Continuing yesterday’s studies about Power Game Scores. . . .I promised today to address the following questions:

What is the average Power Score?

            The average for 2010 was 48.7.  The average for the years 2000 to 2009 was 49.1.

            My intention, obviously, was to center the stat at 50.00.   Experimenting with a variety of values, this was as close as I could come to 50.00 while also achieving the other goals of the system.

Is the average Power Score about the same now as it was years ago?  How much has it changed

over time?

            It has changed.   It would be my preference, of course, to have a scoring system that was constant over time, but unfortunately that doesn’t describe the real world.   These are the averages for the seven decades of the study:

To what extent do Power Games tend also to be Quality Games?

            They do not.   Overall, games in which Power Scores are less than 50 tend to be more effective than games in which Power Scores are higher than 50.

            To start with, let’s take the 35 highest Power Scores of the 2000-2009 decade, and the 35 lowest, just because that creates something that kind of looks like a single-season record, but with impossibly extreme totals:

            Extreme games tend to be effective games.   The reason this is true is that the system gives the highest and lowest scores to games in which the pitcher has pitched the most innings, which makes the margins better than the means.   Let’s divide the starts 2000-2009 into quartiles:

            You can see here that the extreme games get better results than the mid-range outings.   However, when you combine the top two groups, the top 50% of games in Power Score yield a winning percentage of .487 and an ERA of 4.62.   The lowest 50% yield a winning percentage of .507 and an ERA of 4.49.

            This is true in all decades of the study; however, it is becoming dramatically less true over time.    In the 1960s the "power games"—the top 50%--had a winning percentage of .428, and an ERA of 4.08.   The bottom 50% had a winning percentage of .544, and an ERA of 3.20.  

            As baseball has become more of a power game, the power pitchers have generally caught up to the control pitchers.   But as of 2011, the "control" outings—as measured by this system—are still a little bit better.

Does a pitcher like Nolan Ryan throw 60% Power Games, or 70%, or what?

            Nolan Ryan had Power Scores of 50 or higher in 85.4% of his career starts—660 out of 773.    Surprisingly, this is not the highest percentage in the study; it’s fifth-highest.   But he has more starts than the top four guys combined.

Who are the greatest Power Pitchers in history, by this method?

            Among pitchers with 100 or more starts, these are the pitchers with the highest percentages of Power Starts, defining a Power Start as a start with a Power Score of 50 or higher:

Who are the greatest Control Pitchers?

            The number one control-type pitcher of the last 50 years, by this method, was Jim Barr, who pitched from 1971 to 1983, mostly with the Giants.   Barr, who pitched 230+ innings five times but never struck out 100 batters in a season, had only 15 starts out of 252 in which his Power Score was 50 or higher.

Are "Power Tendencies" stronger or weaker than "quality tendencies?"

            As measured by these two methods—the Game Score and the Power Score—the Power/Control tendencies are far, far stronger than the quality tendencies.   Jim Barr had Power Scores under 50 in 94% of his starts; Herb Score was 50 or over in 90% of his.    No pitcher is anywhere near 90% of his starts being either good or bad.    If you have good outings in 70% of your starts you’ll win the Cy Young Award; if you have bad outings in 70% you’ll get released.   Power/Control type is far more consistent than performance level.

            Got a few questions left.

To what extent is "Power" a function of youth?

To what extent is "Power" predictive of future performance?

Of what actual use in the Power Game Score?

            These are the tougher questions, the questions that can’t be dispensed with so quickly.   We’ll get to those tomorrow, in the final part of our three-part series.