Posted by MikeT23 on 6/27/2016 7:52:00 AM (view original):
It was probably the managers' call. Stupid managers. BL would be a much better manager because he knows a hit is better than a sacrifice fly.

A lot of sac bunts were, historically, bad decisions. Obviously the kind of pitcher who hits .160 with no walks is a much stronger bet to lay down a sac bunt than do anything else useful, so it's not always a bad decision. But sac bunts used to be way too common, and they've been dropping basically forever. Here's a chart of team sac hits over time:



Most of the data before the mid '20s is off this scale, but to see that it's actually been continuing to drop persistently since that time I needed to shrink the graph. It's pretty obvious that as strategy has evolved over time it's become increasingly obvious to actual baseball people that throwing away any of their 27 outs to move a guy 1 base isn't necessarily a smart strategy. Outs are your one limited offensive resource in this game. It's usually not a great idea to waste them.

Posted by MikeT23 on 6/27/2016 12:03:00 PM (view original):
So, really, you can't say "It's not just the steroid era. It's 1920-on" and have a legit argument.

You're talking apples and oranges. The run values of events change depending on the scoring environment, but they change in both directions.

In lower run scoring environments, hits are less valuable, outs are less costly. In higher run scoring environments, hits are more valuable, outs are more costly.

Evaluating the relationship between team runs and team K's (or hits or OBP or whatever) can be done on a broad scale and should be done on a broad scale.

Also, while the correlation is very weak, over the period 1990-2015 there is a statistically significant negative relationship between team SH and team runs.

This from the guy who said every front office in baseball agrees with him that a sacrifice fly is a bad outcome for his team.

Posted by Jtpsops on 6/27/2016 12:14:00 PM (view original):
This from the guy who said every front office in baseball agrees with him that a sacrifice fly is a bad outcome for his team.

What do you think dahs is saying in the post above yours?

Posted by dahsdebater on 6/27/2016 12:08:00 PM (view original):
Also, while the correlation is very weak, over the period 1990-2015 there is a statistically significant negative relationship between team SH and team runs.

Question for you...

If something has a very weak correlation, say r=0.25, the fact that it's statistically significant just means that there isn't a sample error, correct? It doesn't mean we should view the correlation as more predictive, right?

Or has it been too long since I took statistics in college?

^ genuine question, not an attempt at any sort of gotcha.

Posted by bad_luck on 6/27/2016 12:49:00 PM (view original):

Posted by dahsdebater on 6/27/2016 12:08:00 PM (view original):
Also, while the correlation is very weak, over the period 1990-2015 there is a statistically significant negative relationship between team SH and team runs.

Question for you...

If something has a very weak correlation, say r=0.25, the fact that it's statistically significant just means that there isn't a sample error, correct? It doesn't mean we should view the correlation as more predictive, right?

Or has it been too long since I took statistics in college?

That's basically right.

In this context, the statistical significance is a test of whether something impacts run scoring at a certain confidence level. The regression matters how much it impacts it.

The R^2 for SH is like .015. It's a very small factor, which you'd expect, given that very few sac hits even happen. But there is a negative relationship between sacrifices and runs.

Posted by bad_luck on 6/27/2016 12:09:00 PM (view original):

Posted by MikeT23 on 6/27/2016 12:03:00 PM (view original):
So, really, you can't say "It's not just the steroid era. It's 1920-on" and have a legit argument.

You're talking apples and oranges. The run values of events change depending on the scoring environment, but they change in both directions.

In lower run scoring environments, hits are less valuable, outs are less costly. In higher run scoring environments, hits are more valuable, outs are more costly.

Evaluating the relationship between team runs and team K's (or hits or OBP or whatever) can be done on a broad scale and should be done on a broad scale.

In low scoring environment, when runs are at a premium, wouldn't runs (and therefore the events that produce runs, such as hits), be more valuable?

Posted by tecwrg on 6/27/2016 1:37:00 PM (view original):

Posted by bad_luck on 6/27/2016 12:09:00 PM (view original):

Posted by MikeT23 on 6/27/2016 12:03:00 PM (view original):
So, really, you can't say "It's not just the steroid era. It's 1920-on" and have a legit argument.

You're talking apples and oranges. The run values of events change depending on the scoring environment, but they change in both directions.

In lower run scoring environments, hits are less valuable, outs are less costly. In higher run scoring environments, hits are more valuable, outs are more costly.

Evaluating the relationship between team runs and team K's (or hits or OBP or whatever) can be done on a broad scale and should be done on a broad scale.

In low scoring environment, when runs are at a premium, wouldn't runs (and therefore the events that produce runs, such as hits), be more valuable?

Maybe it's a little counterintuitive, but in terms of X event was worth 0.Y runs, lower run scoring environments produce lower values.

*by lower I mean closer to zero, positive or negative.

Rather than "counterintuitive", I'd go with "stupid".

I was trying to give you a chance to redeem yourself with something intelligent to back up your statement. Shockingly, you failed.

If something is in less abundance, such as hits or runs , then each one that you get is more valuable.

For example, in a high run scoring era, if it typically takes scoring five runs to win a game, then a single run only gets you 20% of the way there.

In a low run scoring era, if it typically takes scoring three runs to win a game, then a single run gets you 33% of the way there.

Runs are more valuable in a low run scoring era. It therefore follows that the hits that help you score those runs are also more valuable.

You're both right...

In terms of run value, when other hits are less likely to occur, each hit is less valuable. But in terms of game (win) value, its more valuable.

Posted by tecwrg on 6/27/2016 1:51:00 PM (view original):
Rather than "counterintuitive", I'd go with "stupid".

I was trying to give you a chance to redeem yourself with something intelligent to back up your statement. Shockingly, you failed.

If something is in less abundance, such as hits or runs , then each one that you get is more valuable.

Runs are more valuable. But we aren't measuring the value of runs. We're measuring the value of events and runs are the measurement we use.

Maybe try acting like an adult when you disagree with someone. It might get you a little further in life.

The reality is that the big reason why Ks become more important in low-run scoring environments is that the value of productive outs relative to standard outs changes much more slowly with the overall offensive environment than the cost of losing an out. So double plays become a lot less important when nobody's scoring.