First Pitch - Sabermetric Tuesdays:  Lefty/Righty Offensive Team Splits

Sabermetric Tuesday gives us an opportunity to approach fantasy strategies with a statistical focus.  A common theme in fantasy baseball, especially in head-to-head leagues, is the concept of "pitch-and-ditch."  This strategy calls for acquiring a free agent pitcher for one start and then dumping him to either free-up the roster spot or to acquire another pitcher.  The aim here is basically to accumulate strikeouts and wins and punt the ERA and WHIP scoring categories.

There are a multitude of factors that we consider to determine the best candidate in any given day:

• A Pitcher's strikeout rate - often measured by K/9 or K/I
• Run Support - Its tough to chase the "W," but in general, it is always better to have your pitcher offensively supported by a team like the Red Sox than the Padres
• Ballpark Factor - Pitchers scheduled to start in "pitcher-friendly" parks like Petco Park and AT&T Park compared to hitter-friendly parks like Citizens Bank or Wrigley Field
• Quality Start %'s - Earlier in the season, I wrote about the Quality Start and its predictive ability for the W... you can read it here
• WHIP - Usually a good predictor of a player's ability to keep runners off base and, ultimately, from scoring
• Upward / downward trending players - the "hot-hand" theory of picking players that have put together a recent string of good starts
• Opposing team's offense / starting pitcher - does it look like a good match-up?  Does the opposing pitcher have a good match-up as well?
  

We could make an exhaustive list, but those are typically the main ones that are considered.  With that said, I have been analyzing lefty/righty offensive splits as an additional factor and have had relatively good results.  This is the theory: as a dissenting factor when choosing a pitch-and-ditch player, take a look at the handedness of the pitcher and the opposing team's success against that hand.  Eg. If I'm considering picking up a lefty to face the Nationals, how well do the Nationals hit lefties versus righties?  Is the split significant enough for me to stay away from that pitcher and go with another option? 

The Analysis

Here's the link to the full list of data.  As you can see, it is sorted alphabetically by individual team's offensive performance against lefties and righties.  Totals and averages are subtotaled by split (a legend of the data points can be found here). 

I carved the data into %'s by team based on out-performing or under-performing MLB averages for Runs, HR, Avg, OBP, and OPS.  Here are the results:



Here's how to read this chart. As an example, the Diamondbacks score 37% more runs against lefties than the rest of the MLB average.

As you can see, the following teams perform below the MLB average (green shaded boxes) across the aforementioned 5 statistical categories when facing lefties: White Sox, Twins, A's, Tampa Bay, and Toronto.  Teams that tend to perform below the league average against righties are: Rockies, Royals, Angels, Dodgers, Padres, Giants, Mariners, and Nationals. 

The teams in the yellow boxes strike out, as a team, at higher averages than the rest of the league.  Against lefties, these teams are the Diamondbacks, Reds, Marlins, and Phillies.  Against righties, the teams to note are the Marlins, Padres, Indians, and Rockies. 

Its easy to conclude from this chart that the Padres are the weakest offensive team on this list.  They lead in underperforming relative to the rest of the league in just about every offensive category against both lefties and righties as well as K's.  The top teams to avoid are the Cubs, Red Sox, Rangers, and Yankees as they out-perform the rest of the league offensively, yet trend below the league when it comes to K's.

So, when you're considering pitch & ditch candidates this weekend, consider taking a glance at the chart above as a sanity check.  For instance, even if your best free agent option is pitching against a great offensive team like the Phillies, if you only need K's (and are willing to "give-up" on WHIP and ERA for the week), a spot starter against Philly might not be a bad option.   Good luck!  -- Joe