Had a hypothesis, looked at the data, came to a new conclusion.

czman

Well-known member
Joined:
May 7, 2013
Posts:
2,210
Liked Posts:
545
For those who do not want to read the entire thing. SO= real bad.
I think it is somewhat important that I lay the ground work for this. I had this idea that the outfield defense was more important now than ever before. There were two main reasons for this.

1) With the prevalence of the shift the need for great defense would in theory be diminished. You can’t really shift too much in the OF because there is too much space and not enough players. If you shift yourself into nonstandard position in the IF it means you are also leaving gaping holes for a ball to be hit that can easily turn into extra bases if the OF is slow. I feel like the IF shift puts more pressure on OFs and less on IFs.

2) With the decrease in PEDs there are fewer HRs. I thought there would be more balls hit in the OF though and that those fly balls would mean more 2b/3b. Well I was wrong on this. I decided to look at the data for 1996-2015, 20 years worth. I was very surprised what I found though.

I am not going to bore you with everything. Here are the nuts and bolts though. I used a specific data set for most of this. I took AB and subtracted SO and then used that as a basis for Balls in play. I know there are sacrifices, but they are too small a % make me do all the extra work. I also have data for PA %s based on total hits.

The HR/2b/3B rates have been just about stagnant when you look at balls hit in play.

The extra base hit rate to AB-SO not including HRs has not deviated more than 6 per 1000 during that 20 year time. The HR rate has not deviated more than 7 per 1000 and the total XB hit has not deviated more than 12 per 1000. That is extremely small.

If you draw a nice clean line at 2007 you are dealing with an average HR difference of 3 per 1000. If you look at the HR rate to total hits it is only 2 more per 1000. While the extra base hits minus HRs is only 1 more per 1000.

I would call that significant. Here is where I started scratching my head. Why are Runs down if the rates are basically identical? Well the answer is SOs. If you draw the line at 2008, where SOs really start to take off, you see a drastic drop in Runs per game.

From the data that I collected it looks like SOs are the most impactful statistic for league average run scoring. HRs are probably secondary. I want to point out now, that I am looking at hue data sets. Entire seasons. There are going to be teams that anecdotally defy the trend in any given season.

Example:
If you look at 2015, 2012, 2008, 2007, 1997 you have league average HR rates between 163 and 166.
2015: 164 HRs --- 1248 SO ---- 4.25 RPG
2012: 164 HRs --- 1214 SO ---- 4.32 RPG
2008: 163 HRs --- 1096 SO ---- 4.65 RPG
2007: 165 HRs --- 1073 SO ---- 4.80 RPG
1997: 166 HRs --- 1069 SO ---- 4.77 RPG
In all 5 of these years the HR rates do not change much, but the SOs do.

I think you can still blame some of this on PEDs, but there is no way to prove it. It makes sense to suggest that if I can swing with more force, but use less of my full available strength I can still maintain enough bat control to make contact and put the ball in the seats. I would buy that, I just can’t prove it.

My real take away though is that the opportunity cost of SOs is far greater than what I used to believe. I have heard on MLB TV people say that a SO is only a little more costly than a regular out. The problem with that is not all babip are outs. Babip does not even seem to be that important when it comes to actual runs scored.

Not making an out is by far the most important thing to scoring runs. I know that seems obvious. A SO is almost always an out, you are not going to get to first safely on dropped 3rd strikes enough. Making contact and putting the ball in play gives you about a 30% chance to reach safely. Extra base hits / PA have not exceeded 54 per 1000 in the time that I looked at. They are just not frequent enough. Timely hitting is really just putting the ball in play when you think about it.

From the Cubs point of few:

The Cubs need fewer SOs as a team.

The Cubs pitchers lead the majors in SOs. This is why I think the bad OF defense did not hurt them as much as it could have. I don’t know if the Cubs can do this again. If they move closer to the league average it will put a lot more balls in play for the shaky defense. I also suspect, since SO are the worst thing for an offense, that the Cubs RA may have been lower than it should have been based on the defense that they play.

Sorry for the length.
 

beckdawg

Well-known member
Joined:
Oct 31, 2012
Posts:
11,750
Liked Posts:
3,741
I've long argued this was a problem. It's not just that SO have no BABIP. Outs in play can also be productive if they advance runners. I think you can argue Bryant can get away with it because he has a pretty absurd .378 BABIP(which appears to be maintainable) which gave him a .275 BA despite his 30%+ k rate. Honestly I said this at least 2 years ago maybe more. I think the game is going do that whole "what's old is new again" type thing and you're going to see teams wain off of steroid era thinking and go back to 80's style baseball which focused heavily on contact and defense. I don't think you're going to totally see the ideas from the steroid era abandoned with OBP still being a big deal. It also helps that the royals have championed this type of play and they won the WS so surely it will be some what in vogue.

What I'm curious in personally is whether or not stolen bases make a large come back. They were largely viewed as not cost effective in the steroid era. However, in the 80's you had guys stealing 100 bases a year which is double what you see in a typical year now. With runs being harder to score I could see the return of small ball tactics being popular again.
 

czman

Well-known member
Joined:
May 7, 2013
Posts:
2,210
Liked Posts:
545
It is not just babip, you are correct. It is also the fact that the extra base hit rate to balls in play has not changed. Extra doubles and triples score runs to. I think to often people think about SO vs out or small chance at a single.

I was very surprised to see the HR rate basically identical once you remove SO.
 

CSF77

Well-known member
Joined:
Apr 16, 2013
Posts:
18,688
Liked Posts:
2,846
Location:
San Diego
SB will factor but in the 80's most parks were astro turf and having a speed based team payed off with the ball shooting faster through the infield. They could use the turf to bunt single more. In general those fields spurred that era. After they were removed the steroid era started with grass becoming the predominate field type again.

Teams need a balance. OBA at the top. Speed can factor but should not outweigh OBA. Middle of the order SLG should be the main factor. Get those extra base types up. Bottom of the order should be contact/D spec types. Pick up those bonus runs.
 

beckdawg

Well-known member
Joined:
Oct 31, 2012
Posts:
11,750
Liked Posts:
3,741
Teams need a balance. OBA at the top. Speed can factor but should not outweigh OBA. Middle of the order SLG should be the main factor. Get those extra base types up. Bottom of the order should be contact/D spec types. Pick up those bonus runs.

Lot of different ways to go about it but to this point, I think KC would disagree. Not saying that's the only way to win but typically when you talk OBP you're talking walks and they were last in the majors there. Honestly, I think part of the problem with SB though is football and basketball have killed a lot of the faster black players for baseball. That plus the lack of teams emphasizing it in the minors has a bit of a trickle down so that you just really don't mold players much that way anymore. As such, if we're going to approach the 70-100 SB mark again I think it may be 5-7 years before it happens because by in large the guys in the minors now, you don't have many 70-80 speed grades.

Buxton is an 80 runner. Moncada is a 65. Trea Turner is a 75. Jose Peraza is a 75. Manuel Margot is a 65. Ozhaino Albies is a 70. Raul Mondesi is a 70. Tim Anderson is a 70. Forrest Wall is a 70. Jorge Mateo is an 80. Anthony Alford is a 70. So, you're talking about 11 of the top 100 players are 65 or better grades from MLB.com who tends to be a bit more optimistic than other places. Of those players, Only Buxton and Moncada have more than 10ish HR power and a lot of them don't even have that so you often end up with guys who run but that's about all they do. Best case those types end up as Juan Pierre or Luis Castillo types but more realistically you see your Tony Campana's of the world.
 

JP Hochbaum

Well-known member
Joined:
May 22, 2012
Posts:
2,068
Liked Posts:
1,297
"The extra base hit rate to AB-SO not including HRs has not deviated more than 6 per 1000 during that 20 year time. The HR rate has not deviated more than 7 per 1000 and the total XB hit has not deviated more than 12 per 1000. That is extremely small."

Confused at how you got to the 1000 number. IS that Atbats minus strikeouts?
 

JP Hochbaum

Well-known member
Joined:
May 22, 2012
Posts:
2,068
Liked Posts:
1,297
IF anyone else in interested I am working up some other sabermetrics stats to improve upon WAR.

I have this one called WAM (wins above the mean).

https://sportsstatsandscience.wordpress.com/2015/10/22/another-new-stat-contact-wam/

"The technical explanation of what it does is taking any contact instance resulting in a safe hit (single, double, triple, and a home run) and using how often a player gets that per at bat. I then take the league mean (average) of the same per at bat number and get a number for each player of how well they performed above or below the league mean by using z-scores for each contact instance."
 

DanTown

Well-known member
Joined:
Mar 31, 2009
Posts:
2,446
Liked Posts:
509
Lot of different ways to go about it but to this point, I think KC would disagree. Not saying that's the only way to win but typically when you talk OBP you're talking walks and they were last in the majors there. Honestly, I think part of the problem with SB though is football and basketball have killed a lot of the faster black players for baseball. That plus the lack of teams emphasizing it in the minors has a bit of a trickle down so that you just really don't mold players much that way anymore. As such, if we're going to approach the 70-100 SB mark again I think it may be 5-7 years before it happens because by in large the guys in the minors now, you don't have many 70-80 speed grades.

Buxton is an 80 runner. Moncada is a 65. Trea Turner is a 75. Jose Peraza is a 75. Manuel Margot is a 65. Ozhaino Albies is a 70. Raul Mondesi is a 70. Tim Anderson is a 70. Forrest Wall is a 70. Jorge Mateo is an 80. Anthony Alford is a 70. So, you're talking about 11 of the top 100 players are 65 or better grades from MLB.com who tends to be a bit more optimistic than other places. Of those players, Only Buxton and Moncada have more than 10ish HR power and a lot of them don't even have that so you often end up with guys who run but that's about all they do. Best case those types end up as Juan Pierre or Luis Castillo types but more realistically you see your Tony Campana's of the world.

The thing about KC is that the ENTIRE TEAM has to be a contact team or else the value is lost. I think the thing people don't realize is that if you stack a bunch of hitters who each share a similar elite skill, no matter what they're elite at, you have a team that's very hard to beat, especially when they hit well.
 

czman

Well-known member
Joined:
May 7, 2013
Posts:
2,210
Liked Posts:
545
"The extra base hit rate to AB-SO not including HRs has not deviated more than 6 per 1000 during that 20 year time. The HR rate has not deviated more than 7 per 1000 and the total XB hit has not deviated more than 12 per 1000. That is extremely small."

Confused at how you got to the 1000 number. IS that Atbats minus strikeouts?

Yes. I ran the number every way you can and still have the data set so I can work more if I want. But I used AB - SO because it provides a good measure of balls hit in play. I also have data PA, but it was less important to this discussion.
 

czman

Well-known member
Joined:
May 7, 2013
Posts:
2,210
Liked Posts:
545
IF anyone else in interested I am working up some other sabermetrics stats to improve upon WAR.

I have this one called WAM (wins above the mean).

https://sportsstatsandscience.wordpress.com/2015/10/22/another-new-stat-contact-wam/

"The technical explanation of what it does is taking any contact instance resulting in a safe hit (single, double, triple, and a home run) and using how often a player gets that per at bat. I then take the league mean (average) of the same per at bat number and get a number for each player of how well they performed above or below the league mean by using z-scores for each contact instance."

I think the inherent problem you are going to run into is you are completely disregarding defense. I don't think you can get an approximation of win total id you only look at what a team produces. A better measure may to shoot for RPG (runs per game). I think that may be a little more accurate with the data that you are looking at.
 

JP Hochbaum

Well-known member
Joined:
May 22, 2012
Posts:
2,068
Liked Posts:
1,297
I think the inherent problem you are going to run into is you are completely disregarding defense. I don't think you can get an approximation of win total id you only look at what a team produces. A better measure may to shoot for RPG (runs per game). I think that may be a little more accurate with the data that you are looking at.

Yeah I plan on creating a defensive WAM and a non-contact WAM as well.
 

JP Hochbaum

Well-known member
Joined:
May 22, 2012
Posts:
2,068
Liked Posts:
1,297
Yes. I ran the number every way you can and still have the data set so I can work more if I want. But I used AB - SO because it provides a good measure of balls hit in play. I also have data PA, but it was less important to this discussion.

That just seems super low to me, rate wise. SHocked is all.
 

czman

Well-known member
Joined:
May 7, 2013
Posts:
2,210
Liked Posts:
545
That just seems super low to me, rate wise. SHocked is all.

Remember I am talking about the deviation between seasons, not the rate in any one given season. I started this thinking that the "PED" era would have a much higher HR rate to balls in play and the "non-PED" era would have a higher 2b/3b rate. They did. The rates differences were insignificant though.

I did most of this before the Cubs played the Mets. Just did not post it.

The one thing that I took note of though is the Murphy HR off Arrieta. It was not a great pitch. The ball was hit square, but not with a lot of power. It just carried. My bet is 99 times out for a hundred that swing goes foul or is a fly ball hit/out. On one day with the right wind, and atmospheric conditions in the right ball park with a Hot hitter who was sizing everything up square the ball carried out.

That kind of sealed it for me. Making contact, even contact that most times won't get you positive results, is far better than making no contact. The HR rate has not changed. Now, iif Bryant did cut his strikeouts down by 100 he would probably lose some babip. He would probably also hit another 5-7 HRs and 6-9 2b/3bs. Some of those HRs would be basket jobs, but a HR is a HR no matter how far it flies.
 

Top