Still in the Driver’s Seat

The post-season odds report at Baseball Prospectus still likes the Sox to win the series despite last night's loss. They have the Sox at 50.5% to take the division series, down from the erotically-charged 69% before Game 1. The Yanks boosted their chances all the way to 76% from 64% after taking the first game in the Bronx. This means, it seems to me, that the dorkazoidals at BP, or at least their robot machines, saw the Sox as a bigger favorite than the Yanks to win its division series before the post-season started.

I'll interpret the numbers here for you all: It would be an epic choke for either team to lose its first round series. That, or odds reports for 5-game play-off series should probably be ignored.

30 comments… add one

  • Realizing your tongue is planted in cheek, these odds make no sense! What are the historical numbers for visiting teams that lose the first game of a 5 game series? The Sox aren’t that much better than the Angels, if better at all.
    Coolstandings has the Angels at 70.1%, comparatively.

    SF October 9, 2009, 10:05 am
  • “This means, it seems to me, that the dorkazoidals at BP, or at least their robot machines, saw the Sox as a bigger favorite than the Yanks to win its division series before the post-season started.”
    Yeah, that seems utterly bizarre, Going by run differential(yes, I know not as accurate as WAR or second/third order wins when measuring talent level), the Sox were only about 1 or 1.5 wins better than the Halos this year. The Yankees were like 11 wins better than Minnesota – and they had a plethora of other advantages going into the series (no Morneau, worn out Twins team, etc). So it does seem very strange that the Sox would be the bigger favorites.
    The Yanks being 65% faves seems about right (before GM1), The Sox being 69% seems way, way out of line – and not just in hindsight after last night. Honestly I think the Sox are better, by a decent bit, than the Angels, but the home field advantage alone should make up for a lot of that difference (Sox under .500 on the road exacerbates that)

    Mark-YF October 9, 2009, 10:12 am
  • SF, I was very surprised when I saw these numbers, too. It looks like the computer viewed the Sox as the strongest team going into the post-season, which I think is a fair view. But wouldn’t the difference between the Angels and Twins and the fact of home and away games make the Yanks a stronger favorite? Very weird to me.

    Nick-YF October 9, 2009, 10:15 am
  • Mark – I think run differential (Pythag) has a higher correlation to winning percent than WAR. I just read that recently somewhere.
    Found it: http://www.fangraphs.com/blogs/index.php/war-it-works/
    Although WAR has a pretty good correlation as well, but the +/- record would fit into a 5 game series, so it really comes down to randomness.

    dw (sf) October 9, 2009, 10:26 am
  • which I think is a fair view
    Eh, I don’t want to get into a downward spiral about the fine grain differences between our teams, but the Sox just aren’t/weren’t as good as the Yanks. The computer has no clue.

    SF October 9, 2009, 10:29 am
  • Yes, RD has a higher correlation to winning percentage but not necessarily talent level. I think when predicting future success rather than correlating to past success WAR is a little better (comparatively if not absolutely).
    Anyway, that was an aside because any way you measure it, the difference in talent between Yanks-Twins and Sox-Angels isn’t remotely close. The Yanks should be huge favorites and the Sox should be slight favorites.
    Honestly I think, and it’s touched upon in that link, the biggest wildcard for WAR as it translates to wins is baserunning, which is highly volatile anyway and can fluctuate wildly from year to year (personally I think it’s largely attributable to coaching and luck). One of the biggest reason the Yankee offense underperformed so badly last year was because they were by far the worst baserunning team in the league (while in 2007 with most of the same players they were the third-best!) – I remember reading that their swing in baserunning runs from 2007-2008 was something like -60! I suspect that happened to the Rays this year, as well.

    Mark-YF October 9, 2009, 10:41 am
  • Mark – The Yankees baserunning was totally affected by Swisher, that guy ran into allot of outs! :)

    dw (sf) October 9, 2009, 10:45 am
  • Higher correlation between win-loss and runs scored/against doesn’t necessarily mean it’s better – WAR is is context-independent (which I think the article goes about, but not too much) while Runs are depends obviously on context. If you want to translate WAR to runs, you can think of it as the number of “expected” runs given a regular distribution of the probabilities..

    Lar October 9, 2009, 10:52 am
  • It seems like you’re right, but I actually haven’t seen the Yanks’ baserunning stats this year.
    Although funny aside: I just quick Googled “Yankees 2009 baserunning” and the first hit was “Swisher’s clever baserunning helps Yanks win 2-1.”

    Mark-YF October 9, 2009, 10:53 am
  • SF – no offense, but I would say the computer usually probably has more clue than us, haha.
    Given these numbers, it’s a surprise the Sox didn’t win 10-0 yesterday.

    Lar October 9, 2009, 10:54 am
  • Mark – that’s funny!
    I remember the games against the sox and Swisher’s baserunning blunders really stood out to me.

    dw (sf) October 9, 2009, 11:02 am
  • Further derailing the thread, Basbell prospectus has a pretty cool baserunning metric here:
    http://baseballprospectus.com/statistics/sortable/index.php?cid=468149
    Looks like Swish was a below-average baserunner, but cost the team less than one run on the year. The yanks as a whole were -6.34 runs on the year, thanks largely to Posada’s awesome -8 runs on the bases.

    Mark-YF October 9, 2009, 11:04 am
  • All, It is not the computer’s fault. It is the fault of the programmers or designers which apparently applied faulty logic to determine the odds. However I do not find it surprising that the Sox would have been the favorite against the LAAA’s since as all Sox (and Angel) Fans know the Sox constantly clobber the Angels. It would also not surprise me that the Sox may still be a slight favorite, for if the Sox win tonight they would be a heavy favorite, given past history and what would now be a 3 game series with 2 of the games at the Fenway Bandbox. The Sox must be a favorite tonite with Beckett vs Weaver, hence the Sox as still a slight favorite overall. Of course if the Sox blow it tonite…

    Joe October 9, 2009, 11:06 am
  • Joe – yeah, I agree with the Sox being the favorite. That’s not surprising, the Sox are a better team. Even with the Angels having home field, I would expect them to be slightly favored. but 69% means they are overwhelming, prohibitive favorites, which I just don’t see.

    Mark-YF October 9, 2009, 11:16 am
  • I think the biggest surprise isn’t that the Sox were so heavily favored to begin with, but rather that they’re STILL favored despite losing the first game.

    Atheose October 9, 2009, 11:18 am
  • 5 game series…winning the first game is a big deal:
    lose the 1st one, and you have to win 3 of 4…in this particular case, only 2 of those 4 can be at home…
    win the first one, and you only have to play .500 ball…in this particular case with only 2 of the maximum of 4 remaining games on the road…
    i understand why the sox might still be a favorite, and particularly in tonight’s game with beckett and a better bullpen…but what am i missing beyond post season reputation?…we are talking about jered weaver, not jeff, right?…he and beckett look similar for this year anyway…

    dc October 9, 2009, 11:46 am
  • Odds reports, as we all know, are especially worthless in a five-game series. In the last 10 years, the Sox have rallied back from 2-0 down to win a five-game series twice (1999 and 2003). They’ve rallied from a two-game (or greater) deficit with five games left in a series to win it twice (2004 and 2007) and force a deciding game once (2008).
    That’s five times in 10 years that the Sox have been at the brink of elimination, down by at least two games in a series and rallied to either even the odds or win it outright.
    In other words, it’s five games. Anything can happen. /stating the obvious
    Thank God we have Beckett going tonight though. Otherwise I’d be planning for next year already.

    Paul SF October 9, 2009, 12:10 pm
  • Odds reports are not useless if they’re right.. haha..

    Lar October 9, 2009, 12:45 pm
  • OK, so I need some help from all you technological geniuses. Is there a way to get 880 on the iPhone? In order to listen to 880 on their website you need to have flash installed and you are not able to install due to iPhone restrictions. Does anyone have any suggestions or maybe there is something I am missing. Going to the game tonight, but would really like to tune into the radio as well. Thanks! (Sorry to break up the thread)

    John - YF October 9, 2009, 1:16 pm
  • John
    you need the MLB app to listen to the radio broadcast at the game. It may be worthless since the delay will be considerable and the connection spotty due to heavy internet use at the stadium. I recommend a $15 radio from duane reade! Where you sitting? Ill be up there too…

    sam-YF October 9, 2009, 1:20 pm
  • Thanks Sam. Yeah I figured it was something complicated/expensive like that.
    I am in Right Field, Middle Level. Headed in around 3:30, hopefully be there by 4:15 the latest. Where are you sitting?

    John - YF October 9, 2009, 1:33 pm
  • Im in sec 420A where my season tix are….ill be rolling in basically right at game time.

    sam-YF October 9, 2009, 1:36 pm
  • Well if you see a big guy wearing a dark green SPARTANS sweatshirt and Yankees hat, say hello. Enjoy pal! Hopefully two things happen, 1: we get the game in and 2: we go home happy!

    John - YF October 9, 2009, 1:40 pm
  • An epic choke if the Halos beat the Sox? Nah, not unless Boston gets swept. The Angels have a better record, and a much better lineup than previous years, not to mention home field advantage.
    One possible explanation for the odds: Josh “Postseason God” Beckett is starting Game 2 for the Sox.
    Boston does need to win tonight. If they get out of Anaheim with one win, it doesn’t matter if it wad Game 1 or 2 and I like their chances. Lose tonight and the prognosis is grim.

    Hudson October 9, 2009, 1:43 pm
  • RE: phone questions. I learned that my cell phone has a regular ol’ radio receiver in it. It was buried back in the features (my wife actually stumbled on it). The fancy smart phones don’t have such a thing?

    dw (sf) October 9, 2009, 2:03 pm
  • The iPhone 3GS was supposed to have an fm receiver, but they scrapped the idea. The next generation iPhone might have one though.
    John, I think the free mlb.com app lets you listen to some of the games on radio. I bought the mlb.tv app so I wouldn’t know.

    Atheose October 9, 2009, 2:41 pm
  • I think Dave Pinto @ baseball musings has the app to listen now that Ath mentions it. He said something about it at his site some time ago.

    dw (sf) October 9, 2009, 2:47 pm
  • If Beckett’s curve is on (which I understand from some of you SFs has been the main hiccup of late) then I think they stand a great chance to win tonight, which would give them homefield advantage. It’s on him – and he has shown a pretty good capacity for shouldering such loads in October.
    Who is # 3 for Boston – Buccholz? That would worry me a bit but maybe I’m not giving him enough credit.

    IronHorse October 9, 2009, 4:00 pm
  • Yep, Buchholz for Game 3. He struggled in his last two starts, so I’m not too confident about that game.

    Atheose October 9, 2009, 4:12 pm
  • By the way, SG over at Replacement Level Yankees Weblog has run his own projections after the game 1 results. Those projections are based on the rosters the teams are ACTUALLY using (no Brad Pennys, Cody Ransoms, etc) and it incorporates not only 2009 performance but also 2007 and 2008. Further, it accounts for which starting pitchers have already been used and can thus only pitch once more.
    Those projections have the Sox winning the series at 37.2%.

    Rob in CT October 9, 2009, 4:25 pm

Leave a Comment