Feb
05

2009 Yankees didn’t fare well in one-run games, blowouts

By

The 2009 Yankees won more games than almost everyone expected. That’s almost always the case when a team wins over 100 games, but it holds particularly true in the three-team dogfight that is the AL East. They got to the 103 win mark because they were a good team that got hot at the right time, but also because they got lucky in some ways. Their 915 runs scored and 753 runs allowed works out to a 95-67 Pythagorean record, which they outperformed by eight wins. Does this mean the Yankees got lucky?

That’s a tough question to answer. Their Pythag record suggests that they did, but that takes gross runs scored and gross runs allowed and factors them into an equation. It doesn’t take into account how the team scored those runs. Over the course of a 162-game season, the reasoning goes, the issues of how a team scores runs should even out. Sometimes it doesn’t. Perhaps we can learn a bit more about the nature of the 2009 season by looking at a breakdown of the results.

Brandon Heipp of Walk Like A Sabermetrician looks at the 2009 run distribution of all 30 MLB teams. He breaks games down into one-run games, blowouts (five or more runs difference), and the games in between. Here’s the odd thing: The Yankees underperformed their record in both one-run games and blowouts. It’s the latter that seems odd, since a team with an offense like the Yankees figures, intuitively, to have a favorable record in blowouts.

The Yankees played in 38 one-run games and went 22-16, a .579 win percentage. Overall the Yankees had a .635 win percentage, and had a .653 win percentage in non-one-run games. They played in 51 blowouts and went 32-19, a .627 win percentage. This is more in line with their season total, but still a bit below. Still, no playoff team had as big a difference in blowouts as the Yankees. In those middle games, obviously, the Yankees killed, going 49-24 in 73 games, a .671 win percentage.

I’m not sure we can discern much from this data. It’s just interesting to see that while the Yankees had an over .500 win percentage in all three categories, that they still didn’t do exceptionally well in blowouts. I would have thought, since they outperformed their Pythag, that maybe they were inordinately good in one-run games. They weren’t. Though, maybe — and this is just a guess — maybe their tempered record in blowouts led to their Pythag underestimating their record.

Heipp also looks at the team win percentage when scoring X number of runs. When a team scores one run, for example, it wins 7.5 percent of those games. The magic number, it appears, is four runs, as that crossed the .500 line. It also adds the highest marginal win percentage value over the value before it. That is, the jump from scoring three to scoring four runs, in terms of winning percentage, is .186. Going from four to five runs adds just .106 to the win percentage.

Here’s another interesting bit: the Yankees played in 28 games where they scored four runs, and won just 14 of them. The average MLB team played in 21 games where they scored four runs, but we should have figured that the Yankees were above average there. I would say that reflects poorly on their pitching and defense, but then I saw that they outperformed the average when scoring two and three runs. When scoring two runs, the average MLB team won 20.8 percent of the time, while the Yankees won 28.6 percent. They won 41.2 percent of games when they scored three runs, against am MLB average of 33.7 percent. In both instances they played in fewer games than MLB average.

Again, I’m not exactly sure what we can take this data to mean. I’m not sure that, by itself, we can take it to mean anything definitive. I do think it’s interesting to note these trends. In some ways it bucks intuition. In other ways it gives us another way to view the 2009 Yankees as a team. They did well here, but not well there. Since this post contains a lot of random data, we’ll close with another random bit. The Yankees allowed 11 runs just once this season — and won the game. When scoring 11 runs, MLB teams went 110-6. Glad that the Yankees counted for one of those six.

Categories : Offense
  • Will

    With regard to run differential, I think you can remove Wang’s starts from the equation because he was obviously pitching with an injury.

    • Slugger27

      im sure CMW played a major role in the blowouts part of the equation… nevertheless, i dont think that means u can just remove him from the equation

      it simply means we should keep CMW’s steve blass-like meltdown into consideration when examining whether or not the yankees are below average in blowouts

      • DF

        I think though that Wang’s near-historic meltdown is skewing the data. Pythagorean records assume a normal distribution of runs allowed, and Wang’s disaster was anything but normal. Since the Yankees removed him rather quickly, I think it’s screwing up the Pythag, and the Yankees probably didn’t overperform quite as much as it’s suggesting.

        • DF

          Just to clarify my point, it’s not that I think you can just take Wang out and pretend he didn’t happen (TWSS), but including his numbers in the Pythag isn’t representative of the 2009 Yankees’ true talent level. When the team gave up 22 runs against the Indians, that wasn’t a normal blowout, and it’s unlikely it would have happened had any other starter started the game, even Phil Hughes during his ineffective periods.

          • Will

            I agree. You can’t take random events out of the equation, but Wang is one that is very specific. It’s not like removing a game in which the Yankee starter had to leave because of an injury because it’s likely something similar happened to a Yankee opponent. With Wang, however, the Yankees were using a pitcher who was injured, so it specifically detracted from their pitching staff’s performance. Unless you assume that the Yankees faced 6-7 injured starters, that’s not a factor that would be normalized.

            • bexarama

              I think someone did some study where they took out the five biggest blowouts both ways and then worked out the data to prove that, without those games, the Yankees “should” have been a team that won 98 games or so.

              Without looking it up, I feel like the Yankees generally overperform their Pythag, though. I know the 2004 Yankees should have been like an 88-win team but they won 100+ games.

              • Chris

                Can’t you say the same thing about any team? If you remove the biggest losses, then the remaining games would ‘show’ that the team should have won more games. If you want to be fair, don’t you need to take out both the biggest wins and the biggest losses?

                For the Yankees, the biggest run differentials would be:
                5 worst Losses: -18, -13, -10, -10, -9 (two were courtesy of Wang)

                5 best wins: 10, 10, 10, 11, 15

                The net of removing those 10 games would improve the Yankees run differential by +4 runs. That’s not a big difference over the course of the season…

                • bexarama

                  Yeah – that’s what this person did. It’d be totally unfair to just throw out the five biggest losses without throwing out the five biggest wins too. I can’t find the article right now though, grrr.

    • Zack

      You can do that with every team.

      Take out DiceK’s stats pre-DL because he was pitching through leg/shoulder injuries.
      Take out Oliver Perez’s stats because he had a bad knee.
      Take out some of Joe Saunders’ bad starts because he was pitching through a bad shoulder prior to his DL trip.

      That’s 3 other SP after thinking for 2mins, I’m sure every team had a hurt pitcher that affect their run differential

      • Slugger27

        ya that was kinda my point… its not fair to remove them from the study, cuz then youd have to do it for the other teams’ outliers

        but to take the conclusion of the numbers with a grain of salt because of the CMW phenomenon i dont think is unfair at all

        • Zack

          no i understand, but the danger is that we take their run differential with a grain of salt but we often forget about other teams because we dont pay attention to their Anthony Claggets or injured SP messing up the run differential.

    • hutch

      “With regard to run differential, I think you can remove Wang’s starts from the equation because he was obviously pitching with an injury.”

      Actually, Wang had essentially no effect whatsoever on the overall record: he contributed to two +5 losses and one +5 win. (See schedule/dates below)

      And, the oddity here is that the Yankees performed BETTER pre-ASB in +5 RD games, than they did post-ASB when they had the overall best record in baseball in the May-Sept period:

      Pre-ASB Record: 15-8
      Post-ASB Record: 16-11

      So, really the question is what was different in the second half than the first (and it’s strange because the team had the best record from May, what 15th on?).

      Statistical Anomaly Alert:

      Months with the fewest +5 results — June (5) and July (5)
      Months with the most: August (12) and September (11)

      ** CMW Pitching

      4/6 (l)
      4/9 (w)
      4/11 (w)
      4/13 (l) **
      4/14 (w)
      4/16 (l)
      4/18 (l) **
      4/25 (l)
      4/28 (w)
      5/9 (l)
      5/13 (w)
      5/19 (w)
      5/20 (w)
      5/25 (w)
      5/27 (w) ** (Relief App)
      5/30 (w)
      6/2 (w)
      6/9 (l)
      6/14 (w)
      6/26 (w)
      6/27 (w)
      7/7 (w)
      7/11 (l)

      15-8

      7/24 (w)
      7/27 (w)
      7/31 (l)
      8/1 (l)
      8/6 (w)
      8/8 (w)
      8/13 (w)
      8/16 (l)
      8/18 (w)
      8/21 (w)
      8/22 (l)
      8/26 (w)
      8/27 (l)
      8/29 (w)
      8/30 (w)
      9/2 (w)
      9/4 (l)
      9/6 (l)
      9/7 (w)
      9/11 (l)
      9/13 (w)
      9/15 (l)
      9/19 (w)
      9/20 (l)
      9/28 (w)
      10/2 (l)
      10/4 (w)

      16-11

      btw, somewhere either I’m missing a +5 game or they counted 1 too many as I see 31-18 record in blowouts.

  • Zack

    “The Yankees played in 38 one-run games and went 22-16, a .579 win percentage”

    Imagine if they didnt have all those walk-offs

    • JohnC

      What would AJ do with all that spare shaving cream?

  • Manimal

    Pre-Arod we got blown out every other day… doesnt surprise me.

    • Slugger27

      arods 1st pitch homer is probably 1 of my 5 favorite baseball memories of all time… talk about a big “FUCK YOU” to everyone that hates him

      • bexarama

        I literally screamed “BOOM, THAT JUST HAPPENED!”

    • bexarama

      Wasn’t this more due to the bullpen going from awful to a strength? Not that I’m trying to diminish what A-Rod did this year.

      • Bo

        Hughes saving the pen had as much to do with the success as a-rod coming back.

        • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

          Teixeira, Sabathia, Burnett, and Swisher all dramatically outproducing their replacements and Matsui and Posada bouncing back from injury probably had more to do than either Hughes or ARod, though.

          Hughes to the pen is not the single magic bullet that saved the season the way you like to portray that it is.

  • dkidd

    good thing the yankees took david cone’s advice:

    don’t wait to yank wang

    /last chance for adolescent wang joke’d

  • Rose

    I was reading a stat on PA (link below) and the following statement caught my eye:

    Certainly, playing at home is important (the home team wins a 1 run game roughly 17% of the time).

    The home team wins a 1 run game only 17% of the time? That doesn’t sound right. But perhaps they were including other factors other than just “Home Team” and “Away Team” in a 1 run game?

    Reference:
    http://www.pinstripealley.com/stories/2

    • bexarama

      I think they mean out of the total games played at home, a 1-run victory by the home team will happen 17% of the time.

    • pat

      I think it means 17% of a team’s home wins are by one run.

    • Rose

      Yeah, I thought about that immediately after I posted it.

      Either way, it’s still a pretty recent (just a few days ago) and relevant article with a neat graph on it for you to check out.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

      That’s a horribly misworded attribution by jscape2000’s part. The article he sourced and “quoted” is a sports gambling site.

      Original article: http://www.sportsbettingstats......nlines.asp
      And original quote:

      Some basic statistics: In baseball, the home team wins by one run about 17% of the time, and the road team wins by one run a little over 10% of the time. This disparity is due to the fact that in the bottom of the ninth or later inning, the game is immediately over when the home team is ahead by one run. This turns a lot of games into one-run wins, which in turn makes that one run extremely valuable. That run gains even more value when the total is low, which makes it much harder to win by more than one run. Conversely, that run is less valuable for high totals. The value of the run also varies slightly depending on the money-line for the game.

      What the site that Pinstripe Alley sourced is actually saying is that 17% of ALL baseball games (including non-one-run games) are won by the home team by a one run margin. 10% of ALL baseball games are won by the road team with by a one run margin. The remaining 73% of ALL baseball games are decided by a margin of greater than one run.

      Obviously, 17% vs. 10% means that 62% of all one-run games are won by the home team.

      • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

        Damn broken blockquote closetag FAIL.

  • Beamish

    One word: “Mo”

    In the “other games” the differential is +/- 4 runs. In the +2 to +3 (and some nervous +4) wins we had Mo and Mo delivered constantly – 44 Saves.

    I am too lazy to do much more math than that. I have engineering to do.

    • king of fruitless hypotheticals

      Let me fix that for you, since you don’t want to do the math:

      One word: “Mo”

      Fixed.

  • Bo

    I’m shocked they got better in close games once Hughes went into the pen. Shocked

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

      I’m shocked they got better in close games once Hughes went into the pen. Shocked

      Yankees W/L in one run games, by month, 2009:
      Apr: 2-1
      May: 4-3
      :: Phil Hughes joins the bullpen ::
      Jun: 2-4
      Jul: 6-2
      Aug: 2-2
      Sep: 5-2

      Phil the starter: 6-4 (60%)
      Phil the reliever: 15-10 (60%)

      ZOMG HOW DID PHIL DO THAT HE’S SO AMAZING IN THE BULLPEN!!!!

      • http://theyankeeu.com Matt Imbrogno
        • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

          Hell yeah.

      • http://www.bomberbanter.com/ BomberBanter.com

        who was in the bullpen when they dropped hughes in there? Jose Veras? Edwar Ramirez? not exactly a cast of all stars. they let Veras go and Ramirez hardly pitched for the rest of the year.

    • Rick in Boston

      I’m shocked that your comment lacks fact.

      Hughes pitched in 12 1-run games in 2009. Two he started (the Yankees went 1-1 in those games). In the remaining 10 games, the Yankees lost 6 of them; Hughes was credited with one of the losses.

      B-Ref: 1
      Bo: 0

      • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada
        • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

          I propose that the winner of the 2010 RAB Fantasy Baseball League receive the boversimplification definition inscribed on a mug/t-shirt/Zazzle product of his or her choice to serve as his/her championship trophy.

          • http://en.wikipedia.org/wiki/Templeton_Peck Templeton “Brendog” Peck

            2nd’d

      • Rick in Boston

        Whoops…15 games. But still, 15 games.

  • http://www.bomberbanter.com/ BomberBanter.com

    Ha, while I was reading this, I couldn’t help but think of Wang, and it looks like I’m not alone. Obviously, the Yankees used some under qualified starting pitchers last year: Wang stunk, Hughes wasn’t that good, Joba had some good starts but struggled, Mitre and Gaudan were not good… With the addition of Javier Vazquez, you figure you’re only worried about the 5th starter, or maybe 30+ starts during the year as opposed to 60+, so the Yankees’ pitching will probably surrender a lot fewer blow out loses.
    One easy way to guess at a team’s record is to look at their record from the year before (say 2008, 89 wins) and then look at the roster fluctuation (added Burnett, Sabathia and Teixeira), which you can easily call an improvement over who they replaced and guess how many more wins these guys were good for. Now there are advanced stats you can use to calculate this, but I think it’s just easier to guess; 2 for each guy seemed reasonable to me, and I picked the Yankees to win 95 games. This didn’t take into account other teams’ roster fluctuations or the Yankees improvements in other areas (bullpen, defense), and the Yankees jacked it up to 103 wins. But, wins aren’t everything – the 04 Yankees won over 100 games, and take a look at that roster; the 09 yankees would DESTROY the 04 Yankees in a 7 games series and mess them up so bad their mothers wouldn’t recognize them in a 5 game series. Both 04 and 09 were pretty close in win totals, yet the 09 team is so much better. strangely enough, it seems as though wins aren’t everything.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

      Ha, while I was reading this, I couldn’t help but think of Wang, and it looks like I’m not alone.

      … nah, too easy.

      • http://www.bomberbanter.com/ BomberBanter.com

        OK, let me re phrase:
        “Ha, while I was reading this, I couldn’t help but think of Chien-Ming Wang, and it looks like I’m not alone.”

        Tommie, you should have – it would have been well deserved – and well played.