MLB Advanced Analytics - Prediction Machine

MLB Advanced Analytics

Marlins Baseball

As a sport, baseball is loaded with statistics and words that have become household terms even for non-sports fans.

For instance, one might say they “struck out” during a job interview or their “batting average” fixing items around the house is low. These stats are so commonly known that they’ve transcended America’s Pastime, and even your brother-in-law who knows nothing about sports, has a basic understanding of them. Staple stats like these, as well as others like ERA and slugging percentage, have some use to handicappers, but partly because this data is so well known and easily accessible to the public, its value is limited.

Then, on the other side of the spectrum, there are advanced analytics. These are more fringe baseball statistics that even lifelong baseball fans might not be familiar with. These formulas can help fill in the gaps left by standard baseball stats. Professional bettors are constantly trying to see a clear picture of a baseball team or player, and using advanced analytics is like graduating from a black and white screen to high definition color. These metrics help make sense of the more basic stats and give handicappers, sportsbooks, and baseball executives insight into what’s really going on.

Below, I’ll introduce you to three advanced baseball analytics I lean on to examine starting pitchers while handicapping games. Here’s a couple disclaimers before we get started:

1.) These analytics can be trusted more when they include more data. A little more than 20% through the MLB season, 2023 analytics are less predictive this time of year than they will be in September.

2.) This article isn’t meant to be all encompassing, it’s meant to be an introductory primer. The world of advanced analytics is deep. I’m talking, Matthew McConaughey’s voiceover in a car commercial deep. We couldn’t possibly include everything in this limited space, and the intent is to make it as easily digestible as possible.

GB % – Ground Ball Percentage for pitchers is calculated by dividing the number of ground balls induced by the number of balls in play. This equation can be especially useful to handicappers when there are outside factors that might cause more home runs in a game than usual. For instance, a hitter friendly ballpark, or strong winds blowing out. Let’s say there is a game in Colorado with 16 MPH winds blowing out to straight-away center field and starting pitcher A has a 65% GB% and starting pitcher B has a 45% ground ball %. Pitcher A may be more likely to find himself in trouble, especially vs. a lineup with a lot of home runs. Ground balls are also more likely to induce double plays and help pitchers out of jams with men on base.

BABIP – Batting Average on Balls in play is calculated like batting average, but it removes sacrifice flies, strikeouts and home runs. By removing those outcomes, we get a clearer picture of how often a ball lands for a hit when it’s in the field of play. While there are many ways a bettor could use this stat for batters and pitchers, I use it to see which starting pitchers might be regressing toward the mean. In other words, the league average for BABIP is roughly .300, so if a pitcher has been struggling this season and has an abnormally high BABIP, they may have been a bit unlucky thus far. Keep in mind, there are exceptions with this stat. For instance, with really skilled pitchers like Clayton Kershaw who consistently generate weak hit balls, he’s less likely to regress toward the mean. Kershaw’s BABIP is consistently well below .300 each season so it’d be a mistake to assume that number will suddenly trend upward causing him to struggle.

xERA – The Expected Earned Run Average calculation is more complex than Ground ball % or BABIP. The formula utilizes factors like exit velocity of batted balls and launch angles to determine the Expected Weighted On Base Average (xwOBA) and then converts it to an ERA scale. As a bettor, I like to compare a starting pitchers ERA to their xERA to get a clearer picture of how they’re performing. If these numbers are pretty close together, then I move along. However if there is a large gap between these numbers, it prompts me to look deeper and see how that gap can be explained.

If you’re new to the world of advanced analytics, these terms and the alphabet soup can be intimidating and confusing at first. The more you incorporate them into your handicapping and your daily analysis language, the more it becomes second nature.