Thursday, June 28, 2018

BABIP and how it should be used in handicapping


BABIP: How can we use it in handicapping

The term BABIP is the batting average of batted balls in play. The MLB BABIP average is .298 so we are going to compare pitchers to whether they are above or below that average. 

The idea that BABIP can by itself be a deciding factor is not very accurate. In fact, by itself has very little to the outcome of a game. Relating it to a mindset that everything has a median value and will eventually work towards that median, will be the prevailing use for BABIP. This is an old school thought but not totally lost in concept. Let's see how this works. If a pitcher has a BABIP of .328, he is above the median or "average" .298 discussed above. The prevalent thought is that he will regress back to the average of.298 thus lowering the BABIP. It means that this pitcher will get more outs on batted balls than he has so far. Makes some sense and conceptually is correct. Most handicappers that use BABIP stop here and this how they view run prevention capabilities of that pitcher. Again, that is not entirely inaccurate and can provide great insight into how the game will play out. The theory is that if you are above the average then you have been somewhat unlucky in where the balls were hit. The pitchers that are below the average are considered to have had better luck as the balls hit in play have gone for outs. I think this is not a complete picture. If you combine some other elements of run prevention, you might see that there are reasons for the high or low rates of BABIP and then get a complete picture that can certainly be utilized to ascertain the run prevention capabilities of the pitcher. Let's take this further to get a great handle on it.

Sabermetrics provides another stat called SwStr% (Swing Strike Percentage). It measures the swings batters take and miss the pitches thrown. The MLB average is 10.1%. This is an indication of the quality of pitches the pitcher can throw. Many people describe it as "stuff". You will hear an announcer say "that guy has great stuff". If a pitcher has a below average SwStr% plus a high BABIP, that means his stuff is less quality and the batters are getting good contact with him. Thus, the high BABIP is deserved and not unlucky. It should continue with some variance but still above the average. The opposite is also true. If a pitcher has a high SwStr% and a low BABIP, then his stuff is causing hitters to make poor contact that produces outs. Now let's see if we can go a bit further.

Let's bring in GB% (ground ball percentage). There is a correlation that ground balls produce fewer runs. Although a true statement, ground ball percentage is not by itself a key element to how many runs will get scored. If a pitcher has a high GB%, with a low BABIP and a high SwStr%, then he gets balls in play hit on the ground for outs and batters do not make good contact. This means a lot! It takes 2 hits to score 1 run in MLB. However, if the balls in play are in the air, there is greater likelihood that they will be extra-base hits and may reduce the number of hits required to score. Big edge to those who can keep the ball down. Let's add one more.

We will bring into play hard hit percentage. This measures how often a ball in play was hit soft, medium or hard. The more hard-hit balls, the less quality of pitches the pitcher has. The batters are not getting fooled and are making hard contact. The average is 31.4% in MLB. If a pitcher has a high BABIP, with a low SwStr%, and a low GB%, plus a high hard-hit rate, then he should be set for lots of scoring against him. The balls in play are earned for hits, they are hit in the air, and they are hit hard which will produce extra-base hits and more run-scoring opportunities. 

When we combine measurables that relate to the pitcher's ability to prevent run scoring, we can get a complete picture of the overall ability of the pitcher. BABIP has tremendous value and should be looked at to help determine outcomes of games. It can stand alone, but without some help, we could make judgments that are misleading to complete picture. Make sure to analyze the entire package and not just a component of it.
  
  

No comments:

Post a Comment