What started out as a quick All-Star break project has mushroomed into a full-blown research project. I set out to answer what is seemingly a simple question: Are in-season changes to park factors real and therefore actionable?

This is being written under the umbrella of DFS strategy but the concept transcends just DFS. If an in-season change to the way a park plays is real then our expectations for players in seasonal leagues should change, influencing roster management, player acquisition and trades.

The germination came about way back in April when the ball was flying out of Petco Park as if it were a mile-high. I put a note in my Toddy-do folder to look at this later in the season to possibly gain an edge come August, in both my seasonal leagues and DFS play.

Fast forward to the All-Star break and I crunched some numbers, but while doing so unleashed a couple decade's worth of pent up frustrations. Up until now I've been able to make peace with the fact that using park factors in player projection and analysis is better than not using them, irrespective of how inherently flawed they may be. Admittedly, my rationale was largely based on seasonal league analysis, falling back on the following adjustments:

  • A player projected to hit 20 homers for the season with a neutral home park should hit 22 or 23 if his home park favors homers and 17 or 18 if it depresses power.
  • A player projected to hit 30 homers for the season with a neutral home park should hit 33 or 34 if his home park favors homers and 35 or 36 if it depresses power
  • A player projected to hit 40 homers for the season with a neutral home park should hit 44 or 45 if his home park favors homers and 35 or 36 if it depresses power

Note these assume an extreme park factor in either direction - more on that in ensuing postings. Most factors are +/- 10 percent from neutral meaning for every ten homers projected to hit at home, a player would gain or lose one depending on the park.

In terms of seasonal play, where a typical active roster has 13 or 14 hitters, the flaws would even out and it wasn't worth losing sleep over. I ultimately concluded the best course of action was to just plug them into my spreadsheet and lose sleep over far more pressing issues.

The logical extension to DFS is if the difference between a player hitting half his games in Coors Field and half in Petco Park is six or eight homers, when you distill it down to one game out of 162, how much of an effect does the park exert? Those that delve out advice must feel it is relevant as it's an integral aspect of their analysis, mostly in an anecdotal as opposed to quantitative sense.

So here's where the wheels in my head started turning and the whole notion that regardless of the impact, it's fueled by a faulty input bothers me. My head has been able to justify their use but in my heart, I know park factors are fraught with error that we ignore, much the way we rationalize the questionable actions of a close friend.

Believe it or not, I've never been to a therapist but I've watched ample TV to know when you have a problem, or in my case, an obsession, it's best to talk things out. I'm obssessed with how the fantasy industry can rely so much on an entity that in my mind is so unreliable. As such, beginning now and extending until I'm satisfied I'm properly incorporating park factors into my analysis, writing and game play, I'll be using this space as my therapist's couch, so to speak. Don't worry, we're not talking months but we're not talking days either.

We'll begin with an introduction to park factors, what they try to accomplish and some of the factors that contribute to their effect. Next time, the discussion will explain why I've labeled them as flawed and unreliable despite their obvious acceptance and application throughout the industry, present company included.

In short, park factors, or park indices as they're also called, are designed to flesh out all bias and quantify how a park will play in comparison to a neutral environment. The most common factors are for runs, hits and home runs but every stat has its own index. A couple of the more important examples that are often overlooked are strikeout and walk indices.

The dimensions of a venue are obviously integral to how it plays but there are several other physical factors, listed here:

  • DIstance to outfield walls
  • Height of walls
  • Amount and placement of foul territory
  • Batter's Eye
  • Lighting for night games

There's also a bevy of weather related influences:

  • Temperature
  • Humidity
  • Barometric pressure
  • Altitude
  • Wind

These don't just affect the manner the park plays but they can also impact how well a pitcher can grip the ball and the pitch's movement. Temperature, humidity and altitude can have a direct effect on ballplayers - how they feel, how loose they can get, affecting their stamina, etc.

Somethng as seemingly esoteric as the umpires rotation can influence the statistics dictating how a park plays.

The missing element, of course, is the quality of both the home team and that of their collective opponents. This will be discussed in greater detail next time but at least on paper, the calculation is designed to flesh out this bias, as well as everything else already discussed.

NEXT TIME: Six-year look back at the variance with runs and home runs indices