K or 4-3

[1992casey]

---

Players with a lower contact ratio generally have a lower SLG.

---

Don't you mean "higher SLG"?

---

Russ typed it right. Quoth his study:

• Not a very high correlation, but there's a slight trend for batting average to increase with contact rate.
  
• Better correlation than BA and as expected, OBP increases with a decrease in contact rate. This makes sense, since players that strike out a lot often times have their share of walks as well.
  
• SLG increases with decreasing contact rate. It also correlates better with contact rate than either BA or OBP.
  
• OPS correlates better with contact rate than anything else. There's a clear inverse relationship between contact rate and OPS.

This is very important: Russ mentions that "correlation does not imply causation". The results simply indicate that SLG and contact rate are "interconnected".

[dadofandrew]

OPS correlates better with contact rate than anything else. There's a clear inverse relationship between contact rate and OPS.

 

I believe this is why I noticed Ron Shandler's (among others) forecasts are so awful, because he uses "increased contact rate" as a positive trait among hitters, while in reality, just the opposite is true.

[Ennder]

Shandler doesn't really use contact rate except for BA anymore. He even has a blurb in the book talking about how he used to be mistaken about how big an effect it has. There is a good correlation between contact rate and batting eye compared to batting average.

[rluzinski]

You can define contact rate a number of ways. I choose to use (AB-K)/PA, although most would probably do it as a function of total AB. Anyway, doing it my way there was a positive correlation between BA and contact rate but it was very weak.

[BrewCityUnit]

A groundOUT by definition results in an out. The "strikouts don't matter" camp has always compared a strikeout to a BIP out. You are confusing that by comparing a strikeout to a BIP who's result is still unknown. It's obviously better to have a chance of reaching vs. none, but that is not what has ever been argued. It's this huge misunderstanding that has caused countless unecessary arguments at brewerfan.net.

 

You can't have a chance to ground out if you don't make contact, so therefore I think making contact is better than striking out.

[BillScottCanRake]

You can't have a chance to ground out if you don't make contact, so therefore I think making contact is better than striking out.

 

And making authoritative contact is better than making contact for the sake of not striking out. And swinging hard enough to make worthwhile contact might result in striking out, but in the end will result will often be fewer outs.

[Member]

From a statistical standpoint:

 

GIDP's have a lower correlation with winning percentage, at least over the last half-century, than strikeouts.

 

GIDP's correlation with winning percentage, on a team-by-team basis between 1956 and 2005, of -.039. Strikeouts by batters have a correlation of -.103. All outs by batters (at bats less hits, plus sacrifices, caught stealing and GIDP's) have a -.046 correlation.

 

For the statistically challenged: correlation is a really fancy measure of how well two groups of similar statistics (like, for example, runs and wins) "match up" with each other. This measure, called a "correlation coefficient", is a number between 1 and -1. The closer the number is to either 1 or -1, the higher the correlation between the two sets of data. The closer the number is to zero, the less correlated the two sets of data are.

 

For example, doubles have a higher correlation to winning percentage (.187) than triples (.073) over the last 50 years. On-Base percentage has the best correlation with winning percentage during that time frame (.485).

 

Back to the discussion: a groundout is only about half as bad as a strikeout when you view it in terms of the ultimate goal - winning. That should be intuitive, as has been pointed out earlier - more things can happen if you put the ball in play, as opposed to not doing so.

[endaround]

Actually that is bad statisitcs. GIDP assumes a man is on first. Better teams have more men on base hence more opportunities for GIDP.

[rluzinski]

You can't have a chance to ground out if you don't make contact, so therefore I think making contact is better than striking out.

 

What you quoted and what you responded to aren't the same thing at all. There continues to be a huge amount of confusion over what's even being argued.

 

These arguments always start with a comment along these lines:

 

"Yeh, he hits for average and power, but he strikes out too much." The inevitable reponse is, "He can strike out all he wants if he's going to keep maching the ball." The logic is simply that if you conpare two players and each have a .300/.350/.500 line but one has 50 more Ks, the difference is marginal. The K hater then responds with,"But if he didn't strikeout as much he'd be a better player." While true, there's two problems with that notion:

 

A. One can never simply trade strikeouts without losing something else, be it walks, power or both. Ask Hardy.

 

B. If that player could theoretically K less and it theoretically raised his stats, hen the two players wouldn't be equal anyway!

 

The argument from the "Ks don't matter" camp is that a strikeout is the same as any other OUT. That's it. Arguing about what might happen if a player who strikes out alot tried to make more contact is a completely seperate issue.

[rluzinski]

Actually that is bad statisitcs. GIDP assumes a man is on first. Better teams have more men on base hence more opportunities for GIDP.

 

You beat me to it.

[mothershipconnection]

Russ typed it right. Quoth his study:

* SLG increases with decreasing contact rate

That's what I'm saying: as one goes down the other goes up.

Edit: I see he went back and fixed it. Russ, don't make typos when dealing with graphs and the like, as it makes the rest of us feel like we're taking crazy pills and can't understand the graph. I know you just did it to confuse me.

[rluzinski]

Russ, don't make typos when dealing with graphs and the like, as it makes the rest of us feel like we're taking crazy pills and can't understand the graph. I know you just did it to confuse me.

 

Sorry, I make mistakes like that all the time. http://forum.brewerfan.net/images/smilies/smile.gif

[Dollarbill13]

Where is the thread about flyball outs vs. groundball outs?

[1992casey]

---

Where is the thread about flyball outs vs. groundball outs?

---

Are you suggesting that there used to be one or that there ought to be one?

[rluzinski]

The flyball and groundball thing is interesting. FB pitchers have lower BABIP but since HRs are pretty much a function of FBs, GB pitchers have the benefit of giving up less HRs.

[gregmag]

Russ, that last big post of yours was great. I've tried to explain that distinction before, but you really nailed it.

 

One thing I wanted to throw in is that some kind of curve must define the relationship between contact rate and productivity. The inverse relationship between contact rate and OPS is very interesting, because it's so counterintuitive. The intuition -- that contact is good -- isn't completely wrong, is it? A guy who never makes contact won't be productive. So there has to be a tipping point at which contact becomes a problem. I'm just wondering if anybody knows where that point is, to flesh out the discussion of contact rates.

 

Greg.

[DuWayne Steurer]

not to beat a dead horse, but I find it interesting that Schroeder is talking about how J.J. Hardy "really turned it around" in the 2nd half last year because (specifically) he makes contact.

 

I know we've been over it, but J.J.'s numbers went through the roof when he started striking out more. I'm not saying the strikeouts helped, I think it was just a byproduct of being more aggresive and trying to drive the ball, instead of dinking and dunking the ball around the infield.

[rluzinski]

but J.J.'s numbers went through the roof when he started striking out more. I'm not saying the strikeouts helped, I think it was just a byproduct of being more aggresive and trying to drive the ball, instead of dinking and dunking the ball around the infield.

 

I'm going to talk about that a bit in the next WOAHH SOLVD.

[brettac1]

Couldn't you also say that luck may have played a part in that, as well? He was making so much contact and got so few hits, and that luck turning around, coupled with his more pull-happy approach, turned him into Derek Jeter for the second half?

[heresrobin]

Do you think the opinion that strikeouts are so bad stems from what we were taught growing up? For instance, making contact and striking out less is more important...when you're still learning how to hit, defenses suck and the chances of double plays are almost nil, like when you were growing up playing little league, Babe Ruth, and high school ball. Just a random thought.

[DuWayne Steurer]

You could say that, but J.J. himself said he wasn't being aggresive at the plate, and just wasn't hitting the ball hard at all.

 

Luck very well MAY have played a part in his terrible average, but for Schroeder to say that "contact" was the reason of his turnaround, that just can't be said to be true. He was making a lot LESS contact in the 2nd half, yet his OBP, SLG and BA all took significant jumps.

[brettac1]

Yeah, I know that Schroeder is dead wrong - I'm right with you there. I'm just saying that it was a combo of those things that made him hit so good.

[rluzinski]

Couldn't you also say that luck may have played a part in that, as well? He was making so much contact and got so few hits, and that luck turning around, coupled with his more pull-happy approach, turned him into Derek Jeter for the second half?

 

No doubt luck played a part. Was are only talking about 100 AB pre-hardy breakout. Still, it's obvious the difference in the quality of contact and the numbers back that up.

[Member]

Actually, I did show that strikeouts have a greater correlation to winning than your average, run-of-the-mill "out" - most of which involve putting the ball in play.

 

Balls In Play Outs (AB minus hits, plus SH and SF and GIDP, minus SO) are actually neutral in correlation to winning percentage (.007 from 1956 onward). Strikeouts have a negative correlation of -.103 with winning percentage, which, even though it's not a huge correlation, it's a lot more significant than GIDP or triples.

 

A better measure of success is the number of plate appearances per out: PAO correlates at a very high +.495. with winning (since 1956), better than OBP (+.485) and only behind Batting Runs (+.676).

[rluzinski]

A better measure of success is the number of plate appearances per out: PAO correlates at a very high +.495. with winning (since 1956), better than OBP (+.485).

 

Outs = PA x (1 - OBP)

 

so:

 

PAO = 1 / (1 - OBP)

 

So winning correlates slightly better with the inverse of OBP than OBP? The difference between a low team OBP and high OBP is maybe .05 at most, PAO is essentially linear anyway. I don't get it.

 

Also, a lot of the correlation coefficients are so low that they may not even be statistically significant. You are doing it by year? So 50 years x ~25 teams = ~1500 data points?

 

Do you have a spreadsheet of this?