Visualizing Performance: Good AJ and Bad AJ

Written by Josh W. | 17 August 2010

Good AJ: 8/15 at KCR: 8 innings, 4 hits, 1 ER, 6 SO, 3 BB, 8/6 GB/FB ratio

good_aj_countour_plot

Red is bad for AJ, blue is good. The dark green is pretty much neutral.The lime green is slightly positive for the hitter. The small spot of red really far out of the strikezone is a hit by pitch.

In this start, AJ was very effective up in the zone. He really only got hit in one area, middle-out (from the perspective of a right handed batter).

Here are his pitch statistics, courtesy of brooksbaseball.net

Pitch Type	Avg Speed	Max Speed	Avg H-Break	Avg V-Break	Count	Strikes / %	Swinging Strikes / %	Linear Weights	Time to Plate
FF (FourSeam Fastball)	93.76	96	-5.54	10.58	49	30 / 61.22%	2 / 4.08%	-0.8929	0.405
SI (Sinker)	93.92	96.2	-9.61	6.74	20	11 / 55.00%	2 / 10.00%	-0.3992	0.405
CH (Changeup)	85.71	90	-1.64	0.68	7	3 / 42.86%	0 / 0.00%	0.0109	0.443
KC (Knuckle-Curve)	83.88	86.4	3.00	-2.08	27	20 / 74.07%	2 / 7.41%	-2.8634	0.453

*The velocity is a little off. Pitch f/x seems to have some issues in Kansas City.

What's notable here is that AJ actually threw more strikes with his curve than any other pitch. HIs curve also had much less movement that usual, probably allowing him to command the pitch better.

Bad AJ: 8/2 at TOR: 4.2 innings, 8 hits, 8 ER, 4 SO, 2 BB, 9/7 GB/FB ratio

bad_aj_contour_plot

Red is bad for AJ, blue is good. The tan color is almost neutral, though slightly trending towards the hitter. Keep in mind the red here describes a higher run value than the red in the first graph.

As you can see, pretty much everywhere ended up being either neutral or very hitter dominated. The Blue Jays demolished anything thrown on the inner half of the plate (from the perspective of a righthanded batter). However, if you look close enough, you can see a small blue sliver. AJ can sleep safely knowing he absolutely DOMINATED that dot in the strikezone.

Pitch Type	Avg Speed	Max Speed	Avg H-Break	Avg V-Break	Count	Strikes / %	Swinging Strikes / %	Linear Weights	Time to Plate
FF (FourSeam Fastball)	91.73	93.2	-4.98	8.46	29	19 / 65.52%	1 / 3.45%	-0.4587	0.405
SI (Sinker)	91.99	94.4	-10.60	3.74	30	23 / 76.67%	3 / 10.00%	3.5834	0.408
CH (Changeup)	86.40	89.3	-6.50	1.77	8	6 / 75.00%	1 / 12.50%	-0.5266	0.428
KC (Knuckle-Curve)	80.30	82.3	3.34	-5.67	27	10 / 37.04%	1 / 3.70%	2.2061	0.466

Of note here is the poor fastball AJ had on this day. His velocity was down about 1.4 mph and he had less movement than usual too (with the fourseam). He also struggled to command his curve.

Comparison

Thanks to Trip Somers' pitch f/x tool, we can see that AJ used different release points in these two starts.

Bad AJ:

release_points_bad_aj

Good AJ:

release_points_good_aj

As you can see in these graphs, his release points were significantly lower in his start against the Royals; most of his pitches seem clustered below 6 feet. In his bad start, his release points look to be much closer to the 6 foot mark. I don't know how much a difference of a few inches makes, but it's possible it plays a part in his command, which was much better in his good start than his bad start:

Bad AJ:

pitch_locations_bad_aj

Good AJ:

pitch_locations_good_aj

In his good start, you can see that his pitches almost run across a diagnol line, with a lot being up and in and down and away (from the perspective of a right-handed batter). In his bad start, there is no such consistent trend.

Caveat:

With this post, I'm not trying to determine the difference between "good AJ" and "bad AJ" (I also think his inconsistency is overblown); that's quite complicated and there are seeminlgy limitless variables. This is just one good start and one bad start. I just wanted to visualize the differences in the two starts using pitch f/x data.