 3.1
2005 ALDS LAA Bartolo Colon 4.4 NYY Mike Mussina 2.6
2005 NLDS STL Chris Carpenter 4.8 SDP Jake Peavy 4.5
2005 NLDS HOU Andy Pettitte 5.8 ATL Tim Hudson 2.9
2006 ALDS DET Nate Robertson 3.4 NYY Chien-Ming Wang 5.4
2006 ALDS OAK Barry Zito 4.7 MIN Johan Santana 7.0
2006 NLDS NYM John Maine 1.0 LAD Derek Lowe 4.7
2006 NLDS STL Chris Carpenter 5.0 SDP Jake Peavy 1.7
2007 ALDS BOS Josh Beckett 4.7 LAA John Lackey 6.3
2007 ALDS CLE CC Sabathia 6.8 NYY Chien-Ming Wang 3.9
2007 NLDS ARI Brandon Webb 6.1 CHC Carlos Zambrano 2.9
2007 NLDS COL Jeff Francis 2.2 PHI Cole Hamels 4.1
2008 ALDS BOS Jon Lester 5.6 LAA John Lackey 3.6
2008 ALDS TBR James Shields 3.6 CHW Javier Vazquez 3.1
2008 NLDS LAD Derek Lowe 3.6 CHC Ryan Dempster 5.3
2008 NLDS PHI Cole Hamels 4.4 MIL Yovani Gallardo 1.1
2009 ALDS NYY CC Sabathia 4.3 MIN Brian Duensing 1.3
2009 ALDS LAA John Lackey 2.9 BOS Jon Lester 5.6
2009 NLDS LAD Randy Wolf 3.3 STL Chris Carpenter 5.9
2009 NLDS PHI Cliff Lee 5.0 COL Ubaldo Jimenez 5.1

In 60 Division Series matchups, knowing which team's Game 1 starter had the better regular-season by WAR would tell you the correct series winner 33 times (55%). If you picked the winning team at random, you'd have a 50% chance of choosing the correct team, so knowing which team had the better Game 1 starter tells you very little about who the eventual series victor will be.

21 Responses to “More Random Division Series Notes”

  1. Johnny Twisto Says:

    I looked at the times when the G1 SP was at least 2.0 WAR better than his opponent, so we could be relatively sure he really did have a better season. No guarantees on my counting correctly, but it looks like the team with the clearly superior G1 pitcher still only won 19 of 36 (53%).

  2. The Goof Says:

    Actually, Yanks and Dodgers faced in 1941, 1947, 1949, 1952, 1953, 1955, 1956, 1963, 1977, 1978 and 1981. Hence, four times from 1952-1956, not five. Yanks and Giants played in the Series three years in a row and five times in 17 years. Knowing this off the top of my head makes me a nerd of the first order.

  3. The Goof Says:

    Four times from 1950-1956 is what I meant, but same thing, really.

  4. The Goof Says:

    Starting Randy Johnson in Game 1? Kiss of death! 0-4

  5. Neil Paine Says:

    Good catch, Goof. I can't figure out why my query returned 5 in the 7-year stretch from 50-56, but you're definitely right, they "only" played 4 times there.

  6. Neil Paine Says:

    Randy Johnson lost Game 1s when he had 8.8-to-2.7 and 8.2-to-2.0 WAR advantages. Yikes!

  7. BSK Says:

    Were there any teams that did not have their best pitcher start game 1? Sometimes this could be the result of reputation, but I'm sure there are other times where teams are using their top pitchers down to the final games and may have to push back their start. How would the list look if we compared best pitcher to best pitcher?

  8. TheGoof Says:

    You know the playoffs this year should be great when even That 70s Announcer (Craig Sager) looks sharp.

  9. TheGoof Says:

    I take it back; I saw Sager from afar that first time. My goodness, that's a scary getup.

  10. Ryan Says:

    Did you use the play index to get the list of #1 pitchers in LDS match-ups? And if so how?

    It's really interesting.

  11. Yanks/Twins – We’ve Seen This Before | Says:

    [...] stuff via Neil Paine: The 2010 Yankees-Twins ALDS represents just the 8th time in the WS era (1903-2010) that the same [...]

  12. Neil Paine Says:

    Re: #10 - I did. You can use the game finder to come up with a list of starting pitchers by round and game # of series:

    Seasonal WAR was more difficult, and involved converting the names on that list to Retrosheet IDs and cross-referencing a WAR file (not really something you can do readily on the site).

  13. John Autin Says:

    Jesse Crain just gave up a 2-run, tie-breaking HR to Teixeira in the 7th.

    That's 4 postseason appearances by Crain, none of them good.
    -- 2004 ALDS game 3: Entered with a man on 2nd and 2 out, allowed a run-scoring single.
    -- 2006 ALDS game 1: Began the 9th trailing 2-1, gave up a HR to the first batter.
    -- 2006 ALDS game 3: Entered in the 7th trailing 4-2, 2 out, 2 on; first batter reached on error, then he walked in a run and allowed a 3-run double.
    -- 2010 ALDS game 1: Began the 7th in a 4-4 tie, got an out, then allowed single-HR-single and was pulled.

  14. Ryan Says:

    Thanks Neil!

    I am interested if the same percentage holds over a series first 2 games. We hear so often about how "two good starters" are all you need to win in a short series.

    I'm interested if history supports this oft repeated refrain.

    Someone else mentioned this, but maybe starters like Brian Duensing in last year's LDS, should not count towards the total. Meaning guys who weren't number 1 or 2 starters, but had to start those games because the #1 or #2 guys were not available.

  15. John Autin Says:

    It seems ironic that Joe Girardi is now in charge of deciding how to balance Jorge Posada's potent bat against his well-known defensive shortcomings.

    Posada started tonight in ALDS game 1. He has let in a run on a passed ball, but also has hit 2 for 4 with a ribbie and a run. I started thinking about this while noting that Posada was still behind the plate in the bottom of the 7th, after the Yankees took a 6-4 lead; Posada was due to lead off the 8th for NYY.

    Joe Girardi played for the Yankees from 1997-99. He was the #1 catcher in '97, with Posada the backup, but Posada took over the starting role in '98.

    In 5 postseason games in '97, #1 catcher Girardi got all 5 starts.
    In 25 postseason games in '98-99, #1 catcher Posada started just 11 times, while #2 Girardi started 14 times.

    Whether those decisions by Joe Torre were "right" at the time, the fact is that Posada has hit much worse in the postseason than in the regular season -- a .736 OPS in 441 postseason PAs, vs. .856 OPS in the regular season.

    Now Girardi is in charge. Last year, Posada got the lion's share of postseason playing time over Jose Molina -- Posada started 10 of the 15 games, and got 58 PAs to Molina's 7.

    This year, the Yankees' backup catcher is Francisco Cervelli. (Technically, he caught a little more than Posada, but I stick by my ranking.) Cervelli is about equal to Girardi as a hitter: decent BA, no power. But superficially, at least, he's no great improvement over Posada behind the plate, with 55 SB/14% CS, and 37 (WP+PB) in 80 starts and 724 defensive innings; Posada allowed 72 SB/15% CS, with 40 (WP+PB) in 78 starts and 678 defensive innings.

    So maybe that's why Posada is still catching as we go to the bottom of the 9th.

  16. John Autin Says:

    (Upon seeing Joe Mauer ground out to Mariano Rivera: )

    Has there ever been a pitcher who induced so many weak groundouts to the pitcher?
    Rivera has assisted on almost 10% of the non-strikeout outs in his career.
    By comparison, contemporary closer Trevor Hoffman has assisted on less than 4% of his non-strikeout outs.

    Ah, but Rivera's GB/FB ratio is about twice Hoffman's, so maybe Rivera's assists just reflect his general GB tendencies.

    Since the Play Index doesn't cover fielding, nor GB/FB ratios, I'll take just one more shot with a pitcher I know to have a very high GB/FB ratio, even though he's not a reliever: Brandon Webb.

    Webb has a much higher GB/FB ratio than Rivera, 1.82 to 1.03.
    Still, Webb has assisted on a lower percentage of his non-strikeout outs, 9.3% to 9.7%.

    Incredibly unscientific as that "study" was, I now feel more certain of my initial assumption.

    P.S. Does anyone know of an efficient way to check this kind of thing?

  17. Larry R. Says:

    And this, my friends, is why they play the games. Take all the stats on any given day and throw them out the window. Who's going to win a ballgame, any ballgame...flip a coin.

  18. Johnny Twisto Says:

    John, Molina only got those starts last season because the team had decided to make him Burnett's personal catcher about halfway through the season. There was some controversy about whether that would continue in the postseason. I'd bet that Posada started every postseason game from 2000 to 2008, unless he was hurt. I am not sure what the plans are this year but I would not be surprised if Posada starts all the games at catcher. Cervelli is superior defensively, but that isn't difficult. From what I've seen, I think a lot of the problems in throwing out basestealers this year is on the pitchers. Posada's one slightly positive defensive skill throughout his career has been his arm -- when not recovering from a couple shoulder surgeries, he has generally thrown pretty well with a good release. He may not be throwing as well now as he once did but the poor % is not all his fault.

  19. Keeping Score: Parallel Paths of Halladay and Lee - Says:

    [...] Game 1 of the Yankees-Twins division series marked the fourth time the two teams have faced each other in the last eight postseasons, only the eighth such instance in the World Series era. This decade alone, they joined the Red Sox-Angels (four matchups from 2004 to 2009) and Braves-Astros (five matchups from 1997 to 2005). You can read about the other familiar playoff matchups at the blog. [...]

  20. Mike Says:

    This is just a list of who started game 1, not the team leader in WAR for pitchers. If a team had to clinch at the end of the regular season, like Milwaukee in 2008, they used their ace (Sabathia, 4.8 WAR in 130 IP for Mil) to clinch so he didnt start game 1 of the playoffs. Also Yovanni Gallardo accumulated his 1.1 WAR in only 24 IP that season.

  21. Great 1-2-3 Rotation Combos In a Best-of-7 Series » Baseball-Reference Blog » Blog Archive Says:

    [...] week I posted about how often the team with the best Game 1 starter by WAR wins the Division Series (only 55%, as [...]