Fangraphs Piece on Strikeouts
There was a fairly long discussion last week on how important strikeouts were for hitters. Serendipitously, Steve Slowinski addresses it in a good post here.
10 months ago
jessef
19 comments
0 recs |
Email
PrintComments
heh, it is pretty funny
how closely this resembled our discussion.
this new predictive stat is pretty cool too: ShH or Should Hit. basically attempts to be a fielding independent hitting predictor
WAR... WAR never changes.
Excellent article
The logic is so straightforward, and the arithmetic so simple, that I’m surprised it wasn’t done many years ago, or at least shortly after the BABIP observation was made.
One other point: in his evaluations of minor-league hitters, Woodman usually looks at K rates, and downgrades a hitter if he has a high K rate. The idea is that this usually indicates holes in his swing. These will be exploited as he moves up to higher levels, leading to lower offensive output. I wonder if the same logic can not be applied, at least in part, to young major-league hitters (Travis Snider?). Will the pitchers not eventually exploit this weakness?
As was brought up in the thread here.
Strikeouts have a high correlation with power stats. If you strike out a lot and don’t hit for power than yes you are a pretty bad hitter. Like some of the fangraph post responses say, this is nothing too new here.
that wasn't what the discussion was about
the argument was that a K is equally bad as a ball put in to play
I demand to be heard.
you never responded to my comments debunking that sentiment
in response to http://www.bluebirdbanter.com/2011/8/4/2344855/brett-lawrie-called-up-to-blue-jays#73893167, I said:
However, you’ve ignored the possibility that power and strikeouts may be correlated because players need to be productive at the plate in some way. If they strike out a lot, it means they’re not going to be productive hitting for average so they either need to take a lot of walks or hit for power.
Furthermore, you also used one of the worst examples of the correlation between power and K-rate you could find. Bautista actually demonstrates an inverse correlation between power and strikeouts.
Prior to 2010, Bautista had an isolated power of .161 (exactly league average power) and struck out roughly once every five plate appearances (slightly about league average K-rates). In 2010, when Bautista really started to show power, he cut his strikeouts down to 17% and this season has cut his K’s further to below 15%.
"Look at me! I'm Tomokazu Ohka of the Montreal Expos!"
Anecdotal evidence doesn't really work
There have been statistical studies like I said (and linked) which correlate the 2. Yes Bautista has dropped his K rate while increasing power. That is very valuable (see Pujols) and very rare. But in general power comes with higher K rates.
As Dave Cameron said in a chat last month:
Comment From Mike
Do you think there is a disconnect where ‘stat guys’ value pitchers strikeouts and largely ignore hitters strikeouts?
Dave Cameron: Nope. For hitters, strikeouts are heavily correlated with power, so the more a hitter strikes out, the more likely he also is to hit a lot of home runs and be a valuable performer. That is not true of pitchers – high K pitchers don’t give up more home runs as a result.
Comment From Mike
So based on your strikeout/power argument, Austin Jackson should be one of the biggest power bats?
Dave Cameron: I didn’t say strikeouts = power, I said that most hitters who strike out a lot also hit for a lot of power. Obviously, a guy who strikes out a lot and doesn’t hit for power won’t be very good, which is one of the reasons why we’ve been far more down on Jackson than most mainstream analysts.
Comment From PR
“strikeouts are heavily correlated with power, so the more a hitter strikes out, the more likely he also is to hit a lot of home runs and be a valuable performer.” You make it sound like strikeouts help your power. Isn’t the reality that if you strikeout a lot, you better be able to hit for power (or do something else great) or you arent in the majors
Dave Cameron: It’s a sliding scale. Guys can either choose to take big cuts that make the ball go far when they make contact at the expense of making less contact, or they can try to hit the ball as often as possible at the expense of hitting it hard.
So yes Pikachu a ball in play has a slightly (I think there is some statistical evidence that it is a few percentage better) to be useful than a strike out. But like Cameron said, there is heavy correlation between K rate and power numbers, and that is what mainly drives the narrative that strikeouts are no worse than other outs from a saber perspective.
The anecdotal evidence was yours
I merely pointed out that it made no sense.
Has it ever occurred to you that Dave Cameron could — gasp — be wrong about something?
Please look at this, then continue the argument.
http://www.tangotiger.net/strikeout.html
"Look at me! I'm Tomokazu Ohka of the Montreal Expos!"
I've read that.
And while sure Cameron could be wrong, and so could the baseball prospectus writer and other writers could be as well.
But that study you link is hardly conclusive. They talk about the hitters with the biggest drops in K rates and the ones with the biggest rise in K rates. Then they take the 100 biggest offenders on each side.
So a player like Alex Rodriguez, who has had between 97 and 139 Ks in every season would never be on the list. He’s been too consistent to even be part of the sample. Or Sammy Sosa, who was remarkably consistent in K’ing about 1 per game over his career and never changed his approach.
Since the sample is so narrowly limited to people who changed their K rates in either direction the study is inherently flawed. So yes you’re going to have something a bit more conclusive than a post from TangoTiger to make me disregard the opinions of Dave Cameron and others.
Tango's study demonstrated the relationship between
increased strikeouts and increased homeruns but also demonstrated that the increase in homeruns occurred at the expense of doubles and triples; thus, the impact of homering more was outweighed by the drawbacks of striking out more.
Please link a study demonstrating that the increase in homeruns outweighs the increase in strikeouts. I understand that “Dave Cameron and others” are of the opinion that it does, but Dave Cameron was also of the opinion that the Seattle Mariners and the Minnesota Twins were the sixth best organizations in MLB in 2010 and 2011, so excuse me if I don’t think that he’s infallible.
"Look at me! I'm Tomokazu Ohka of the Montreal Expos!"
#6org
And apparently Tom Tango is of lesser repute than is Dave Cameron. Who knew!
"Upon further evaluation, though, the Blue Jays reversed course. Doctors suggested that the inconsistent workload in the bullpen would increase the chances of McGowan suffering another injury. "
And I went and looked at the tango post you mentioned in the other thread
And yes Pikachu is right that it is slightly more likely to result in something good happening. But the reason that drives the narrative of “A K is as good as any other type of out” is what Dave Cameron talks about in the strong correlation between power stats and strikeout rates.
So this balances out. A K is slightly less likely to be productive than a ball in play, but players who K more tend to be more productive than players who don’t.
you were fine until
But players who K more tend to be more productive than players who don’t.
Of the top 35 players in wOBA, 5 K over 20% of the time, while 16 K under 15% of the time (including 2 under 10%)
(The proportions are almost the same for wRC+)
"Upon further evaluation, though, the Blue Jays reversed course. Doctors suggested that the inconsistent workload in the bullpen would increase the chances of McGowan suffering another injury. "
a ball in play
isn’t “slightly more likely” to result in a good outcome. a strikout is an out over 99% of the time. a fly ball is an out 78% of the time.
furthermore, I think you are putting a lot of stock in correlation, though there’s little reason for the effect to be causative.
WAR... WAR never changes.
Even the article that Jessef linked
Shows that a strikeout is only a few percentage points worse than another out without taking into account the correlation that exists between Power and K rate.
as jessef has said many times
the argument is very flawed since players who are in the MLB and strike out a lot HAVE to be good at hitting for power and taking walks, else they’re not in the MLB. striking out more probably doesn’t make you hit for more power. being an MLBer who strikes out a lot necessitates hitting for power
WAR... WAR never changes.
i think we're arguing different things here
you keep bringing up the “players who K a lot are power hitters”, but that wasn’t the argument. The argument was that a K is equally as bad as a ball put into play for an out, which its not. the difference might be miniscule just like slowinski says, but its there, and you denied that.
I demand to be heard.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
