How Pitchers Are Attacking Gardner

A few games into the season, friend of the blog Brien Jackson noted that pitchers were beginning to throw Brett Gardner more first pitch strikes. He suggested that Gardner’s penchant for taking pitches early in the count would begin to hamper his effectiveness, and that he needed to show the ability to get hits on those pitches to keep pitchers honest. I took umbrage with this suggestion, noting that pitchers had not changed their approach between 2009 and 2010 despite the fact that Gardner had already established himself as a “bat on the shoulder” type of hitter in 2009. I also pointed out that, as Brien stated, Gardner could always adjust his approach:

Furthermore, even if pitchers do alter course in attacking Gardner, I am not so certain that this would represent a problem for Brett. As Joe Pawlikowski of RAB said to me:

“It’s not as though pitchers can just choose to throw him more strikes without facing other consequences. If they’re going to consciously throw him more strikes, it increases the chances that they’re going to make mistakes within the zone. That’s going to create other opportunities for Gardner.”

Gardner is a high contact, high BABIP hitter, and more pitches thrown in the strike zone should result in an increased hit total that should replace the walks he had been drawing under the old approach. Putting this all together, I do not think there is any evidence to suggest that pitchers are inclined to throw Brett more strikes than they did in the past, and if they do, I doubt it greatly impacts his effectiveness.

However, as this season continued to progress and Gardner continued to fall behind in counts by refraining from swinging until it was 0-2, I wondered whether he was in trouble offensively unless he made some adjustments. However, over the last week, he has entirely turned things around (10 walks and 7 hits in 24 PA’s), and the swing data suggests that not much has changed between 2010 and 2011:

Season	O-Swing%	Z-Swing%	Swing%	O-Contact%	Z-Contact%	Contact%	Zone%	F-Strike%	SwStr%
2008	26.0 %	48.5 %	38.2 %	75.4 %	96.5 %	89.9 %	54.1 %	59.6 %	3.7 %
2009	17.2 %	51.0 %	34.2 %	74.2 %	92.4 %	87.8 %	50.2 %	57.4 %	4.1 %
2010	18.2 %	44.7 %	31.0 %	74.8 %	97.5 %	90.6 %	48.3 %	56.2 %	2.9 %
2011	18.1 %	50.3 %	34.1 %	79.4 %	93.6 %	89.8 %	49.9 %	52.2 %	3.2 %

As you can see, pitchers are now actually throwing Brett fewer first pitch strikes (F-Strike%) and about the same number of pitches in the strike zone (Zone%) as they have in the past. So what happened to cause Gardner’s numbers to adjust like this? Well, one possibility is that we were simply dealing with too small a sample earlier in the year, as swing data takes some time to stabilize. Using small samples that are almost entirely based upon at-bats that came during a slump may skew the data to make a player look worse than he actually is. This is the likeliest explanation. (Note: The sample is still not exceedingly large, and I would be cautious about drawing any strong conclusions from the data at this point.)

However, it is possible to tease another explanation from the O-Swing% (Percentage of pitches a batter swings at outside the strike zone) and Z-Swing% (Percentage of pitches a batter swings at inside the strike zone) data, although I caution that there are concerns about measurement error with those metrics AND they take longer than basic Swing% (which measures the total percentage of pitches the batter swings at) to stabilize. One could argue that the adjustment process that Brien, Joe, and I discussed actually took place over the first few weeks of the season. Let’s lay out a possible timeline:

1) Pitchers began to throw Brett more strikes, which reflected in the data and caused Brett to struggle.
2) As such, he started to swing at more pitches inside the the strike zone, as his elevated Z-Swing% suggests.
3) In the 6 games before he heated up, Gardner homered 3 times.
4) Since that point, Gardner has walked 9 times in 24 at-bats.

There is certainly a picture to be painted here from the ebb and flow of the data, whereby Gardner began to react to pitchers throwing more strikes by swinging more in the zone (elevating his Z-Swing%), which led to some power, which caused pitchers to adjust back to their previous approach, causing the F-Strike% and Zone% numbers to return to expected levels. Although this seems compelling, I am doubtful of this conclusion for a few reasons. Firstly, the sample size argument made above is strong and does not require us to create a narrative or depend on anecdotal observation. More importantly, I have trouble believing that it took pitchers two years to adjust to Gardner the first time and then just a week for them to drop their new approach due to a few home runs. While I plan on keeping a close eye on these numbers all year, I expect that at the end of the season they will look reasonably similar to Gardner’s career swing rates.