THE BOOK--Playing The Percentages In Baseball

Tuesday, September 23, 2008

The zero-level player

Here’s one thing everyone can try:

1. Go here: http://tangotiger.net/markov.html
2. Click CALCULATE
3. You’ll see this: 4.905 : Runs Scored per Game
AVG / OBP / SLG: 0.270 / 0.341 / 0.405
4. Take 8/9ths of that, which is 4.36
5. Click the BACK button, and change the value of AB, to something, and then click CALCULATE, and keep revising the value of AB until y ou get 4.36 runs scored. I’ll save you the trouble and tell you to put in AB=38.985. You get these results:
4.360 : Runs Scored per Game
AVG / OBP / SLG: 0.257 / 0.326 / 0.385

So, what did we just do? Well, we constructed a team of players that scores 4.36 runs per game. But, originally, we had a team of players that scored 4.905 runs per game, meaning that each one “created in isolation” 0.545 runs per game. 8 of those guys “creates” 4.360 runs per game.

Their OBP was .341. And if we take 8/9ths of that, and presume the 9th guy has a .000 OBP, the new average would be .303.

However, we just figured that for a team to score 4.36 runs per game, their OBP would need to be .326. In order to get that, you take 8/9th of .341, and 1/9th of .206.

What we have therefore is a guy with an OBP of .206 added to a team of 8 guys with an OBP of .341 to get you a team average of .326. And that team will score the same amount as 8/9ths of an average team.

In effect, the guy with the .206 OBP added “zero” runs. Indeed, all his positive contributions was undone by his negative contributions, as whatever he added with his hits and HR was wiped out by all his outs that undid all the good work by the other 8 players.

Now, go back to my link. Change the AB so that you get an OBP of .206. (Change AB=63.96). You will see a team of 9 such “zero-type” players scoring:
1.550 : Runs Scored per Game

So, by this process, a team of “zero-type” players will score 1.55 runs per game (roughly one-third of an average team). Such a team, given average fielding and average pitching, will win roughly 12-15% of the time.

Now, let’s try to be symmetrical about pitchers as well. In this illustration, the “zero-type” hitter has an OBP of .206 in a league of .341. The Odds Ratio method would say that the defense would have to have an OBP of .508 to be “equivalent”. That is, a team of hitters hitting .206 facing a team of pitchers allow .508 will result in a .341 OBP.

Go back to my Markov, and set the AB=23.56

You get: 14.733 : Runs Scored per Game

That number is 3 times the average.

(Nice symmetry right? The “zero-type” hitters are one-third of league average, while the “zero-type” pitchers are three times league average.)

The winning percentage for a team of zero-type hitters with zero-type pitchers is a bit less than 1% of games won.