Announcement

Collapse
No announcement yet.

The 2015 projected $ amount in the forecaster

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • The 2015 projected $ amount in the forecaster

    Is it based on nl/al 5x5 or 4x4?
    12 team NL only...11 keeper Max...260/360 in season.

  • #2
    For several years now, we have used 5x5 values in the book. However, the values are a bit distorted because playing time tends to be over-projected in the aggregate.
    "If you torture data long enough, they will confess." -- Ronald Coase

    Comment


    • #3
      OK thanks Michael...And where would the difference lie in Playerlink predictions vs. Forecaster predictions for a pitcher at this point of the preseason? For instance someone like Teheran is $17 in forecaster vs. $15 in player link on this site? I see small differences in his predicted stats...Why is that?
      12 team NL only...11 keeper Max...260/360 in season.

      Comment


      • #4
        The short answer is don't worry about what is in the Forecaster because the projected statistics on-line at BaseballHQ.com are always more current.

        If a player's projected statistics in the Forecaster and on-line are identical and only the dollar amount changed, it is because the rest of the pool of players differs quite a bit. If you are asking why projections themselves change, the reasons are:
        (1) playing time changes (IP for pitchers and AB+BB for batters) happen frequently and are processed overnight.
        (2) statistics are adjusted whenever a player's home ballpark changes.
        (3) even though most of this work happens before the Forecaster is published, sometimes we change our minds about a player, based on information not previously in the Forecaster.
        (4) sometimes we spot glitches or errors and correct them.
        This late in the off-season, categories (3) and (4) are less likely to occur.

        You can read the official version of this explanation at the first FAQ on this page.
        "If you torture data long enough, they will confess." -- Ronald Coase

        Comment


        • #5
          Yes that was what was curious to me and why I chose a pitcher...I figured the player pool may have something to do with it, but playing time? Why has a pitcher like Teheran's playing time (and therefore ip and k) been adjusted?
          12 team NL only...11 keeper Max...260/360 in season.

          Comment


          • #6
            I can try to explain the process, but the short answer is not to worry about differences between the projections in the Forecaster and what appears on line.

            Teheran was projected for 201 IP in the Forecaster and now is projected for 189 IP on line, so we're talking about a difference of just 12 innings pitched.

            Projections While We Work on the Forecaster

            BaseballHQ has a projection engine that produced our initial draft of projections not long after the 2014 regular season ended. The BaseballHQ writer assigned to write the first draft of the player's comment (you can see the names in fine print at the bottom of the page in the Forecaster) reviews the projection as part of reviewing the player and writing the comment. We're not hesitant about layering on adjustments to the projection when we are aware of other information about the player. However, our focus at this point is from the individual player's viewpoint: what his skills are and where they are most likely heading. We aren't concerned very much about playing time and are not concerned at all about the team context, which changes for many players before draft day.

            Projections on BaseballHQ.com

            We start with the same projections as the Forecaster. Most of the work on the skills portion of the projection, the rate statistics, was already done by this point. Still, especially early in the off-season, you might see changes in the rate statistics.

            Most of our efforts now though are on the playing time. Our process no longer is on the individual player. We have a playing time analyst for each team that updates the playing time on a team depth chart. That's what drives the playing time. It changes whenever the playing time analyst changes the percentages, with overnight statistic updates to apply those percentages to the projection files, the Custom Draft Guide, and our other statistical files & tools. We're not worried about whether a player's playing time is higher or lower than what it was listed at in the Forecaster: we're looking for our best YTD estimate of playing time based on today's information and the team context.

            Note that playing time estimates will continue to be fluid. This tends to affect players who may or may not make the opening day roster, more than a guy like Teheran. If you look at the total innings pitched on the Braves depth chart (there isn't a total listed, but you can add them up) you will see that we often overproject playing time during most of the off-season but that later in spring training the totals typically will collapse down to 100%.
            "If you torture data long enough, they will confess." -- Ronald Coase

            Comment


            • #7
              See the Consumer Advisory in the front of the Forecaster (page 7, I believe) for a full explanation.

              Comment

              Working...
              X