Gordon and making contact

by kcdc1 on Apr 15, 2011 1:35 PM EDT

This isn't a real post so much as it is a conversation topic and a request for someone more capable to dig a little deeper.

Just before the start of the regular season, Jeff wrote up a profile of Alex Gordon's career hitting tendencies. There are reports that Gordon worked all off-season on recreating his swing to be more level and hopefully make more contact. It's really early still, but with the reports that he's changed his swing, the different looking results, and the appearance that it might look a little different to the untrained eye, it might be worth updating that profile and comparing this season's results to his career profile.

Just looking at fangraphs, Gordon's K% is down to 20% from a career average of 25%. If that's sustainable, I'm guessing it would add about 15-20 points to Gordon's batting average, making him a .265 to .280 hitter. With his power and walks, it seems cutting his K% to 20% would make him a much improved player.

Diving a little deeper, his contact rate is at 80.6% compared to his career rate of 76.4%. Given the small sample size, I'm not sure if that's statistically significant yet, but it's at least interesting. The contact figure that most jumps of the page is his 67.7% contact rate on swings at balls outside the strike zone. His career rate is 54.6%. Presumably, a lot (a majority?) of Gordon's strikeouts are on missed swings at balls out of the zone, so if that improvement is sustainable, it would go a long way in reducing Gordon's K%. Even in a small sample, I'd be surprised if that 13% jump isn't statistically significant. I'd also love to see those contact rates broken out by splits against lefties, against whom Gordon has particularly struggled to make contact in his career.

Maybe I'm fishing for any signs of life from Gordon, but the improved early contact rates look promising.

Tweet Comment 11 comments | 0 recs |

Do you like this story?

Comments

Any guesses what his K% will be at the end of the year?

I’m guessing the improved contact is at least mostly for real. I’ll go 21.5%.

by kcdc1 on Apr 15, 2011 1:56 PM EDT reply actions

He looks a lot better from last year

He had a hole in his swing

.......Just a beer drinking KC Chiefs & KC Royals fan.......
POWDER BLUE BOMBERS!!!!!

by Mas Cervezas on Apr 15, 2011 2:03 PM EDT via mobile reply actions

K rates
vs LHP 26.1%
vs RHP 15.6%
I will take these in a heart beat

He has been killing lefty pitchers with some help from a high BABIP:
0.435/0.458/0.609 with a BABIP of 0.588

Almost 60% of balls put in play from LHP have turned into hits

- .-. ..- … – / – …. . / .—. .-. - .. . … …

by Jeff Zimmerman on Apr 15, 2011 2:04 PM EDT reply actions

In his career to date, it's been 25% and 25%

So while his overall K% is improved, against lefties, it hasn’t been better this year. The sample sizes on K% are too small to mean much, and when you break it out into splits, the sample sizes are even smaller.

I’d be more interested in contact percentage splits vs lefties and righties, where since you get a handful of pitches per PA, you have a larger, hopefully more meaningful sample.

by kcdc1 on Apr 15, 2011 2:19 PM EDT up reply actions

He's not striking nearly as much

But strictly speaking his contact on swings isn’t that much higher so far this season (80.6% as opposed to 78% last season — and that’s after last night — it was right at 78% before then). It’s an improvement, I guess, but sample size and all…

It’s nice, but I’m not going to get excited, because if I did that, I’d need to panic even more over the terrible walk rate (5.2%, although this may get him an extension with this team) so far, and his extreme ground-ball tendencies so far this season. Maybe the hope is to turn him into DDJ part II? Like I said, I’m not worried yet, either, but then I can’t get excited.

I have faith in Gordon, but I’m not going to go nuts and jinx him. Certainly, there is a lot of uncertainty and upside. This isn’t Jeff Francouer, where people get blinded by his age and Dayton Moore’s alleged ability to turn terrible players into good ones based on things that happened 5+ years ago and ignore a huge recent sample size of worthlessness. But I have to admit that while Gordon’s overall results have been nice so far, if we’re going to ignore sample size, there are at least as many things to worry about as to celebrate.

Making watching baseball as fun as doing your taxes.
My Twitter feed.

by Matt Klaassen on Apr 15, 2011 2:43 PM EDT reply actions

In addition to my response below

Gordon’s contact rates were also improved last year compared to his career averages. There was talk of him working to level out his swing during his AAA stay last year, so maybe the 78% from last year can seen as early returns on an effort to make more contact. His O-Contact% was also improved last year at 61.6%, which also seems support the case that Gordon was reworking his swing to make more contact. Of course, in 2010, Gordon struck out 25.6% of the time which is more or less in line with his career average, so make what you will of the 2010 numbers. I have a hunch that as the year continues, the improved contact story line will gain more traction.

by kcdc1 on Apr 15, 2011 3:09 PM EDT up reply actions

Reply to both

I’ve found that Contact and Swing rates actually correlate better to actually performance than O-Contact and O-Swing.

His Swing rates are way up, too (this correlates negatively with walk rates). The sample sizes, relatively speaking, are pretty much the same compared to what they are measuring.

Moreover, if you want to go off of the “level out the swing” theory of explanation…. well, that would also explain why he’s hitting so many worm-burners so far.

Again, my point is not to say Gordon is bad. Just that there’s at least as much negative evidence as positive regarding his “new approach.”

Making watching baseball as fun as doing your taxes.
My Twitter feed.

by Matt Klaassen on Apr 15, 2011 3:25 PM EDT up reply actions

The swing rate is a totally separate issue from the contact rate. If we’re focusing on whether Gordon has reworked his swing so that he doesn’t miss the ball as often, his swing rate won’t tell us much.

On the other hand, an increase in swing rate would lower both his K% and BB%, so it’s possible that Gordon has just had a more aggressive approach at the plate this year. Cutting strikeouts by not going as deep in counts and thereby reducing walks isn’t going to help his value much. That said, I’m not patricularly worried about Gordon’s patience at the plate since he walked a ton in spring training.

It seems to me that Gordon had a pronounced upper-cut in his swing in prior years, and that the swing motion resulted in him being well below average in making contact on balls out of the strike zone. Jeff mentioned in his article that his career plate discipline rates were in line with MLB averages with the exception of O-Contact% which was ~8% below MLB average. If Gordon has worked to eliminate the upper-cut in his swing, the O-Contact% seems to be a good place to start looking for changes.

by kcdc1 on Apr 15, 2011 4:01 PM EDT up reply actions

Sure, on the contact issue

1you may be right, then. I know that was your point. On the other hand, I was addressing the broader issue that if we look at his offense generally (and all of this ignores the small sample, of course), there’s are as many or more reasons to be pessimistic about his overall approach. Indeed, I would be more excited about the increase in O-Contact if it wasn’t accompanied by a discouraging and massive increase in O-Swing.

And if the new swing is generating a bit more contact, it is at least as closely connected with a big increase in ground balls, which is nice for a few more balls in play, but generally is not a good thing for hitters other than guys who have no hope other than slapping it into play and hoping they get lucky (e.g., Escobar).

Making watching baseball as fun as doing your taxes.
My Twitter feed.

by Matt Klaassen on Apr 15, 2011 4:50 PM EDT up reply actions

Regarding sample

The sample sizes for K% and BB% are measured in plate appearances, while the sample sizes for contact rates are measured in number of swings. We’ve got a bigger sample for the latter. As for 80.6% this year compared to 78% last year and 76.4% for his career, I doubt that the uptick is statistically significant at this point. But the jump in O-Contact% from 54.6% in his career to 67.7% this season very well could be. How many swings are in the sample?

We can look at each data point on its own to determine whether its sample is significant. The K%, BB%, and GB% likely have not stabilized, and I doubt we have statistically significant evidence to believe that they’ve changed. The O-Contact% might be a different story.

by kcdc1 on Apr 15, 2011 2:59 PM EDT reply actions