You Are Here > Baseball-Reference.com > Blog >

SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all B-R content, so you can quickly and easily find the content you want.

Also, our existing B-R blog rss feed will be redirected to the new site's feed.

Baseball-Reference.com ยป Sports Reference

For more from Andy and the gang, check out their new site High Heat Stats.

Three 5-WAR pitchers on one team … but will they win it all?

Posted by John Autin on September 6, 2011

Barring a collapse by one of the Big Three, the Phillies will become the 17th team since 1893 to have 3 pitchers with at least 5 Wins Above Replacement (using B-R's WAR formula).

If you believe that pitching is the main key to championships, this list might give you pause; only 2 teams with this kind of "big three" have won the World Series, and none after 1912.

Rk Year Lg Tm Standings Pitchers
1 2011 NL Philadelphia Phillies 1st in Div. Roy Halladay / Cole Hamels / Cliff Lee
2 2005 NL Houston Astros Pennant Roger Clemens / Roy Oswalt / Andy Pettitte
3 1996 NL Atlanta Braves Pennant Tom Glavine / Greg Maddux / John Smoltz
4 1974 NL New York Mets 5th, 71-91 Jerry Koosman / Jon Matlack / Tom Seaver
5 1973 AL Chicago White Sox 5th, 77-85 Stan Bahnsen / Terry Forster / Wilbur Wood
6 1973 AL Detroit Tigers 3rd, 85-77 Joe Coleman / John Hiller / Mickey Lolich
7 1970 NL Chicago Cubs 2nd, 84-78 Bill Hands / Ken Holtzman / Fergie Jenkins
8 1968 NL San Francisco Giants 2nd, 88-74 Bobby Bolin / Juan Marichal / Gaylord Perry
9 1960 NL St. Louis Cardinals 3rd, 86-68 Ernie Broglio / Larry Jackson / Lindy McDaniel
10 1956 AL Cleveland Indians 2nd, 88-66 Bob Lemon / Herb Score / Early Wynn
11 1925 NL Cincinnati Reds 3rd, 80-73 Pete Donohue / Dolf Luque / Eppa Rixey
12 1913 AL Chicago White Sox 5th, 78-74 Eddie Cicotte / Reb Russell / Jim Scott
13 1913 NL New York Giants Pennant Rube Marquard / Christy Mathewson / Jeff Tesreau
14 1912 AL Boston Red Sox W.S. CHAMP Ray Collins / Buck O'Brien / Smoky Joe Wood
15 1912 NL New York Giants Pennant Rube Marquard / Christy Mathewson / Jeff Tesreau
16 1909 NL Chicago Cubs 2nd, 104-49 Mordecai Brown / Orval Overall / Ed Reulbach
17 1907 NL Chicago Cubs W.S. CHAMP Mordecai Brown / Carl Lundgren / Orval Overall
Provided by Baseball-Reference.com: View Play Index Tool Used
Generated 9/5/2011.

But no, I'm not suggesting this has any bearing on the 2011 Phillies.

Three of these 17 teams had a reliever among their big three: the 1973 White Sox (Terry Forster, 173 IP including 12 starts) and Tigers (John Hiller, 125 IP, 1.44 ERA, record 38 saves, 10-5 record), and the 1960 Cardinals (Lindy McDaniel). None of those teams were serious contenders.

Now, the Phils' big three already have at least 5.3 WAR apiece, so they may all wind up with 6 WAR. Which of the teams above would meet a higher threshold?

Interestingly, though perhaps coincidentally, the teams' fortunes fade as the WAR threshold is raised. Teams with 3 pitchers at 5.5+ WAR won 1 pennant and no WS -- the '96 Braves, '25 Reds, '13 White Sox, '09 Cubs. And the only other team with 3 pitchers at 6 WAR was the 1913 White Sox, who finished 5th, just above .500.

Again, I'm not trying to make a point about the 2011 Phillies. I just think the table is one piece of evidence contrary to the popular belief that a team with 3 excellent starters has a great shot to win it all. However, it could also be said that, in the wild-card era, only 2 teams have made the tourney with three 5-WAR pitchers, and both won the first 2 series.

One last point: None of the greatest records in history show up on this list -- no 2001 Mariners, no 1998 Yankees, no 1954 Indians, no 1906 Cubs, etc. Of the 18 teams that won at least 106 regular-season games since 1901, only the 1907 Cubs (107-45) made the "big three" list.

Your thoughts?

This entry was posted on Tuesday, September 6th, 2011 at 12:34 am and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

66 Responses to “Three 5-WAR pitchers on one team … but will they win it all?”

  1. The 1998 Yankees (in addition to an offense that scored 965 runs) had about 16 WAR out of their rotation (David Wells, David Cone, Andy Pettitte, Hideki Irabu, and a combination of El Duque and Ramiro Mendoza).

    That's the same as having 3 5-WAR guys and a conglomerate of scrubs composing the other two spots (much like the 2005 Astros).

  2. @1, John -- Well, no, it really isn't the same, especially if you're thinking about a postseason series.

  3. I think the argument about having three big starters is primarily about how they'll help a team win a playoff series -- not that they'll necessarily lead to regular season success. A team that makes the playoffs is likely to be at least solid all-around, and having those three big starters in a short series might give them an edge.

    The point is to say that you'd really want to see how teams with those three 5+ WAR starters did in the playoffs. Looks like their series record is 6-4, which, of course, is much too small a sample size to make any conclusions. Which makes this a pretty boring comment!

  4. Eh. The Phillies already have too many positive indicators going against them--namely, the team that finishes with the best regular season record almost never wins the World Series. I think they'll make the World Series but lose to the Yankees. Something tells me the Yankees offense will trump the Phils' pitching, and retreads like Burnett, Colon and Garcia will be maddening just enough to get the Yanks to the seventh inning with a lead, and then it will be over. Yanks' bullpen is stronger. (Gosh, I hope I'm wrong, but that's the way I see it.)

  5. "Something tells me the Yankees offense will trump the Phils' pitching"

    That worked well against Lee in 2009

  6. And last season for that matter

  7. So the last two teams to win the World Series with three 5-WAR pitchers on one team were the Boston Red Sox and the Chicago Cubs in the early 1900s?

    Memo to Philly fans: PRAY YOU LOSE THIS YEAR!! If not, you're heading toward nearly 100 years of painful losing before your next championship.

  8. You need to factor in the pitcher's performance with respect to his own team. Team performance affects the winning record of the pitchers.

    Team wins minus losses followed by wins minus losses for the pitchers
    The 3 pitchers account for a majority of the team differential
    58 (41)
    55 (38)
    55 (36)
    50 (36)
    50 (31)
    40 (25) 2011 Phillies
    38 is a 100 win season
    30 (25)
    BELOW THIS LINE THE REST OF THE PITCHERS ON THE TEAM ARE BELOW 0.500
    22 (28)
    18 is a 90 win season
    18 (25)
    16 (21)
    14 (27)
    8 (14)
    7 (15)
    6 (15)
    4 (12)
    -8 (-4) but the 3 recorded 48 of the 77 team wins.
    -20 (2)

  9. I think just having 3 great pitchers is not a sign of having a great team around them, as evidenced by the fact that most of these teams missed the playoffs, which the Phillies will not. It's the same reason why the Dodgers aren't good despite having the WAR leader and possibly the Cy Young award winner. Adding 15 wins above replacement still only gets you to 60 wins. The other 30 have to come from somewhere else.

  10. Atlanta's 3 pitchers pitched great in the WS in 1996. They were 2-3 in 5 games despite giving up only 5 ERs. Atlanta outscored the Yankees 26-18. They were always ahead or within 1 after 7 innings except game 6 when they were behind 3-1.

    Houston's 3 pitchers did not do well, but the games were close The White Sox scored 3 on Clemens in 2 innings to won 5-3. They scored 2 on Pettite in 6 IP to win 7-6. They scored 5 on Oswalt in 7 innings to win 7-5 in 14 innings. Backe pitched game 4 with 7 SHO innings and the Astros lost 1-0. All four losses went to relievers. Two 1 run and two 2 run games.

  11. @9

    100% correct, but if you're good enough to make the playoffs, it helps because teams often go to a 3 man rotation.

    Here's the team win differential and the the combined differential for Cicotte, Russell and Scott starting in 1913.

    1913 +4 (+12) 5th (rest of team -8 wins) All three have WAR above 5
    1914 -14, (-16) 6th (rest of team +2 wins)
    1915 +32, (+15) 3rd (rest of team +17 wins)
    1916 +24, (+8) 2nd (rest of team +16 wins)
    1917 +46, (+25) (rest of team +21 wins) Won World Series

    I saw someting interesting between the 1913 and 1915 seasons
    In parenthesis position in league
    1913 ERA 2.33 (1) Runs Allowed 3.22 (1) Runs Scored 3.18 (8) -0.04
    1914 ERA 2.48 (2) Runs Allowed 3.62 (5) Runs Scored 3.10 (8) -0.52
    1915 ERA 2.43 (3) Runs Allowed 3.28 (3) Runs Scored 4.63 (1) 1.35
    1916 ERA 2.36 (1) Runs Allowed 3.21 (2) Runs Scored 3.88 (3) 0.67
    1917 ERA 2.16 (1) Runs Allowed 2.96 (2) Runs Scored 4.21 (1) 1.25

    So you can see how the fluctuations in these from year to year track the change in win differential.

  12. @8, Charles -- I don't understand your point here, particularly "Team performance affects the winning record of the pitchers.". I'm sure you know that actual wins and losses play no role in Wins Above Replacement, but I'm still at a loss to grasp your point. Can you clarify? Thanks.

  13. Good writing. I was reading this thinking that it would be harder to get 5 WAR in 32 starts than in 40 so this would be a stronger group of pitchers than those achieved by 4 man rotations. Then the part about higher WAR thresholds having poorer results punctured the argument before it got worded.

  14. @12, JA, maybe what Charles is looking at is the winning record of the team as a whole-which, while not impacting WAR, does affect where they finish in the standings. One thrust of the article seems to be that having 3 or more 5+WAR pitchers isn't a guarantee of playoff success (or even playoff appearance). So it's reasonable to look at overall team performance as expressed in their win loss record.

  15. Stan Bahnsen? Was that the worst 5 WAR season for a pitcher in the last fifty years? It seems it was accomplished more because he started more than a quarter of his team's games than for any real skill at pitching.

  16. The problem with this argument is teams 4-17 on this list featured pitchers with IP totals that are impossible to reach today (275+ for starters, 100+ for relievers), and since IP is a major component of WAR, this skews the results.

    Look at it this way - the 1970 Cubs trio pitched 60% of the team's innings and contributed 18.2 WAR. On the other hand, the 2011 Phillies trio pitched 47% of the innings with 17.7 WAR, the 2005 Astros trio pitched 47% of the innings with 18.3 WAR, and the 1996 Braves trio pitched 50% of the innings with 18 WAR.

  17. @14 You got it Mike.

    Forget why they can't win the WS, my question became, why can't they do better in the regular season?

    When I said the team affects the winning record, I only meant the W-L record. I don't know enough about WAR to get into a discussion about that. Any parameter that I can't see in a box score is less interesting to me.

    I'm showing that 3 pitchers with high WAR will have better won-loss records if they play for a better team. 3 pitchers with a high WAR will not get you into the playoffs unless the other players can show they can win. I took the 17 teams and rank ordered then by wins minus losses then I looked at the personal wins and losses of those 3 pitchers which is a parameter influenced by the hitters on the team. Sure enough, as the team win differential goes up, so does the ovarall winning record of the 3 pitchers. If you look down the list, you'll see the right side number becoming bigger than the left. That means the rest of the team pitchers have a losing record. You can make the playoffs with less than 90 wins, but if one of those 3 guys is injured, you could be in trouble. The NYM had 20 more losses than wins, the 3 WAR pitchers combined won only 2 more games than they lost. The question was why can't you succeed with 3 high WAR players? I showed they could dominate the team won-loss record, no matter how good or bad the team is, but a poor team isn't going to give all 3 pitchers a good record.

    In post 11, I showed a team that went nowhere until they got some good hitters. I showed that the team and the same 3 pitchers improved their W-L record as the hitting improved, even with a fairly constant ERA. This agrees with the trend in post 8 with a single team, but taking a closer look at run scoring to identify how the team got better.

    Hope that clears it up..

  18. And in post 10, I gave an example where the high WAR starters pitched well, but the offense and or relievers couldn't get them the wins.

    The 2nd WS example we have the offense overcoming sub-par performances by the starters, but the relievers couldn't keep them in the game.

  19. Lawrence Azrin Says:

    I was surprised to see only one year of the Braves great trio of Maddux/ Glavine/ Smolz on this list (1996), so I looked up what years all of them were together: 1993 to 2002.

    However, Smoltz missed all of 2000, and was the closer in 2001 and 2002, so they were all starting for the Braves from only 1993 to 1999. Still, it's surprising that they "synced up" only the one year. I was also surprised not to see any of the great pitching staffs of the Baltimore Orioles from 1969-1972 with Palmer, McNally and Cuellar.

    I think in the postseason, having two truly dominant pitchers (like the 2001 Diamondbacks, with Randy Johnson and Schilling), is better than having three excellent pitchers.

    @15/ Howard Rosen - Stan Bahsen's 113 ERA+ in 282 innings in 1973 certainly shows "real skill at pitching" to me. I _do_ question how he could be 9th in pitcher WAR (5.9) for the 1973 AL, without being in the Top-10 in ERA+, innings, or WHIP. Maybe he was in a hitters park, or had bad defenses behind him?

  20. Charles, thanks for the follow-up.

    I'm puzzled by your stance on "any parameter that I can't see in a box score." Does that mean that you view a shutout pitched against the 1908 Cardinals (who averaged 2.42 R/G) as equivalent to one pitched against the 1995 Rockies in Coors Field (8.12 R/G)?

  21. Lawrence @19 -- Yep, you might say the '73 White Sox had bad defense -- combined defensive WAR of -8.1. That was pretty typical of the White Sox in the mid-'70s.

  22. That '74 Mets team is such an odd team. Matlack actually had a losing record that year and Seaver was .500. They probably had the worst fielding outfield I ever saw with Hahn, Staub, Theodore, Jones and Schneck. Millan was done as a good fielder. I think Cleon Jones was the only player who batted above .270 with a .284 avg.

    I never really understood why they had players like Schneck, T. Martinez and Don Hahn because they couldn't field and they couldn't hit.

  23. @20

    If I feel the need to look at R/G I will. I understand what WAR tells us. I meant I have no interest in learning how to calculate it.

  24. Well, to be fair, one of those turn of the century teams was the 1907 Cubs. And we all know the long-term World Series dominance that has resulted since.

    /cards fan.

  25. surprised the A's of hudson, zito, mulder, lilly aren't on here

  26. Interesting there were two AL teams to do this in the first year of the DH, the only time in the live ball era two teams did it in the same year.

    Seems like the initial reaction by AL managers to the DH rule was "Thank goodness I don't have to think about pinch-hitting for my pitcher in a tight game. Now I can just leave him out there as long as I want."

    An indication of this is this list of the number of AL pitchers reaching 250 IP.
    - 1969-72: 10, 13, 17, 17
    - 1973-76: 22, 23, 16, 13

    Certainly, a perceptable spike in the first two years of the DH.

  27. It's usually the most balanced teams that have the best chances at winning the WS. When we say 'great pitching,' what does that mean? Does it mean that they are great because their earned run average is lower than league average? Perhaps that they win most of their decisions? Does it mean that they have dominating stuff that correlates to high strikeout rates?

    I think most will agree that Greg Maddux was one of the best pitchers of his era and all time, however, his strikeout rate over his career merely hovered around 6. If we look at his postseason numbers, he has a losing record too. Perhaps it's not just as simple as having great pitchers, but great strikeout pitchers, good defense, etc.

    Take the Giants from last year; the Phillies [arguably] had the more acclaimed starters and they had even better seasonal results. However, the Giants had better defense, a decent lineup, shutdown bullpen, pitching depth, and a little bit of luck.

    There's a lot more than pitching that goes into a WS berth.

  28. As a control group to Doug's data @26, here are the number of National League pitchers reaching 250 IP in those same seasons:
    -- 1969-72: 14, 14, 16, 10 = total of 54
    -- 1973-76: 10, 11, 10, 10 = total of 41

  29. From 2001 to 2010 there were 80 playoff teams. There were 26 total teams with at least 3 pitchers (60% of their games as starts, qualified for ERA title) with a WAR greater than or equal to 3. 19 of those teams made the playoffs. Two won World Series titles, 4 lost WS titles. Five teams had at least 3 pitchers in 2005 (Astros, Nationals, Red Sox, Angels, White Sox). The White Sox have the longest current streak from 2005-2010. There was at least 1 team every year.

  30. i am fortunate to have at the core of my baseball knowledge the idea that winning is a TEAM effort, even in the realm of pitching. Pitching may be essential, but it is better to have more good pitchers than a handful of dominant ones? where did I get this core idea? because I grew up as a fan of the big red machine...and I remember the attention the media gave to the team that they didn't have single stand out pitchers.

    1975 they had six pitchers with 10 wins, yet none with more than 15 wins. Yet they had the 3rd best ERA in the NL and 1st in Saves (with a left-handed and a right-handed closer). The highest WAR for their pitchers that year was 0.2 (Borbon and Gullet). It didn't hurt that they had balanced offense, an entire line up of 3+ WAR.

    1976 was more of the same. Seven pitchers with 10 wins, yet none with more that 15. 5th in ERA and 1st in saves. And the two top pitchers at 0.1 WAR (Billingham and McEnaney).

    Yes, yes, an amazing offense, but part of the MACHINE was this well-rounded pitching staff that had no future hall of famers, but was just pure goodness from top to bottom.

  31. @29, Charles -- I'm glad you're trying to advance the discussion. However, I think a 3-WAR threshold is much too low to be of use here.

    Take those 2005 Red Sox: The 3 starters with at least 3 WAR were Tim Wakefield (4.3), David Wells (3.2) and Matt Clement (3.1).

    Their ERA+ and IP:
    -- Wakefield, 109 ERA+, 225 IP
    -- Wells, 102 ERA+, 184 IP
    -- Clement, 99 ERA+, 191 IP

    If we're talking about the adage that "3 good starters gives you a good shot in the postseason," do those 3 starters meet anyone's working definition of "3 good starters"?

  32. Followup to #31 -- And in fact, those 2005 BoSox were crushed by the ChiSox in the Division Series, getting just 1 quality start in 3 games by those 3 pitchers.

  33. Johnny Twisto Says:

    Seems like the initial reaction by AL managers to the DH rule was "Thank goodness I don't have to think about pinch-hitting for my pitcher in a tight game. Now I can just leave him out there as long as I want."

    IP per start jumped from 6.3 to 6.6 in 1971, two years before the DH, and stayed right there through 1976. So it looks like managers were probably more willing to let their top SP keep pitching, but the overall length of starts was unchanged. (The lesser guys were probably getting pulled quicker, with the increased offense.)

    ***

    The highest WAR for their pitchers [in 1975] was 0.2 (Borbon and Gullet). 1976 was more of the same. ....the two top pitchers at 0.1 WAR (Billingham and McEnaney).

    I was sure this had to be a typo or a mistake. I think you were looking at the batting WAR tables only. Look below them. Gullett in '75 had 3.2 WAR pitching, and Pat Zachry had 3.5 in '76. Those teams weren't known for their pitching, but there's no way they'd have been successful if they had a replacement level pitching staff.

    There has never been a team since 1900 without a pitcher who exceeded 0.2 WAR.

  34. mr twisto, thank you for your help in my navigation of these charts!

  35. I think a better way to approach this rather than looking at the regular season up would be to look at the WS down. What is the average regular season WAR of the top 3 starters for a WS winning team? Is it really that high or is starting pitching not the WS winning element we think it is?

    My thoughts are that WS winning rotations are good but not often the best in baseball. You need good starting pitching, but you don't need excellent starting pitching. I'd bet the average war would be closer to 8, not 15.

  36. [...] 6th, 2011 Recap: Today, John Autin notes that Philly’s staff has 3 5-WAR pitchers (but what will that mean in October?)… Steve Treder gives the 1960s Indians a re-do… BP’s Michael Jong looks at Javier [...]

  37. @31 Agreed

    So the conclusion is: No significant correlation can be found in this decade.

    WAR 3.0 gives a reasonable number of teams to try to attempt to make some sense out of this. 19/26 (73%) of those teams make the playoffs.
    If 3.0 is considered too low, then I would say there's nothing else to pursue.

    Raising the bar won't help because it will make the sample size even smaller than 19 and lower the possibily of pulling out a WS participant, already 6 of 20.

    We start with 300 teams.
    135 are eliminated since they don't have 3 qualifed starters.
    Eliminate 139 because they have no more than 2 starters at 3.0 WAR. That leaves 26, 19 (73%) make the playoffs.
    The other 61 come somewhere in the pool of 274.

    This is funny.
    2005
    Astros, Nationals (81-81), Red Sox, Angels, White Sox met the criteria
    Astros, Red Sox, Angels, White Sox made the playoffs
    White Sox eliminated the Red Sox, Angels and Astros advance
    White Sox beat Angels and Astros advance
    White Sox beat the Astros.
    Red Sox were only blown out in game 1. In the other two, the White Sox held 1 run leads after 6 innings.

    The 2010 Giants was the other WS winner

    The only two of the top 8 playoff teams to meet this criteria in 2011 are the Phillies and Angels.

  38. Johnny Twisto Says:

    Mosc, that would be interesting data. I eagerly await your findings.....

    I guess you could either take the top three WARs on the pitching staff, or the harder route of trying to identify who team considered its top 3 SP by the time of the Series.

    And then you have the occasional mid-season call-up who is one of the top 3 SP, but doesn't have a full season's worth of stats, a la Whitey Ford in 1950.

  39. Johnny Twisto Says:

    (Actually, Ford did not have one of the top 3 WARs on that team, and probably was not yet considered one of their top 3 guys. So, take Mel Stottlemyre in 1964 instead. There will be other examples too. Their numbers will somewhat "corrupt" the data.)

  40. How about Milt Pappas with 3.9 WAR in only 144 IP for the 1970 Cubs ! - with an ERA+ of 167. How many staffs had 4 STARTERS with 3.9 WAR?

    As mentioned above, in 1996 Cox really outmanaged himself by removing Neagle in Game 4 in like the 6th inning to bring in McMichael or some other 10th man in an eleven man staff. The Braves had the lead in this game and basically handed it over to the Yankees. They had taken a 2-0 lead in the Series and were still up 2-1 after a Game 3 loss and leading in this game ( and about to break the Yankees backs 3-1?). But no, take out one of the best #4 starters of the last 40 years and replace him with a middle innings guy....nice

  41. There is a conspicuous absence of a 2000's era Big Three.

    The A's Hudson, Mulder, Zito group were so very close to making this list. In 2003, it was Hudson with 6.7 War (yowza!), Mulder with 5.1 WAR, and Zito with 4.9 WAR, just barely missing the mark.

    To be fair, that was the only year they all came close to this feat. Every other year someone had a bit of a down year (down year by Big Three standards).

  42. @19 Lawrence. I agree that Bahnsen had some skill at pitching but a 113 ERA+ will not result in a 5.9 WAR unless the pitcher is in a four man rotation and thus gets 40+ starts. Daniel Hudson has a 112 ERA+ this year and his current WAR is just 2.0. Relative to other 5.0 WAR pitchers of the last fifty years my guess is that the 1973 version of Stan Bahnsen was among the "worst".

  43. Johnny Twisto Says:

    There have been 599 pitching seasons of 5+ WAR since 161. The median ERA+ is 141. Bahnsen's ERA+ is 24th lowest.

    The worst belongs to Dave Giusti of the '68 Astros. He managed 5.2 WAR despite an ERA+ of 93 and "only" 251 IP. It looks like he was saddled with a historically awful defense. WAR thinks they were 102 runs below average in the field. Just give them an average defense and theoretically they'd go from 72-90 to around 84-78. Incredible. Anyone know of a lower team Total Zone? Anyone have memories how those Astros looked in the field?

  44. Johnny Twisto Says:

    This is interesting. Mike Cuellar went from Houston to Baltimore in 1969. Those O's were known for being tremendous defensively. On the surface, it looks like Cuellar finally found himself with the Orioles, turning from a decent mid-rotation SP into an ace. And yet the only change might have been in his defensive support. According to WAR, his own performance didn't change very much at all, and actually follows a pretty traditional aging curve.

    From '66-'68, he had a .288 BABIP. From '69-'71, it was .244.

    It's just shocking how much a defense can affect our perceptions of a pitcher's talent. But I still feel like we're so far from disentangling all the ways in which pitching and fielding interact (and may never get there).

  45. Jack Cooper Says:

    Just before the All Star break, I compiled a similar list, but instead of looking for three starting pitchers all with WAR above 5.0, I looked at a trio where the minimum WAR was 4.0 and the total (pro-rated to 162 games) was above 15. And I only looked at rotations starting in the lively ball era (or more precisely, 1920).

    Having your worst pitcher of the trio with WAR of 5 is setting the bar awfully high, so there is a fairly small sample to choose from and I don't think it necessarily represents what we would think of as truly great starting trios.

    For the record, the top ten trios I came up with were:

    Team Year Pro-rated WAR Pitchers
    Indians 1956 22.16 Wynn,Score,Lemon
    Cubs 1969 20.97 Hands, Jenkins, Holtzman
    A's 1931 20.44 Grove, Walberg, Earnshaw
    Braves 1995 20.14 Maddux, Glavine, Smoltz
    Mets 1973 19.92 Seaver, Koosman, Matlack
    Giants 1966 19.52 Marichal, Perry, Bolin
    Mets 1974 19.40 Matlack, Seaver, Koosman
    Royals 1994 19.16 Cone, Appier, Gordon
    Dodgers 1981 19.15 Valenzuela, Reuss, Hooton
    Reds 1925 18.64 Luque, Rixey, Donohue

    You might notice a couple of World Series winners in the list; The entire list, excluding this year's Phillies, consists of 44 teams, and in addition to the '81 Dodgers and '95 Braves, includes the '54 Giants as the only other team to go all the way.

    At the time I compiled this list, the trio of Halladay, Lee and Hamels was projected to obliterate the record. Now they are on a path of accumulating 20.93 total WAR, good for third place. It still is an historic rotation; adding Oswalt and Worley to the mix would probably place them at or near the top of the best 5-man rotations.

  46. @43..Thanks Johnny...Did anyone lower than Bahnsen on that list have a higher WAR than his 5.9?

  47. Johnny Twisto Says:

    Yes, four of them. Carl Morton of the '75 Braves had a 6.2 WAR with a 108 ERA+. I don't think I've ever heard of him. I'm pretty sure those Braves teams had some lousy defenses as well. (That's always going to be the major reason when a pitcher's WAR doesn't seem to correlate to his ERA(+).)

    I was just going to write I know Phil Niekro had some big WAR years in the '70s when his traditional record didn't seem to put him quite at that level, and sure enough there he is with an 8.5 WAR for the '77 Braves. 111 ERA+ in 330 IP. That's easily the highest WAR among the ERA+ seasons below Bahnsen. It's easily the lowest ERA+ among all 8+ WAR seasons since 1893. (Next is Mickey Lolich '71 at 125.) Of the 149 8+ WAR seasons, median ERA+ is 167.

    The '77 Braves were apparently 138 runs worse than average in the field. That's our new leader. That's the season Ted Turner tried managing for a game.

  48. @44, JT -- Good point about Cuellar. And let's make it explicit that he went from an apparently putrid defensive team to an outstanding one. (Uh-oh, do I have to check to make sure that a team full of Gold Glovers really was a great defensive team? I guess I should ... be right back!)

    Phew! Yep, the '69 Orioles were a truly great defensive team.

  49. @45, Jack, those are interesting numbers-thanks.
    The bulk of the data just seems to be counterintuitive. The trope recently is that to win games and championships you need at least three top starters. But if five WAR is the cutoff for comparative excellence, there's very little to support that as either a predicate for or a guarantee of success. Not only did only 1 of the 18 teams with 106 or more wins have that type of pitching, only 7/18 teams who did even made it to the post season, and only two won the Series. Couple that with the curious fact that the most successful franchise, by far, the Yankees, have had only one pitcher (Mike Mussina) in the top 50 in career WAR who spent even close to half his career on their roster. It just doesn't seem that top tier pitching is a predicate for winning baseball games: more important is probably solid pitching paired with high quality bats. So, when the data doesn't support the presupposition, maybe it's time to re-examine?

  50. @45, Jack Cooper -- Nice work. I like your method.

    BTW, are you familiar with the method Bill James used in the Historical Baseball Abstract for ranking the "greatest" 1-year outfields, infields, etc.? For each group, he summed the Win Shares of the best guy x 1, the 2nd-best x 2, 3rd-best x 3, etc. I think that method sets too high a value on balance; it rates a group of 6, 5 and 5 higher than a group of 7, 6 and 4. Balance does have some value in the discussion of a "great group," but not that much.

  51. @47..Thanks again, Johnny. Carl Morton was half of the 1970 ROY curse. Both he and Thurman Munson met their ends in their thirties.

    I didn't realize defense affected WAR that much. Does the formula take into account how a pitcher gets his outs? A strikeout pitcher seemingly would be less hurt by a bad D than a groundball pitcher.

  52. General comment -- I think someone above (sorry, too rushed to check right now) said something along the lines of this:

    What looks like great pitching is sometimes more about great defense; and since WAR tries to filter the defense out of the pitching, some of the teams often identified as the best rotations ever don't make the 5-WAR list -- but those teams would still fit the (undefined) criteria of those who repeat the adage that "pitching wins championships." So this analysis may not really be addressing the adage head-on.

    Sorry for not giving appropriate credit, and apologies for any errors in my paraphrasing.

  53. @ 35

    I cut the WAR back to 2.5, removed the inning restriction, but still limited the search to 60% starters (looking for 3). From 2001 to 2011, it pulled out 59 teams, 48 with winning records, 28 playoff teams, 9 World Series teams, 4 winners. (4/59 vs 2/26 in the earlier search)

    There were 241 teams not in this class. 110 with winning records, 131 with losing records, 52 playoff teams, 11 WS teams, 6 winners.

    So that's close to pulling out 1/2 the WS winning teams without going in to determine who the actual starters were. Each league had at least 1 per year even with the 162 inning restriction.

    The big conclusion here is that only 8% of the losing teams have this property vs 30% of the winning teams. In reality, we're linking a parameter belonging to the "third best pitcher" with team success..

  54. @47, JT -- Amazing what tossing 376 innings can do for a solid pitcher's WAR value!

  55. Johnny Twisto Says:

    Carl Morton was half of the 1970 ROY curse.

    Oy, thanks. I didn't even notice the ROY when I looked at his page earlier. I'm a bit ashamed I'd never heard of a ROY (or at least, had completely forgotten him).

    I didn't realize defense affected WAR that much. Does the formula take into account how a pitcher gets his outs? A strikeout pitcher seemingly would be less hurt by a bad D than a groundball pitcher.

    There's no guarantee the defensive effects are "right," but since all the advanced defensive stats seem to find similar ranges of individual defensive performance (usually from around 20 runs above to 20 below average for the best and worst defenders), I'm inclined to think they're in the neighborhood.

    Yes, I am pretty sure B-R WAR applies its estimate of the team's defensive performance to the pitcher's balls in play. A high-K pitcher will not be as affected. I am pretty sure WAR does *not* make any adjustment for the team's area of weakness. E.G., a flyball pitcher on a team with a terrible infield will still receive "credit" for all that bad defense, even if it didn't necessarily hurt him much.

    ***

    Balance does have some value in the discussion of a "great group," but not that much.

    Maybe Bill's formula wasn't the best approach (mathematical precision has never been his forte), but I agree with the concept. I'd never want to see, say, an outfield consisting of Ruth '21 or Bonds '01, plus two stiffs, listed among the best outfields ever, even if their combined WAR (or whatever) put them there.

  56. Good writing. I have read through the other reader's comment and agree with most of them: 3 great pitchers do not make the team, the rest must be of a certain standard to win the championships

  57. @53

    There's a typo. I meant 2001 to 2010.

    There's no advantage in the playoffs, when one of the 28 teams met up with one of the 52 teams, they advanced 7/15 times and took 2/5 WS.

  58. @55, JT: "I'd never want to see, say, an outfield consisting of Ruth '21 or Bonds '01, plus two stiffs, listed among the best outfields ever, even if their combined WAR (or whatever) put them there."

    Agreed -- and that's what I like about Jack's method @45. Setting a minimum threshold insures that everyone in the discussion had a "good year," without overly rewarding perfect balance.

    If I want to test the theory that "3 good starters is a great formula for postseason success," I think a 3-man crew with 6, 6 and 4 WAR is a better test than a group that all have 5 WAR, even though the latter group scores better in the James method.

  59. @47

    Yes, Ted Turner is listed as a manager in baseball-reference.com with an 0-1 record. After firing his manager, he sat on the bench while an assistant manager ran the team. The League commissioner "fired" him after 1 game.

  60. I looked at the last 5 WS. I looked at each game and the WAR and compared it to which pitcher was winning when the first pitcher was relieved. The favored pitcher was ahead 16/25 times, but his team did not necessarily win. If you only count the games were the difference was greater than 0.5, the expected pitcher won the duel, 15-4, but, again, his team did not necessarily win. The last 4 WS went as expected with the favored team based on head to head WAR winning each year. The WAR favored team in 2006, the Tigers lost to the Cards. The WAR favored pitcher was 3-2 in head to head duels in that Series.

  61. So, in the last 5 World Series, if you compare the WARs of the 2 starting pitchers and if the difference is 0.6 or more, the odds are better than 80% that the favored pitcher's team will win, unless that pitcher already lost a duel or pitched in a loss or the unfavored pitcher already won a duel or pitched in a winning game.

  62. Jack Cooper Says:

    I think the impression one could get from looking at WS winners, particularly in the expansion era, is that outstanding starting pitching can carry an otherwise mediocre team.

    There are at least five examples that come to mind: The '65 Dodgers, '69 Mets, '85 Royals, '88 Dodgers and '05 White Sox. Oh, and the '95 Braves. It's true that none of these teams (except the Braves) were characterized by a singularly dominant trio; but these teams all were associated with outstanding starting pitching.

    I can't think of any examples, except maybe the '76 Reds, where an outstanding offence carried mediocre pitching. I think it's mostly an unusual quirk to have more than two outstanding pitchers who are also healthy enough to be really dominant.

    When I compiled my list of dominant starting trios, there were only a few that made repeat appearances; The recent Braves trio cracked the list a few times (before Maddux, Glavine/Smoltz/Avery had an outstanding season), the early 70's Mets trio popped up a couple of times, the late 60's /early 70's Cubs, led by Jenkins, made the list a few times, and the 60's Giants combo featuring Marichal, Perry and Bolin made the list a couple of times.

    While the Mets and Braves each one a title (although the '69 Mets lacked a third outstanding starter), the Giants never won a pennant with Marichal and Perry and the Cubs were pretty awful outside of their starting pitching (which seems strange considering Billy Williams and Ron Santo were still active and playing well).

  63. Gary Gentry of the 1969 Mets was a rookie, but he had the third best WAR in the National League at 3.1 for the "third best WAR pitcher". The Mets played 5 games in the WS with Seaver and Koosman pitching games 1,2,4,5. Gentry was the winner in Game 3, pitching 6.2 innings of a 5-0 victory over Jim Palmer of the Orioles. The #3 best WAR starting pitcher for the Orioles was 20 game winner Dave McNally at 2.0.

    1966 was the only season Bolin was a regular starter. He had the ninth lowest WAR at 4.6 in the NL with Marichal at 9.0 and Perry at 5.8. They finished 1.5 games behind the Dodgers. He had a 5.1 in 1968 as a part-time starter with Marichal at 6.1 and Perry at 6.2. The Giants finished second, 9 games behind the Cards.

    It was not the offense who kept the Cubs from winning a Division title.

    In 1969, the Cubs were 3rd in offense and 4th in pitching in terms of R/G with the largest league differential. In 1970 they 2nd in offense and 3rd in pitching with the largest differential. In 1972 they were 4th in offense and 4th in pitching with the 3rd differential, finishing 2nd in the division each year.
    Why did they not win the division vs the Pirates in 1970. In terms of R/G, the Cubs were comparable to the Reds in both hitting and pitching and the Reds won 102 games vs the Cubs 84.

    The Cubs had a 4.98 to 4.50 R/G scored edge over the Pirates.
    The Pirates had a 4.10 to 4.19 edge in runs allowed.
    Cubs were +0.79 to +0.40 for the Pirates.

    Pittsburgh's starters were 64-58 with a 3.70 ERA
    Chicago's starters were 71-59 with a 3.55 ERA.
    Cincinnati's starters were 57-24 with a 3.44 ERA
    This gives Chicago a +7 over the Pirates

    Pirate's relievers were 25-15 with a 3.71 ERA
    Chicago's relievers were 13-19 with an ERA of 4.63
    Cincinnati's relievers were 45-36 with a 3.99 ERA
    The Pirate's relievers made up the difference to win the division.

    The Cubs starters had strong support, 47-3 in 66 games where the offense score 6 or more runs.
    The Pirates starters were 28-6 in 56 games where the offense score 6 or more runs.

    The Mets had Koosman, Seaver, and Matlack reaching WARs of 4 in 1973, 1974 and 1976
    The won the division in 1973 and were miserable in 1974.
    In 1976 they were 86-76 with the 3 finishing 21 games over 0.500 with ERA+'s between 112 and 127. That means the rest of the team was minus 11.
    The Phillies finished 101-61 with their top 3 game winners going 26 over 0.500 with ERA+'s from 97 to 116, but the rest of the team was 14.

    In 1973, they won their division with the 4th best record in the NL with the #2 defense and #11 offense. They lost in 7 games to Oakland in the WS, with the Mets getting 5 Quality Starts to Oakland's 2. They even outscored the A's 24-21.

  64. The 1970 Cubs were 62-63 vs teams starting RHP, 22-15 vs teams starting LHP. They hit 0.252 in games started by RHP and 0.283 in games started by LHP. So, opposing teams would try to avoid using LH starting pitchers vs the Cubs. That was tied for the lowest number of starts vs LHP's in the League. The Pirates were 68-42 vs RHP, 21-31 vs LHP.

  65. Can I only state exactly what a comfort to come across somebody that truly is aware of exactly what theyre talking about on the web. More people must examine as well as understand why. I find it difficult to feel youre not popular when you undoubtedly have the surprise.

  66. bnc network...

    [...]Three 5-WAR pitchers on one team … but will they win it all? » Baseball-Reference Blog » Blog Archive[...]...