Jump to content

Providing independent coverage of the Minnesota Twins.

The same great Twins Daily coverage, now for the Vikings.

The Store

Recent Blogs


Photo

Article: Geeking Out: OPS Minus Batting Average

  • Please log in to reply
80 replies to this topic

#1 Seth Stohs

Seth Stohs

    Owner

  • Administrators
  • 7,663 posts

Posted 25 June 2013 - 04:06 AM

You can view the page at http://twinsdaily.co...Batting-Average

#2 spycake

spycake

    Senior Member

  • Members
  • 2,046 posts

Posted 25 June 2013 - 06:17 AM

Another way to express it, that might make more sense, is OBP plus isolated power (ISO, which is just SLG - BA).

#3 Badsmerf

Badsmerf

    Senior Member

  • Members
  • 1,732 posts

Posted 25 June 2013 - 06:26 AM

I don't think OPS-BA is a better stat than OPS. BA, while fluctuating based on things like BABIP, LD%, FB/HR%, still is a key stat. Base hits are important. They drive in runs, move runers ect. Subtracting them is basically taking ISO + OBP. This is why power showed to be important in this stat.

I don't think a single stat deserves to tell an entire story on a player. One can make an argument that oWAR does make a good attempt. The 20 guys listed here will look good no matter how you slice it up, they are having monster seasons. When advanced stats come into play are for guys like Plouffe that contribute greatly yet quietly.

#4 Brock Beauchamp

Brock Beauchamp

    Owner

  • Administrators
  • 8,937 posts

Posted 25 June 2013 - 06:29 AM

I think it's the exact opposite. By counting hits twice, OPS is accidentally adding value to singles, which helps offset the fact that OBP is more important than slugging. Not making an out comes first, taking more than one base follows in importance by a pretty healthy margin.

OPS is a good quick and dirty stat but it should not be used to directly compare players because of the OBP/SLG split and the fact that they shouldn't be judged equally. There are a couple of ways to reach an OPS of .800 but if one of the guys has +/- .100 OBP (say, .425 to .325), that .425 OBP player is going to be much more valuable than the slugging-centric .325 OBP player.

Looking only at OPS, Dustin Pedroia and Ian Desmond look like virtually the same player (.812 to .809). Except that Pedroia is getting on base nearly 40% of the time while Desmond is getting on base less than 32% of the time. I think that pretty much everybody would take Pedroia's bat over Desmond, and for good reason. It's better.

Which is not evident by OPS+, where Pedroia trails Desmond 119 to 121 but quite evident in oWAR, where Pedroia leads Desmond 2.6 to 2.2.

#5 ThePuck

ThePuck

    Senior Member

  • Members
  • 3,232 posts

Posted 25 June 2013 - 06:35 AM

I think it's the exact opposite. By counting hits twice, OPS is accidentally adding value to singles, which helps offset the fact that OBP is more important than slugging. Not making an out comes first, taking more than one base follows in importance by a pretty healthy margin.

OPS is a good quick and dirty stat but it should not be used to directly compare players because of the OBP/SLG split and the fact that they shouldn't be judged equally. There are a couple of ways to reach an OPS of .800 but if one of the guys has +/- .100 OBP (say, .425 to .325), that .425 OBP player is going to be much more valuable than the slugging-centric .325 OBP player.

Looking only at OPS, Dustin Pedroia and Ian Desmond look like virtually the same player (.812 to .809). Except that Pedroia is getting on base nearly 40% of the time while Desmond is getting on base less than 32% of the time. I think that pretty much everybody would take Pedroia's bat over Desmond, and for good reason. It's better.

Which is not evident by OPS+, where Pedroia trails Desmond 119 to 121 but quite evident in oWAR, where Pedroia leads Desmond 2.6 to 2.2.


This post makes a lot of sense.

#6 Kwak

Kwak

    Senior Member

  • Members
  • 1,392 posts

Posted 25 June 2013 - 06:49 AM

What's the correlation between OBP and scoring compared to SLG and scoring?

#7 Brock Beauchamp

Brock Beauchamp

    Owner

  • Administrators
  • 8,937 posts

Posted 25 June 2013 - 06:53 AM

What's the correlation between OBP and scoring compared to SLG and scoring?


Runs Scored Correlations

#8 jay

jay

    Senior Member

  • Members
  • 927 posts

Posted 25 June 2013 - 06:55 AM

What's the correlation between OBP and scoring compared to SLG and scoring?


I asked myself the same thing a few minutes ago. Here was the first link: Correlation Between Stats and Runs, etc.

I'll continue to defend OPS even over something like Seth is proposing here, especially if we're talking for the masses. OPS correlates to runs extremely well, something in the neighborhood of .94-.97 ® depending on the source and data pool. Even the most advanced stats can't make a big improvement over that. OPS is something that can be easily explained and understood by casual fans. As for GM analysis, well, I'd hope they're into stuff a lot deeper than both of these...

#9 Brock Beauchamp

Brock Beauchamp

    Owner

  • Administrators
  • 8,937 posts

Posted 25 June 2013 - 07:11 AM

This is a pretty interesting analysis of correlation that goes into a bit more than "this is the correlation number to runs scored" because the topic is a little more complex than that:

What?s more important, OBP, Slugging or OPS? | Reading into the Numbers

"A team that raises its OBP from .300 to .400 is expected to increase it’s runs scoring from 3.7 runs/game to 6.7. The equivalent increase in slugging would be about 131 points, on average. An increase in slugging from .358 to .489 predicts a scoring increase from 3.7 to 5.8 runs/game, an obvious downgrade from the OBP surge. However, if a team increases its OBP purely by walking more, then the 100 point increase in OBP will likely only increase run scoring from 3.7 to less than 5 runs/game. What a GM can take from this is that, while increases in OBP seem to lead to more run scoring than equivalent increases in slugging percentage, this is only the case when these increases come with a mix of walks and hits."

#10 Gernzy

Gernzy

    Senior Member

  • Members
  • 448 posts

Posted 25 June 2013 - 07:16 AM

Wow I didn't realize Cuddyer was having such a great year. Good for him!

#11 Badsmerf

Badsmerf

    Senior Member

  • Members
  • 1,732 posts

Posted 25 June 2013 - 07:24 AM

I think it's the exact opposite. By counting hits twice, OPS is accidentally adding value to singles, which helps offset the fact that OBP is more important than slugging. Not making an out comes first, taking more than one base follows in importance by a pretty healthy margin.

OPS is a good quick and dirty stat but it should not be used to directly compare players because of the OBP/SLG split and the fact that they shouldn't be judged equally. There are a couple of ways to reach an OPS of .800 but if one of the guys has +/- .100 OBP (say, .425 to .325), that .425 OBP player is going to be much more valuable than the slugging-centric .325 OBP player.

Looking only at OPS, Dustin Pedroia and Ian Desmond look like virtually the same player (.812 to .809). Except that Pedroia is getting on base nearly 40% of the time while Desmond is getting on base less than 32% of the time. I think that pretty much everybody would take Pedroia's bat over Desmond, and for good reason. It's better.

Which is not evident by OPS+, where Pedroia trails Desmond 119 to 121 but quite evident in oWAR, where Pedroia leads Desmond 2.6 to 2.2.

I agree and disagree at the same time. I might favor Pedroia slightly more, but not by as much as you're suggesting. Desmond has 12 HR on the year compared to Pedroia's 4. That is pretty significant. I don't want to turn this into a Pedroia vs. Desmond thread, but this is really Pedroia's on base skills vs. Desmond's power. Fortunately for Pedroia, he is an OBP machine and thus gets a slight edge to Desmond.

I know you are bias toward OBP, and for good reason. I'm pretty balanced on BA, OBP, and SLUG. Maybe I am just too used to seeing SLUG instead of ISO. I might feel differently about OPS-BA if we just used ISO instead of SLUG. Maybe it is more indicative. In the case of Desmond vs. Pedroia, this would make Desmond sit at .529 and Pedroia sit at .501. So, you would think OPS would then be better since it favors your position more.
Do or do not. There is no try.

#12 Oldgoat_MN

Oldgoat_MN

    Senior Member

  • Members
  • 688 posts

Posted 25 June 2013 - 07:25 AM

We can't be sure that something like this isn't being used by one of the MLB teams.
During the discussion of who should be the AL MVP last year, an Oakland exec said that their stats had Cabrera leading Trout by a tiny margin.
Pretty sure they were not overly influenced by RBI.

Fun stuff.

#13 Badsmerf

Badsmerf

    Senior Member

  • Members
  • 1,732 posts

Posted 25 June 2013 - 07:27 AM

Have I ever said I have a B.S. in Mathematics? I love the numbers in baseball and how to interpret them. There is almost no 2 people that place the same emphasis on the same categories. One of the many reasons I absolutely love this game.
Do or do not. There is no try.

#14 wabene

wabene

    Member

  • Members
  • 34 posts

Posted 25 June 2013 - 07:34 AM

Wow I didn't realize Cuddyer was having such a great year. Good for him!


A band box in thin air helps

#15 Brock Beauchamp

Brock Beauchamp

    Owner

  • Administrators
  • 8,937 posts

Posted 25 June 2013 - 07:40 AM

I know you are bias toward OBP, and for good reason. I'm pretty balanced on BA, OBP, and SLUG. Maybe I am just too used to seeing SLUG instead of ISO. I might feel differently about OPS-BA if we just used ISO instead of SLUG. Maybe it is more indicative. In the case of Desmond vs. Pedroia, this would make Desmond sit at .529 and Pedroia sit at .501. So, you would think OPS would then be better since it favors your position more.


I also believe in balance or those OBP guys won't score (it's a lot easier to drive home runners with a SLG guy than an OBP guy).

But all things being equal, I think it's harder to find those OBP guys and runs scored correlations tend to favor a .001 increase in OBP more than a .001 increase in SLG.

In the end, you need both... But if we're drafting, I'm taking the OBP guys first because they're more valuable and they'll be off the board more quickly. If the player is both an OBP and SLG monster, well... All the better.

#16 Badsmerf

Badsmerf

    Senior Member

  • Members
  • 1,732 posts

Posted 25 June 2013 - 07:43 AM

A band box in thin air helps

Since his HR/FB ratio is about 12.9 % which is below the 15.1% he enjoyed in 2009, so the thin air isn't making much of an impact here. His LD% is also at 22% which is the highest of his career, not really sure of the thin air make him hit more line drives. O yeah, he is also enjoying a .382 BABIP, which I don't think is an effect of Coors Field. In case you were wondering, his highest BABIP ever is .328 in 2006.

Lets be fair when assessing players.
Do or do not. There is no try.

#17 Forever34

Forever34

    Member

  • Members
  • 30 posts

Posted 25 June 2013 - 07:53 AM

I'm not an expert on advanced metrics, but I'm trying to start a new stat called Total Batting Productivity. It is calculated similar to slugging in that it assigns 1 point for a single, 2 for a double etc., but also subtracts 1 point for a strikeout, 2 for a GiDP, and 3 for those rare GiTP. All of that is divided by total number of at bats.

A similar more advanced stat would be Total Offensive Productivity which not only includes 1 point for walks and HBP but also 1 for stolen bases. It also subtracts 1 for a caught stealing and is divided by total plate appearances.

#18 Guest_USAFChief_*

Guest_USAFChief_*
  • Guests

Posted 25 June 2013 - 07:53 AM

This is a pretty interesting analysis of correlation that goes into a bit more than "this is the correlation number to runs scored" because the topic is a little more complex than that:

What?s more important, OBP, Slugging or OPS? | Reading into the Numbers

"A team that raises its OBP from .300 to .400 is expected to increase it’s runs scoring from 3.7 runs/game to 6.7. The equivalent increase in slugging would be about 131 points, on average. An increase in slugging from .358 to .489 predicts a scoring increase from 3.7 to 5.8 runs/game, an obvious downgrade from the OBP surge. However, if a team increases its OBP purely by walking more, then the 100 point increase in OBP will likely only increase run scoring from 3.7 to less than 5 runs/game. What a GM can take from this is that, while increases in OBP seem to lead to more run scoring than equivalent increases in slugging percentage, this is only the case when these increases come with a mix of walks and hits."


What studies like this miss, or might want to consider, is that in the real world increases in OBP consisting of "a mix of walks and hits" also increase SLG.

As for dropping BA from OPS, IMO we'd all be better off dropping OPS and using the triple slash line instead. All OBP is not created the same. A single is worth more than a walk, for example, and by knowing BA/OBP/SLG you get a pretty clear picture of a hitter...clearer than combining OBP and SLG into one number, IMO.

#19 ThePuck

ThePuck

    Senior Member

  • Members
  • 3,232 posts

Posted 25 June 2013 - 07:54 AM

A band box in thin air helps


He's doing very well away from Coors as well. If his OPS away from Coors was his season OPS, he would rank 7th in the NL for OFs, and 2nd in the AL for OFs.

And most players do better at home anyway. That's not to say that Coors doesn't give him an advantage.

Edited by ThePuck, 25 June 2013 - 07:56 AM.


#20 Badsmerf

Badsmerf

    Senior Member

  • Members
  • 1,732 posts

Posted 25 June 2013 - 07:54 AM

I also believe in balance or those OBP guys won't score (it's a lot easier to drive home runners with a SLG guy than an OBP guy).

But all things being equal, I think it's harder to find those OBP guys and runs scored correlations tend to favor a .001 increase in OBP more than a .001 increase in SLG.

In the end, you need both... But if we're drafting, I'm taking the OBP guys first because they're more valuable and they'll be off the board more quickly. If the player is both an OBP and SLG monster, well... All the better.

OBP skills tend to follow a player much better than SLUG. I agree with valuing OBP in young players because it is so difficult to project power in the MLB.

I don't know if we are arguing over a point or just having a discussion right now.
Do or do not. There is no try.

#21 CRArko

CRArko

    Agent of SHIELD

  • Members
  • 1,694 posts
  • LocationThe Playground
  • Twitter: crarko

Posted 25 June 2013 - 07:58 AM

Have I ever said I have a B.S. in Mathematics? I love the numbers in baseball and how to interpret them. There is almost no 2 people that place the same emphasis on the same categories. One of the many reasons I absolutely love this game.


Huh, me too, although the truths hidden in these numbers is often pretty gray. I would say that Huff should be mandatory reading for anyone who argues baseball statistics.

#22 John Bonnes

John Bonnes

    Owner

  • Administrators
  • 4,977 posts

Posted 25 June 2013 - 08:00 AM

I can't run the numbers myself - I'm a little busy this morning - but would anyone like to do a quick correlation between Seth's OPS-BA stat and Runs scored and share it with us?

#23 Badsmerf

Badsmerf

    Senior Member

  • Members
  • 1,732 posts

Posted 25 June 2013 - 08:01 AM

I'm not an expert on advanced metrics, but I'm trying to start a new stat called Total Batting Productivity. It is calculated similar to slugging in that it assigns 1 point for a single, 2 for a double etc., but also subtracts 1 point for a strikeout, 2 for a GiDP, and 3 for those rare GiTP. All of that is divided by total number of at bats.

A similar more advanced stat would be Total Offensive Productivity which not only includes 1 point for walks and HBP but also 1 for stolen bases. It also subtracts 1 for a caught stealing and is divided by total plate appearances.

Problem with that is you are counting things against a play that he has no control over. For example, a ball hit to SS is an out with no runners on and a GIDP with a runner on first, but not on second and possibly an RBI if the runner on 3b is fast and it is late in the game with the SS playing deep. See, too many variables to make it credible.
Do or do not. There is no try.

#24 jay

jay

    Senior Member

  • Members
  • 927 posts

Posted 25 June 2013 - 08:03 AM

I'm not an expert on advanced metrics, but I'm trying to start a new stat called Total Batting Productivity. It is calculated similar to slugging in that it assigns 1 point for a single, 2 for a double etc., but also subtracts 1 point for a strikeout, 2 for a GiDP, and 3 for those rare GiTP. All of that is divided by total number of at bats.

A similar more advanced stat would be Total Offensive Productivity which not only includes 1 point for walks and HBP but also 1 for stolen bases. It also subtracts 1 for a caught stealing and is divided by total plate appearances.


Interesting ideas. Some feedback: I think a big shortfall is assigning full integer values to all outcomes. IE - A strikeout isn't equally as bad as a single is good.

Stats like wOBA do a pretty good job of measuring all batting outcomes based on their true impact. Here's a link: wOBA | FanGraphs Sabermetrics Library

#25 jay

jay

    Senior Member

  • Members
  • 927 posts

Posted 25 June 2013 - 08:08 AM

John, it showed up in the article I linked to earlier (Correlation Between Stats and Runs, etc.) in the alternate form a few people pointed out (ISO+OBP). The r value to runs was .941 compared to .960 for OPS alone.

#26 John Bonnes

John Bonnes

    Owner

  • Administrators
  • 4,977 posts

Posted 25 June 2013 - 08:16 AM

John, it showed up in the article I linked to earlier (Correlation Between Stats and Runs, etc.) in the alternate form a few people pointed out (ISO+OBP). The r value to runs was .941 compared to .960 for OPS alone.


Thanks. I think this suggests that OPS and OPS-BA are essentially equal in their importance for a team. OPS is SLIGHTLY more important, but that's splitting hairs. They're essentially equivalent.

So whether you count the hits once or twice, it's about the same, although it might be just a little better to count the hits twice, for whatever reason.

#27 Brock Beauchamp

Brock Beauchamp

    Owner

  • Administrators
  • 8,937 posts

Posted 25 June 2013 - 08:19 AM

As for dropping BA from OPS, IMO we'd all be better off dropping OPS and using the triple slash line instead. All OBP is not created the same. A single is worth more than a walk, for example, and by knowing BA/OBP/SLG you get a pretty clear picture of a hitter...clearer than combining OBP and SLG into one number, IMO.


For individual players, definitely. You want more than one stat line to judge the efficacy of a player.

For quick and dirty analysis or team stats, I think OPS has more of a place, as multiple batters tend to cancel out extreme variances in OBP and/or SLG.

#28 Brock Beauchamp

Brock Beauchamp

    Owner

  • Administrators
  • 8,937 posts

Posted 25 June 2013 - 08:21 AM

OBP skills tend to follow a player much better than SLUG. I agree with valuing OBP in young players because it is so difficult to project power in the MLB.

I don't know if we are arguing over a point or just having a discussion right now.


We'll just call it an arcussionment.

#29 jay

jay

    Senior Member

  • Members
  • 927 posts

Posted 25 June 2013 - 08:32 AM

Here's an interesting look at some of Brock's musings on the value of the OBP guy vs the SLG guy. Lineup Styles, Part One - Beyond the Box Score

"The high-OBP lineup scored about 10% more runs than the high-SLG lineup. In the comments section, he notes that the lower wOBA gets, the OBP-heavy lineups decrease production faster, eventually being passed by SLG-heavy lineups."

#30 fairweather

fairweather

    Banned

  • Banned
  • 253 posts

Posted 25 June 2013 - 08:36 AM

I hate batting average. I think every team should be coached to hit the way Toronto coaches their hitters. Just swing hard all the time and hit dingers. Maybe now you people can understand why I'm not a big Mauer fan.