No announcement yet.

Forest and trees

  • Filter
  • Time
  • Show
Clear All
new posts

  • Forest and trees

    I'm inclined to think a lot of compiled and recompiled and overwrought reflections on player ability and strategy goes into the category of missing the forest from the trees. The thing I really want to know is whether move X or player Y wins games, not whether X or Y theoretically increases the probability of winning games if baseball was played as a simulation. I'd like my proof to be a little more concrete. There are plenty of guys with mere mortal numbers who knew how to win, maybe just because they knew when to turn on the switch. Reggie Jackson comes to mind.

    One of the side things that bothers me is that in compiling all that stuff, we have also lost track of what I would say are fundamental inquiries into the basic nature of baseball.

    I've been looking hard at whether something simple -- scoring first -- really changes your prospects. They make a huge deal about this in soccer, notably because scoring first more or less changes your chances of winning from about 45% at kickoff to about 72% (about 10% of games end in draws and 18% are come from ahead losses). The whole game plays out differently depending on who gets the first goal. But at the NFL, it really takes scoring the first 10 points to really change your odds of winning vs kickoff.

    But baseball? Seems like no one knows and no one is even interested even though baseball is very well suited to trying to exploit first to score if it is a real important thing. I've asked around on several boards and consulted with several authors who write esoterica involving data and reflections on data, and it seems no one has even considered the issue of whether scoring first changes outcomes.

    So now I come to a second area that I have worked on -- home field advantage. Is it a thing? Should teams bust their humps to get better seeding for the playoffs? How about travel teams playing at neutral sites -- should they be home or away if that is decided by coin flips or by pool standings?

    Here too, I looked hard at other people's work to see if there is something on the topic, and I found little if anything. All anyone can talk about during the stretch is getting home field for the playoffs, and yet there's no proof one way or the other that having it is advantageous. So I guess my work on the subject makes me the leading expert on the topic.

    And the answer is: Hard to tell. There's plenty of data from 150 years of playing games, but nothing that definitively explains what is going on or what is advantageous or how to act if the opportunity arises to choose your position.

    And I'm going to say if there aren't definitive answers to a straight forward poser like that, then a lot of research in a lot of topics with massive amounts of number crunching isn't explaining much either.

    The historically known phenomenon is that MLB teams win at home about 54% of the time. Lots of travel coaches use that as a basis for acting, though many choose to bat first as an act of psychological warfare. It also occurs to me that anywhere seedings are used to determine who is home, that a team that would choose to bat first if given the opportunity might tank or otherwise game the pool play, and that's pretty ugly.

    There are several operative theories why home teams win more often in the majors, in no particular order: (i) home teams assembled around their parks; (ii) institutional knowledge of the facilities; (iii) home cooking, reduced travel fatigue and visitors partaking in local nightlife; (iv) subconscious officiating bias against visitors due to impact of home crowds; (v) batting last; (vi) unilateral ability of home team to rain out games when that might serve its advantage; (vii) in interleague games, visiting teams not designed to optimize applicable site rules; (viii) visiting teams intimidated or distracted by opposing fans.

    But no hard proof that any of these things was a real thing.

    In 2020, a lot of these things vanished. This created a moment for a controlled study, at least with respect to some of the above factors. The biggest ones were, IMHO, the absence of attendees which should have eliminated officiating bias and distractions and the absence of cross-country travel which should have eliminated most of the wear and tear and grind of a season. In many cases, home cooking was eliminated, with both teams quartered in hotels to prevent outbreaks and to control the movements of young men, so the disadvantage of actually being out of your own bed was negated. The temporary use of the DH in all games should have eliminated the problem of inefficient team assembly for interleague road games.

    And here are my assumptions: that the various Covid rules including 7 inning doubleheader games, California tie breakers, a short sprint rather than a marathon, masks and so forth, the opt-outs and the varied playoff rules that included more teams and more significance to seeding for the playoffs didn't really affect how games were played, or if they did, everything was a wash.

    So here it is: home teams won at a .557 clip in 2020, actually higher than normal. 898 games were played; two were canceled. The projection based on the long term result was 485 home wins, and the actual result was 500.

    It isn't much and I assume it would be within normal year to year fluctuations if 2020 had been a normal year, but with many of the problems associated with being a visitor either eliminated or reduced, the home win rate should have been lower than usual, not higher.

    One thing we can tell wasn't important: inefficient teams in interleague games. The result for AL vs NL was exactly .500. The novelty of the DH to NL teams didn't seem to affect them at all.

    So I'm a little shocked that home teams did better than projectable even after circumstance eliminated most of the factors that theorists considered significant to the prior results, and what wasn't eliminated simply remained the way it always has been.

    I also think the takeaway is going to be that having home games is overrated. The impact of off-the-field decisions and the effect of the absence of crowds wasn't negligible; it was contrary to what you would theorize. And it does not seem likely batting last would be more important when other factors dissolve.

    So pending further research, I would say we know nothing about whether being at home really improves your chances if you are a major league team.

    Now what?

  • #2
    HFA is always about 4%Points or so i think (Home Team winning 54% of the Time ob Teams have same quality).

    scoring first is different than in soccer since you can't play defensive in baseball and run out the time like you can in soccer.

    scoring first thus only matters in that the opponent needs to make up the run.

    here is an expectancy calculator.just scoring a run in the first inning only moves the needle minimally (to about 51%) but if it is consolidated with a "zero" at the end of the first this goes up to 56.

    of course that changes if you score first later, if you score first in the 9th it is more than 80% chance to win.
    I now have my own non commercial blog about training for batspeed and power using my training experience in baseball and track and field.


    • #3
      It seems like something is wrong with the link or with the research. It says that only 2271 games over 62 years (36+ per year across MLB) began with the visitor scoring before making the first out. That seems absurd given the likes of Ichiro, Rickey Henderson, Lou Brock, Pete Rose on and on playing in that period. Something is up with the way that was figured.


      • #4
        Originally posted by rodk View Post

        It seems like something is wrong with the link or with the research. It says that only 2271 games over 62 years (36+ per year across MLB) began with the visitor scoring before making the first out. That seems absurd given the likes of Ichiro, Rickey Henderson, Lou Brock, Pete Rose on and on playing in that period. Something is up with the way that was figured.
        I think the number 2271 reflects the number of games played in which a run was scored before the second batter had a ball or strike on him, which is a subset of those times when a run was scored before there was an out. When you click on ball one, the number goes to 665. When you go to 1-1, the number is 639.

        I also think that the count reflects the situation when the pitch is thrown, not at the end of a given play which may or may not have resulted in an out.


        • #5

          I agree that often baseball fans and even analysts and scouts tunnel vision too often on trees, and sometimes on the forest as well. To me, the bottom line is that if you were to assemble the best theoretical 25-player team for your favorite to manager to win the most games possible, you have to be flexible on what perspective to take, for which players, and when.

          An even more common problem are when fans compare trees that don't really lend themselves to direct comparison. The easiest example is trying compare HR hitters in the Dead Ball era to other eras, or not giving Dead Ball HR hitters their due.

          It's not rocket science. Just calculate an "HR+T" type of stat, similar to how OPS+ is calculated, but compared to the average team that year instead of average player.

          Almost no one cares about Gavvy Cravath having some of the best power hitting seasons of all-time, adjusted for era. In 1915 Cravath led the NL with 24 HRs, but his real achievement is obvious only when considering the average TEAM hit only 28.1 HRs that year. In other words, Cravath hit .853 HRs for every HR the average team hit, or an HR+T of .853.

          In comparison, when Barry Bonds hit 73 HRs in 2001, that equated to a .396 HR+T. To equal Cravath's dominant 1915 season in 2001, someone would have had to hit 153 HRs.

          When you adjust for era that way, the only players in AL or NL history who exceeded Cravath's 1915 season were Ruth (8 times, topped by his 1920 season when he had an HR+T of 1.171) and Gehrig (once).

          I'm not saying Cravath > Bonds. I'm saying almost no one cares about Cravath because he's tied for 767th in all-time HRs. Which again, is because most people have no inkling that they can't compare HR trees in the Dead Ball era with trees in other eras.

          Even for Cobb and Wagner, they aren't remembered for their pure power, but adjusting for era for them would produce results surprising to most fans, even long-timers on this site. Cobb's best HR+T season was .661 in 1909, or 118 HRs in 2001. Honus's best was .530 in 1908, or 95 HRs in 2001. And he didn't even lead the league in HRs.

          Back to the OP's question of forest vs. trees, IMO adjusting for era with an HR+ or HR+T type of stat is one "forest" way of measuring trees on a level playing field, regardless of era.
          Last edited by Unconventional; 09-18-2021, 11:42 PM.


          • #6
            Nearly all of this discussion, IMO, boils down to ensuring an ability to compare eras. Comparison across eras defeats the purpose of what the research is meant to accomplish.

            For example, I would be very, very surprised..... Shocked! (using columnist Bob Ryan's term), if the home field advantage of the current era Baltimore O's is measurable. Further, it would be interesting to compare team's current measure against the Orioles' HFA in their '66-'83 heyday, I'm sure the O's in their heyday had a strong advantage for some reason or another.

            Another example would be the Red Sox HFA within any season. For time immemorial, opponents have not been able to get their head around how to take advantage of the Green Monster. IMHO, Red Sox right handed hitters just accept their ability. They either know they can pull the ball off the Monster or they can't. Lefties either know they can go oppo or they cannot. This adds up to a major and resilient HFA over time.

            And then, IMHO, every team has the benefit of sleeping in their own beds and enjoying the comforts of home in every way. Home cooking doesn't just refer to umpiring. It literally could relate to their wives/GFs, or private chefs, ability to prepare 3 solid meals as often as possible.

            All of that would show up in analyses, I am quite certain. However, comparison across eras will involve variables which can barely be appreciated today. We cannot imagine what train travel was decades ago, how players could sleep on the trains, or whether they told the truth about doing so. The effects of cookie cutter stadiums may have levelled the effects of HFA during the days of cookie cutter stadiums in Philly, Cincy, Pittsburgh and STL. We cannot know what we cannot imagine.
            Catfish Hunter, RIP. Mark Fidrych, RIP. Skip Caray, RIP. Tony Gwynn, #19, RIP

            A fanatic is someone who can't change his mind and won't change the subject. -- Winston Churchill. (Please take note that I've recently become aware of how this quote applies to a certain US president. This is a coincidence, and the quote was first added to this signature too far back to remember when).

            Experience is the hardest teacher. She gives the test first and the lesson later. -- Dan Quisenberry.


            Ad Widget