Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Sunday, September 17, 2017

If you’re the Red Sox, do you want to play the Indians or Astros in the divisional round of the playoffs?

Nick Cafardo’s latest.

Jim Furtado Posted: September 17, 2017 at 03:09 PM | 63 comment(s) Login to Bookmark
  Tags: notes, red sox

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Jose is an Absurd Doubles Machine Posted: September 17, 2017 at 04:28 PM (#5533433)
Win the division first, then worry about it.

I'll take my chances with either. This is a good team and while they will likely be underdogs against either team they were favorites last year and that worked out well.
   2. willcarrolldoesnotsuk Posted: September 17, 2017 at 05:19 PM (#5533473)
Which do I have the more recent signs for?
   3. JJ1986 Posted: September 17, 2017 at 06:33 PM (#5533513)
Yes.
   4. the Hugh Jorgan returns Posted: September 17, 2017 at 06:43 PM (#5533519)
Cleveland's pitching is approaching otherworldly right now, I'd much rather face the team with the weaker pitching. If the Astros or NY for that matter, just continue to bludgeon their way over the next couple of weeks, then there isn't really much you can do about that but try to keep the ball in the park(Rick Porcello need not apply) and catch it when they do hit it.

However in spite of playing well as of late, with NY refusing to lose more then once a week, winning the division should be the only focus at this point. With 13 games left, I'm thinking they will need to go 8-5 to stay above NY as I've got this feeling NY are going to go like 10-3 over the last 2 weeks.
   5. BDC Posted: September 17, 2017 at 07:08 PM (#5533523)
On the whole, there aren't many sports where I'd care about the order my playoff team met others in. College basketball, yes; I would be happy if I was following some middling seed and the top ones in "our" path got knocked off early, leaving "us" to face the Cinderellas. But even there, what happened to the top seeds can happen to you.
   6. Walt Davis Posted: September 17, 2017 at 07:44 PM (#5533532)
We'll see what the long-term costs were but the Yanks cleaned up at the trade deadline:

T Frazier 1.1 WAR
Gray 1.5 WAR
Kahnle 0.7 WAR
Robertson 1.4 WAR
Garcia -0.2 WAR but 100 ERA+ (horrible peripherals)

4.5 wins over 6-7 weeks ... I know, the guys they replaced might not have been strictly replacement level but still that's a big boost.

But yes, first step is to make sure your first "round" opponent isn't the Twins or Angels.
   7. Walt Davis Posted: September 17, 2017 at 08:07 PM (#5533539)
Doesn't seem any useful thread to put this and it does affect most of the teams involved here ... the sorry state of the DH:

2017, min 80 games at DH, OPS+, WAR

Cruz 142, 3.6
EE 124, 2.6
Trumbo 86, 0.2 (how is he above-replacement?)
Holliday 95, 0.0
Hanley 93, -0.2
Beltran 88, -0.2
Kendrys 94, -0.3
VMart 84, -0.6
Moss 90, -0.7
Pujols 83, -1.5

I still don't have a particularly good answer for Trumbo ... he's got 249 innings in RF where he's apparently been non-terrible and he's provided positive baserunning for the first time in his career. But I don't see how a line of 240/293/409 in 580 PA is just -7 Rbat.

Tampa and Minnesota apparently have cobbled together average DH slots and Tex, Oak and the Yanks (not Holliday) aren't terrible. The rest of the league is pretty terrible. League line is 243/319/421, about a 96 OPS+. The average DH OPS (and average LF OPS) is the same as the average SS OPS and below 2B, CF and 3B. Something is very wrong about all of that.

For now it's just a one-year fluke with OPS+ usually around 110 in the last 6 years (2013 was somewhat similar at 101). And last year was easily the best OPS by SS since 2012 and this year beats last year pretty easily.
   8. Joyful Calculus Instructor Posted: September 17, 2017 at 08:11 PM (#5533541)
It doesn't matter. Other than the Braves and Nationals, anyone can be a tough opponent. (And tbh, there's probably a Wyatt Earp effect with respect to those teams)
   9. Where have you gone Brady Anderson? Posted: September 17, 2017 at 08:12 PM (#5533543)
Those DH's are mostly old guys who aren't good enough to play in the field anymore. You put an old, partly broken down player at DH, you shouldn't be surprised when he plays like an old, broken down player.
   10. Sunday silence Posted: September 17, 2017 at 08:30 PM (#5533550)
yeah but what Walt is saying is that OPS+ is usually around 110 for this position. So if your theory was correct you would see a below average OPS+ that was holding steady for year after year. Instead we are not.

*****

Question: is the difference between the ranking of OPS+ and WAR in walts post have to do with baserunning runs (Rbas)??
   11. cardsfanboy Posted: September 17, 2017 at 08:46 PM (#5533555)
Question: is the difference between the ranking of OPS+ and WAR in walts post have to do with baserunning runs (Rbas)??


Some of it is rbase(Hanley, Holliday and Beltran are all negative, while Trumbo is positive) The other difference is pa, with the higher pa, Trumbo and a few others have a higher replacement bonus.
   12. Fancy Crazy Town Banana Pants Handle Posted: September 17, 2017 at 09:12 PM (#5533560)
yeah but what Walt is saying is that OPS+ is usually around 110 for this position. So if your theory was correct you would see a below average OPS+ that was holding steady for year after year. Instead we are not.

There are 15 DH's. Often not full time. Taking an Ortiz or two out of that group can bring the average down in a hurry.
   13. cardsfanboy Posted: September 17, 2017 at 09:31 PM (#5533563)
There are 15 DH's. Often not full time. Taking an Ortiz or two out of that group can bring the average down in a hurry.


In the 45 year history of the Dh, this 97 ops+ is the lowest it's ever been, the next lowest was 101 in 1974 and 1985. And it's usually pretty predictable between 105-114 in most of it's history, this year is an aberration that isn't caused by the loss of one guy.
   14. Ziggy: The Platonic Form of Russell Branyan Posted: September 17, 2017 at 09:46 PM (#5533566)
I'm going to preface this with: I don't know.

That said, my go-to response to everything like this is "probably no reason, weird random stuff happens, give it a while and see if there's a trend, then we'll start looking for explanations."
   15. cardsfanboy Posted: September 17, 2017 at 09:55 PM (#5533567)
Just in case anyone wants to look at the bb-ref split on the dh. Click here for the full list.

Here is a chart with the highlights. Seasonal rate totals for DH, sorted by year. tOPS+ is ops+ in this particular instance.

Year    OPS   BA  OBP  SLG BAbip tOPS+
2017   .741 .244 .320 .421  .285    97
2016   .775 .254 .326 .449  .289   109
2015   .765 .259 .329 .436  .299   111
2014   .736 .247 .317 .419  .286   110
2013   .725 .245 .322 .402  .287   103
2012   .754 .255 .326 .428  .293   108
2011   .764 .263 .337 .427  .298   112
2010   .752 .250 .330 .422  .287   106
2009   .782 .256 .338 .444  .287   108
2008   .771 .255 .338 .433  .286   106
2007   .801 .268 .354 .447  .296   111
2006   .811 .263 .347 .463  .289   110
2005   .775 .258 .337 .437  .287   106
2004   .789 .262 .345 .444  .292   107
2003   .793 .262 .350 .444  .292   110
2002   .785 .262 .342 .444  .291   110
2001   .784 .260 .340 .444  .289   106
2000   .823 .276 .362 .461  .304   110
1999   .834 .276 .363 .471  .303   114
1998   .815 .274 .356 .459  .305   116
1997   .793 .274 .352 .441  .307   110
1996   .831 .277 .365 .466  .307   116
1995   .819 .276 .363 .456  .304   117
1994   .824 .273 .353 .472  .296   116
1993   .757 .262 .332 .425  .286   106
1992   .738 .258 .334 .405  .284   111
1991   .769 .257 .344 .424  .285   117
1990   .721 .251 .326 .395  .281   103
1989   .713 .256 .323 .390  .284   105
1988   .731 .251 .334 .397  .276   110
1987   .770 .253 .337 .433  .276   106
1986   .762 .256 .332 .429  .279   110
1985   .721 .240 .315 .405  .252   101
1984   .748 .256 .330 .418  .275   111
1983   .772 .267 .344 .428  .288   116
1982   .758 .265 .338 .421  .280   112
1981   .708 .247 .326 .383  .264   106
1980   .758 .270 .336 .422  .286   112
1979   .750 .263 .331 .419  .275   106
1978   .723 .253 .320 .403  .268   105
1977   .756 .264 .332 .423  .282   106
1976   .700 .257 .325 .375  .282   105
1975   .718 .254 .327 .391  .272   104
1974   .698 .256 .325 .373  .279   101
1973   .720 .257 .328 .393  .274   105 
   16. Jose is an Absurd Doubles Machine Posted: September 17, 2017 at 10:03 PM (#5533570)
To answer Fancy Pants if my math is right taking a guy with a 164 OPS+ out of a group of fifteen players with a 115 OPS+ would make it about 110-111 for the remaining players so call the impact of one star at about five points of OPS+ so yanking Ortiz out is about a third of the drop.

Is the fact that this is more of an offensive season part of the impact? Do DHs have lower OPS+ in a big offensive year?
   17. Jose is an Absurd Doubles Machine Posted: September 17, 2017 at 10:03 PM (#5533571)
Cfb's chart suggests no, a big offensive year does not make an impact.
   18. Fancy Crazy Town Banana Pants Handle Posted: September 17, 2017 at 10:16 PM (#5533574)
Year    OPS   BA  OBP  SLG BAbip tOPS+
2017   .741 .244 .320 .421  .285    97
2016   .775 .254 .326 .449  .289   109 


That's a difference of 34 points of OPS. So some back of the envelope math, 34*15 is 510, so you need to find roughly a drop of 510 points of OPS total.

Going from Ortiz 2016 to Hanley 2017 gets you about 270, or more than half the way. Add Toronto going from EE to Kendrys Morales and you are over 400. Add Pujols dropping 100 points of OPS (and if not for his contract he would not be DHing... he would not be in baseball), and you are there. It really doesn't take much.
   19. Fancy Crazy Town Banana Pants Handle Posted: September 17, 2017 at 10:21 PM (#5533576)
To answer Fancy Pants if my math is right taking a guy with a 164 OPS+ out of a group of fifteen players with a 115 OPS+ would make it about 110-111 for the remaining players so call the impact of one star at about five points of OPS+ so yanking Ortiz out is about a third of the drop.

The Red Sox didn't replace him with an average DH though. They replaced him with a guy with a 93 OPS+.
   20. Walt Davis Posted: September 17, 2017 at 10:31 PM (#5533580)
I suspect it's just a fluke paired with some bad contracts. The Jays just signed Morales for 3 years, VMart has one year left after this, Pujols we all know about. So they weren't going to lose their jobs this year and the Jays and Angels hope it's just a fluke-y bad year while Detroit is probably not that worried about cutting bait on VMart next year if he's done. Oh yeah, Trumbo's on his new 3-year contract too.

I'm a little surprised there are so many qualifying at 80+ games. It's not unusually high or low but it is higher than the 2012-14 period (a low point it seems) while I would have thought adding the 13th reliever would have forced teams even further away from a "full-time" DH. (That may also be playing a small part in those WAR variations as some guys have reasonable time in the field ... mostly covered by cfb's comment.)

As others have implied, DH is always the position for aging, fading players ... maybe it's moreso this year than usual but I doubt it. And a reasonable number in that list aren't that old -- Moss 33, Morales 34, Trumbo 31, Hanley 33, EE 34. This might well be the last hurrah for Holliday and Beltran ... and would be for Pujols and VMart under other circumstances.
   21. Walt Davis Posted: September 18, 2017 at 03:47 AM (#5533600)
The Red Sox didn't replace him with an average DH though. They replaced him with a guy with a 93 OPS+.

Sure but the average this year is 97 so obviously Hanley isn't pulling that average down. At best, the non-Red Sox DH's this year are averaging a 98 while last year they averaged (based on #16) they'd be averaging about a 105.

Since you insist, 2016 vs 2017 OPS at DH by team

Tor 818 761 -57 (EE to Kendrys)
Tex 701 680 -21
TBR 698 753 +55 (balances Tor)
SEA 852 906 +54 (SEA and OAK balance LAA)
OAK 705 727 +22
NYY 762 761
MIN 762 720 -42
LAA 758 684 -74
KCR 764 701 -63 (Kendrys to Moss, etc)
Hou 677 700 +23
Det 792 715 -77 (Det and CWS balanced by CLE)
CLE 798 900 +102 (Santana et al to EE et al ... the et al have been raking this year)
CWS 711 681 -30
BOS 1045 733 -312
BAL 842 670 -172

So Boston had by far the biggest drop but 9 teams have seen declines, 6 by more than 50 points. Last year 10 teams beat a 740 OPS, this year it's just 6. The DH average OPS dropped 40 points despite league OPS going up 12 points.

The departure of Papi does not explain the massive drop in Balt or what happened in LAA, Det, KC, etc. And the very underlying question is why didn't the Red Sox replace Ortiz with a (pre-2017) average DH? Is this just a fluke (probably) or is something more going on? (Too soon to tell)

BTW, Hanley also is signed through next year and is 528 PA from his option vesting.
   22. SandyRiver Posted: September 18, 2017 at 08:34 AM (#5533613)
And the very underlying question is why didn't the Red Sox replace Ortiz with a (pre-2017) average DH? Is this just a fluke (probably) or is something more going on? (Too soon to tell)

BTW, Hanley also is signed through next year and is 528 PA from his option vesting.


Since Hanley's 2016 OPS+ was 126, they probably thought that they had done that. Unfortunately, his bat reverted to 2015.
   23. Jose is an Absurd Doubles Machine Posted: September 18, 2017 at 08:39 AM (#5533614)
The interesting thing about that list in #21 is that this year is a higher run environment from last year. The AL OPS is up from .744 to .756.

And the very underlying question is why didn't the Red Sox replace Ortiz with a (pre-2017) average DH?


They thought they did. Hanley hit .286/.361/.505/126 last year. I think the Sox expected similar production. My theory is that Hanley needs to be an infielder of some sort to be productive. He has a long track record of success then was terrible in 2015 as a left fielder, had a great year last year as a first baseman, and has been terrible again this year as a DH. There is plenty of counter evidence for this; he was raking to start 2015 then crashed into a wall and injured his shoulder and this year his splits are no different whether he's at first or DH.
   24. BDC Posted: September 18, 2017 at 09:25 AM (#5533619)
Very interesting, Walt and everybody.

I guess a fluke is the best explanation for '17. The guys who have been used regularly at DH don't happen to be very good this year, on the whole.

Overall, cfb's list in #15 may show that the trend from the mid-1990s to now is a little bit down, which may show that there's a trend toward just rotating guys through the DH slot, in which case the position as a whole is going to tend toward dead average, because you've got random starters/decent bench hitters filling it on aggregate. But that's just a guess. There may be no significant trend at all.
   25. Rally Posted: September 18, 2017 at 09:44 AM (#5533627)
Going from Ortiz 2016 to Hanley 2017 gets you about 270, or more than half the way. Add Toronto going from EE to Kendrys Morales and you are over 400. Add Pujols dropping 100 points of OPS (and if not for his contract he would not be DHing... he would not be in baseball), and you are there. It really doesn't take much.


EE isn't in Toronto but he's still in the league mix of DHs. Toronto is worse at the position, Cleveland is better. KC is worse, going from Morales to Moss, etc.

Boston, as mentioned above, is responsible for half the decline in DH production. The next biggest is Baltimore, with a .668 OPS at the spot. That's due to Trumbo having a .648 OPS as a DH. He hasn't been good overall (.699), but for the last two years has been significantly worse when not playing the field. Last season he got a lot of time standing around in right field, Pedro Alvarez had the most PA as a DH and hit well in the role.

Detroit and LA have the next worse declines at DH, as Pujols and Victor Martinez are trying to hit with giant forks stuck in their backs. Then you get KC and Toronto. Cleveland is the only team with a big improvement at the spot, though Seattle and Tampa Bay have .050 OPS improvements as well.

   26. Morty Causa Posted: September 18, 2017 at 09:51 AM (#5533629)
What is the average OPS+ for pitchers--in case the DH is done away with, what can we expect? (I realize it would be necessary to look at the NL for that.) Has the average pitcher OPS+ gotten appreciably worse in the last couple of decades?
   27. Rally Posted: September 18, 2017 at 10:34 AM (#5533641)
I think OPS+ for pitchers is pretty close to zero.
   28. Rally Posted: September 18, 2017 at 11:20 AM (#5533669)
Actually, zero is optimistic

OPS, OPS+, NL only
2017 327, -11
2016 340, -7
2015 331, -6
2014 312, -10

   29. What did Billy Ripken have against ElRoy Face? Posted: September 18, 2017 at 12:18 PM (#5533695)
My theory is that Hanley needs to be an infielder of some sort to be productive. He has a long track record of success then was terrible in 2015 as a left fielder, had a great year last year as a first baseman, and has been terrible again this year as a DH.

Counter-theory: That thing about correlation and causation.
   30. PepTech Posted: September 18, 2017 at 12:36 PM (#5533712)
If you’re the Red Sox, do you want to play the Indians or Astros in the divisional round of the playoffs?
Dear Sox Fans: Go screw yourselves.

#GrumpyMarinerFan
   31. Morty Causa Posted: September 18, 2017 at 12:59 PM (#5533729)
Dear Sox Fans: Go screw yourselves.

Piece of cake.
   32. SoSH U at work Posted: September 18, 2017 at 01:08 PM (#5533734)
If you’re the Red Sox, do you want to play the Indians or Astros in the divisional round of the playoffs?


Dear Sox Fans: Go screw yourselves.

First Division Problems.
   33. PepTech Posted: September 18, 2017 at 01:18 PM (#5533736)
SoSH, that was my first choice as my hashtag there, but figured ID'ing as an M's fan would make me more pitiable.

I actually find the Sox crowd to be much less intolerable than either Yankees or Cards fans. ;)
   34. Steve Balboni's Personal Trainer Posted: September 18, 2017 at 01:38 PM (#5533750)
Hanley has gotten a pretty free ride in his three years in Boston, because Panda was so historically awful. Ramirez is now less than 530 PAs between now and the end of next year of his 5th-year option vesting. Ugh.

Oh, and Panda is 4-for-his-last-58 ABs, .069/.125/.172.
   35. Jose is an Absurd Doubles Machine Posted: September 18, 2017 at 01:43 PM (#5533754)
Oh, and Panda is 4-for-his-last-58 ABs, .069/.125/.172.


So better than he was in Boston?
   36. Traderdave Posted: September 18, 2017 at 01:47 PM (#5533757)
OPS, OPS+, NL only
2017 327, -11
2016 340, -7
2015 331, -6
2014 312, -10


How does OPS+ go negative? Does that mean the team should simply concede the out and not even bother picking up a bat?
   37. PreservedFish Posted: September 18, 2017 at 01:53 PM (#5533763)
I've always paired Hanley Ramirez and Jose Reyes in my mind. Reyes is about 6 months older. Hanley has always been a better hitter, Jose always more athletic, and it's still true.

Hanley has 37.5 WAR (-0.3 this year), Reyes has 37 (also -0.3 this year). Hanley's still a better hitter (107 wRC+ per Zips), but he doesn't have a prayer of ever again playing a skill infield position. Reyes is a 90 wRC+ hitter, he's still pretty lithe, but an average infielder at very best.

Ignoring the contracts, who would you rather have going forward?
   38. Nasty Nate Posted: September 18, 2017 at 01:55 PM (#5533766)
Does that mean the team should simply concede the out and not even bother picking up a bat?
That's not what it means. It just means a really low OPS.
   39. Fancy Crazy Town Banana Pants Handle Posted: September 18, 2017 at 02:43 PM (#5533805)
EE isn't in Toronto but he's still in the league mix of DHs. Toronto is worse at the position, Cleveland is better.

Cleveland's primary DH last season was Carlos Santana, who had an OPS of .866. Basically identical to EE this year. I did look that up at the time, but I didn't want the post to get too convoluted, so I left it out.

Boston, as mentioned above, is responsible for half the decline in DH production.

That is really my point though. It just doesn't take a lot to have a big impact on DH production. If you have 2-3 big drop offs, that is all it takes. The rest is just random fluctuations. Whether one team is 50 OPS points better, or another 70 worse. Those are the kind of fluctuations we see at every position every year. The difference being they have 30 players to average out the big outliers, instead of 15.
   40. Rally Posted: September 18, 2017 at 02:49 PM (#5533808)
How does OPS+ go negative? Does that mean the team should simply concede the out and not even bother picking up a bat?


It's negative if your raw OPS is less than half of the league OPS. The formula is (OBP/lg OBP)*100 + (SLG/lg SLG)*100 - 100.

The -100 at the end is to make a league average performance = 100, not 200. It's much better than conceding an out, if you did that your OPS would be -100.

And that is the lowest it can go, for anyone here contemplating what their metrics would look like against MLB Pitching.
   41. Fancy Crazy Town Banana Pants Handle Posted: September 18, 2017 at 02:52 PM (#5533811)
How does OPS+ go negative? Does that mean the team should simply concede the out and not even bother picking up a bat?

It basically means that their OPS is worse than half that of league average. Since OPS+ basically adds the percentages for OBP and for SLG together to get to the final result.

100 + '-50% OBP' + '-50% SLG' = 0 OPS+

Similarly a 200 OPS+ means an OPS 50% higher than league average, not twice as high. So in a .700 OPS league it is 1.050. An OPS of twice the league average (1.400) would be 300 OPS+.

(Park adjustments etc would apply to all the OPS+ numbers above as well obviously)

ETA: coke, though you can theoretically do worse than -100 in a hitters park for example
   42. Rally Posted: September 18, 2017 at 02:59 PM (#5533815)
ETA: coke, though you can theoretically do worse than -100 in a hitters park for example


How are the park adjustments done? I suppose that could be true, but it might also be something like this:

If you're hitting in a normal park, the league SLG you are compared to is .420. If you are in a hitter's park, it's instead .450. If you do it that way, your OPS+ for an 0-fer will be -100 regardless of the park.

Edit: Looks like that's how it works. Last year Jeff Hoffman and Jordan Lyles were both 0-8 in Coors, OPS+ was -100.
   43. Fancy Crazy Town Banana Pants Handle Posted: September 18, 2017 at 03:34 PM (#5533842)
How are the park adjustments done? I suppose that could be true, but it might also be something like this:

If you're hitting in a normal park, the league SLG you are compared to is .420. If you are in a hitter's park, it's instead .450. If you do it that way, your OPS+ for an 0-fer will be -100 regardless of the park.

Edit: Looks like that's how it works. Last year Jeff Hoffman and Jordan Lyles were both 0-8 in Coors, OPS+ was -100.

Oh boo. Looks like you are right. I only had the primitive form of OPS+ in my head, which BRef lists as
OPS+ = 100 * ( OBP/lgOBP + SLG/lgSLG - 1)/BPF

Which would allow for an OPS+ less than -100. But the more convoluted one (which I therefore can't be bothered to remember) they actually use, does say it applies the BPF to lgOBP and lgSLG
   44. Walt Davis Posted: September 18, 2017 at 07:03 PM (#5533972)
Sure, Ortiz retiring could always have a big impact. Except that it very rarely does -- Edgar's last great year was 2003 and he personally saw a 50 point OPS+ drop (177 points of OPS) in 2004. League average DH just went from 110 to 107.

As noted, if Hanley had just repeated his 2016, the drop in Boston would have been "just" 155 points of OPS, less than the drop when Edgar finally hit the wall. If he'd been replaced with a 2016 average DH, it would have been 265 points (I know, he was pulling up that average a good bit). Instead he was replaced by a 2017 league-average SS. The story in Boston is both Ortiz's retirement and Hanley's cratering.

In Cleveland the jump is mainly that last year they had lousy non-Santana DHs (he had a good number of games at 1B) and this year those guys are raking in non-EE starts.

may show that there's a trend toward just rotating guys through the DH slot

Nope. First, it's always been the case that a lot of teams will take the rotation approach. It's probably down a bit these days relative to 1976 but not by much. I did look at number of players with at least 80 games at DH from 2000-2017 ... won't reproduce it now but it's usually 7-9. From 2012-14 it was lower but the last 3 years it's been back to that level. Not that 80-100 games is exactly a "full-time DH" but again, it's relatively rare to find such beasties at any point.

In theory, by "full-time" we might mean 600+ PAs. In the first 20 years of the DH (excluding the 1981 split season), there were only 28 seasons matching that criterion -- with 5 of those in 1982 and 4 in 1993 and 1976, 1980 and 1987 having none. Over the last 22 seasons (skipping 1994), it's just 21, 5 by Edgar and 5 by Ortiz.

Of course it's mostly older players so maybe 600 is too strong a criteria. I'll drop it to 500 but note that will leave about 200 PA for fill-ins, that's quite a lot. (Slightly fewer now with interleague.)

You do get a lot more but in the first 20 years, it's still just 72 seasons; in the last 22, it's 72 seasons. (Canseco actually got over 500 PA in 1994.) There have been 4 already this year. Some recent history:

2007 6
2008 1
2009 2
2010 3
2011 4
2012 1
2013 4
2014 1
2015 6
2016 4
2017 4+

Some DHs still got reasonable time in the field. Up until the very end, Molitor would take about 10-30% of his starts at 1B. Santana took a lot of 1B starts. EE has been close to 50% at 1B at times. Nelson Cruz has always gotten lots of OF starts before this year, Trumbo has starting getting OF starts again.

Anyway, sure it's a fluke. A mix of Ortiz retiring while his replacement craters and Pujols craters and Det stuck with VMart, etc. About the only DH who has maintained (with the rest of the league rising) is Cruz and maybe EE is close enough. A leaguewide drop of 40 OPS points while league OPS goes up 12 points, DH not out-hitting SS? That's much more than the retirement of Ortiz.

As to things like Trumbo and Hanley needing to play the field ... I'm sure there's the occasional case but I've never bought into this. Pretty much any good hitter given a chance to adapt to the role seems to take to it quite well. IMO, the primary issue is that guys usually aren't moved there until there's pretty much no other choice which means they are marginal players by the time they get there anyway.

Trumbo of course was never a particularly good hitter anyway; Hanley's had similar seasons too -- 95 OPS+ in 2011, 105 in 2012, 89 in 2015. Over those last 7 years now, it's a 115 OPS+; over the last 4 it's 111; 2014-16 (coming into this season) it was 117. He was already a borderline hitter for a full-time DH anyway and he wouldn't be the first guy with defensive limitations to go off a cliff as a hitter at 33. But Hanley's pretty clearly been a guy focused pretty exclusively on his hitting -- if he's going to bounce back, he's probably more likely to do it as a DH than a 1B. It just seems unlikely he'll bounce back to anything more than maybe a 105-110 OPS+.

Hanley's great half-season in 2013 now stands out like a sore thumb in his career stats. It makes a big difference whether you include it in any short-term calculation. OPS+ for 2014-16 was 117; for 2013-16 it's 130. OPS+ for 2010-12 was 111; for 2010-13 it was 123. All data counts but you also shouldn't let a flukey half-season distort your perception. In real time it may not have been so obvious what an outlier that was but with hindsight, it's pretty obvious. Hanley remained a solid hitter but it's clear now that he hasn't been an excellent hitter since 2009 or 2010 (possibly robbed of that by the injuries).
   45. Morty Causa Posted: September 18, 2017 at 07:48 PM (#5534016)
Just as a matter of first impression, I would have thought DHs hit better than what those figures represent. I mean, is it really that hard to find a player who can't field but can hit? I would have said it isn't. Maybe something happens to their hitting when they don't play in the field. Maybe playing in the field does something to their mindset and mentality--make it seem as if they aren't in the game.
   46. Sunday silence Posted: September 18, 2017 at 08:09 PM (#5534030)
Id like to make a pt about OPS and by implication OPS+ which I think has been mentioned but perhaps not really fleshed out very far.

Are we really all that certain that OPS+ is the best single number to evaluate hitting? Cause it seems to me that's its treating OBP on more or less the same valuation as .slug and I dont think both of those are contributing equally to run production.

TO illustrate: I used this years league wide OBP and slug (.325 and .427). THen i produced two hitters who are both 100 OPS, but one is OBP heavy and the other is slugging heavy. Here they are

name....OBP.... .slug
p1 .487 .213
p2 .162 .640

THen I stuck a lineup of P1s and a lineup of P2s into the lineup analysis tool found at: http://www.baseballmusings.com

NOTE both players are 100 OPS players with P1 having OBP 150% that of league avg. and slug 50% of league avg. And p2 is vice versa

P1 lineup produced an average of 5.5 runs/game
p2 lineup produced average of 3.75 runs/game.

Obviously there are criticisms of this raw approach. For one thing you dont have a balanced lineup of obp guys and slug guys, but still not sure this would invalidate the approach. Another criticism would be that it doesnt use real world numbers, nobody is really OBP .162 and slugging .640. Perhaps but having put a few more numbers in the machine I am not sure that changes anything.

OPS+ simply accounts for ball park effects, if I am not mistaken so that shouldnt change the conclusion.

COnclusion: is OPS+ over estimating the value of .slug heavy guys and minimizing the value of good OBP guys?

EDIT: I deleted the link to the lineup analysis tool as it seemed to break the formatting of the page. I dont think its that hard to find just look up Baseball MUsings lineup ANalysis
   47. Sunday silence Posted: September 18, 2017 at 08:23 PM (#5534038)
So I try again this time using a 100 OPS guy who has 120% of league obp and 80% of league slug and compare him to a guy with 100 of both:

player ..... obp .... slug ... runs/game
heavy obp ... .390 .344 5.014
balancedplayer .325 .427 4.64

League avg runs per game is 4.6something so the model seems to be pretty accurate. The guy who is a 100 OPS guy who is league average in both skills scores .37 runs per game less than a guy who is weighted for OBP. Doesnt this bother anyone who uses OPS+ as a measure of off skill??


btw: I am using the 1989-2002 model (the user has a choice of two models here)
   48. Walt Davis Posted: September 19, 2017 at 01:54 AM (#5534163)
Are we really all that certain that OPS+ is the best single number to evaluate hitting? Cause it seems to me that's its treating OBP on more or less the same valuation as .slug and I dont think both of those are contributing equally to run production.

It's well-known that OPS+ under-values OBP. It's mainly used because it's familiar and there's no widely available alternative. The two most obvious alternatives are fangraphs wRC+ and b-r Rbat (or maybe Rbat per PA which already shows why Rbat isn't directly useful).

There's also a technical issue that P-I doesn't provide WAR components in the splits engine. The only other "saber" option is runs created but that's not park-adjusted. Guess that doesn't matter for league comparisons and we find that 2017 DHs have created 1066 while SSs have created 1141 but DHs don't play in NL parks. If we take it as RC/game then it's .509 for DHs and .509 for SS ... and DHs are still behind every other position but C this year. Last year DHs created 1268, a .556/game rate. Once you adjust for inter-league, I believe DH in 2016 just beat out RF for the top spot ... so from tied 1st to tied 7th.

That said:

1) Differences in the real world are rarely so extreme. For 2017 there are 7 players with an OPS+ between 95 and 105 and at least 600 PA (to keep everybody on about the same PA scale). They range from 98 to 102 in OPS+ and from -2 to +4 in Rbat. Ignoring LeMahieu's Cours-inflated stats, the raw OBPs for these players range from 295 (Duvall) to 357 (Markakis) and the SLGs from 390 (Markakis) to 487 (Duvall). Duvall's 295/487 produces one run less than Markakis's 357/390 and 2 runs more than Inciarte's 350/408 (no control for park effects). Duvall and Inciarte are both 100 OPS+, Markakis is at 98. It's just not a big deal.

The biggest "disconnect" is Longoria's 102 OPS+ but -1 Rbat compared with Gardner's 99 OPS+ but +4 Rbat. That's still marginal. And when we start talking about comparing all DHs across years or DHs to 1Bs or even SSs, you're comparing a mix of OBP/SLG combos to another mix anyway. If anything, DHs are likely to be more SLG heavy than the standard player with the same OPS+ so they are generally producing a bit less than that player. Alternatively, most of the historical difference between, say, DH and CF is probably primarily SLG with roughly similar OBPs.

2) It's not quite correct to say they are weighted equally. The formula is OBP/lgOBP + SLG/lgSLG - 1. lgSLG is of course much larger than lgOBP so a 1 point increase in SLG is worth less than a 1 point increase in OBP. This year, for a AL average context, one point of OBP is equivalent to 1.32 points of SLG in OPS+ terms (lgSLG/lgOBP). That's still not as much of a boost as most analysis says it should be (somewhere around 1.6 to 1.8 if I recall).

3) Really the "+" measures should make an adjustment for the standard deviation as well. This probably won't make a big difference and is partly captured by the mean differences (because the mean and variance are related for rate/counting stats) but in the current crazy environment, I wouldn't be surprised to find that the variance of BA and ISO are higher than usual.

Looking back at DH vs SS in 2017, the numbers are essentially identical: OBP 319 vs 318, SLG 420 vs 421 (yes SS are out-slugging them). The main difference is that SSs are more BA-heavy, lower BB, lower ISO ... again these all generally balance out. In 2016, DHs were 328/452 while RF were 332/438. By RC/game they were nearly identical with RFs 4 point edge in OBP not quite matching DHs 14 point edge in SLG, in terms of RC/g. Thinks might differ a bit if you prefer RC/650 PA or RC/27 outs but probably not a lot.
   49. Sunday silence Posted: September 19, 2017 at 02:12 AM (#5534167)
I wasnt trying to use it in the DH discussion in particular but just had some general questions about it. However, thanks for the extended discussion; I'll need a few days to digest all that. thanks again for all that.
   50. Walt Davis Posted: September 19, 2017 at 02:13 AM (#5534168)
Willie Randolph may be the best modern equivalent of what you're interested in. A career line of 276/373/351 and 104 OPS+. That's an OBP+ around 115 (so a SLG+ around 89). I don't know what we'd expect a 104 OPS+ to translate to over 650 PA but probably something like +4. Using that we'd expect him to have produced about 58 runs above-average offensively while he's credited with 120. So that's 62 runs spread across 14.5 seasons, 4-5 runs a year.

The opposite extreme Adam Duvall type in Randolph's context would have an OBP around 290 and a SLG around 450. Conveniently that's very close to Tony Armas. Based on Rbat, that gap is about 8 runs per 650 in Randolph's favor -- 8 runs a year is pretty significant. Or it's about 7 extra WAR over the course of Armas's career. So that's a realistic extreme case.

If b-r added something like wRC+ to a player's main stat line and to P-I, we'd probably shift towards using that. Until then, OPS+ is generally a perfectly good approximation except in extreme cases like Randolph-Armas in which case hopefully the writer points out that Randolph was the more productive offensive player.

EDIT: So for example Randolph has a 104 OPS+ but a 110 wRC+; Armas is at 103 OPS+ and 100 wRC+. Rbat puts Armas at -2 in 5502 PA so also pegs him as dead average. Anybody have any idea what the largest wRC+ vs OPS+ differences are?
   51. Walt Davis Posted: September 19, 2017 at 02:32 AM (#5534172)
Some answers to my query in the edit (wRC+, OPS+)

Luis Castillo 97 92
Butler 115 110
Batista 89 93
Jose Hernandez 86 88
Gallo 113 112
Votto 158 158
Rose 121 118
Mauer 124 126 (surprise ... I figured a higher wRC+ with that OBP)

Maybe I'm forgetting some obvious ones to look up but it seems Randolph probably is one of the most extreme, at least among post-expansion players.
   52. Fancy Crazy Town Banana Pants Handle Posted: September 19, 2017 at 04:59 AM (#5534179)
John McGraw 135 OPS+ 142 wRC+

Biggest I could find.

edit misread a line, removed one entry

Single season, and very limited playing time, but Chone Figgins 2014 (.373 OBP; .267 SLG) was good for 87 OPS+ and 100 wRC+.
   53. Fancy Crazy Town Banana Pants Handle Posted: September 19, 2017 at 05:34 AM (#5534180)
Biggest difference I can find the other way around is Albert Belle 144 OPS+ 139 OPS+ (insert obligatory joke here).
   54. Rally Posted: September 19, 2017 at 08:37 AM (#5534198)
Just as a matter of first impression, I would have thought DHs hit better than what those figures represent. I mean, is it really that hard to find a player who can't field but can hit? I would have said it isn't. Maybe something happens to their hitting when they don't play in the field. Maybe playing in the field does something to their mindset and mentality--make it seem as if they aren't in the game.


It's a combination of things. Many teams don't want to spend much on a DH, so they rotate players in and out. In general, players do hit better when playing the field than when serving as a DH. It's a rare player who has the mindset to succeed in the DH role.

The best hitters in the game are players in their prime years. There are exceptions like Frank Thomas or Jason Giambi but most great hitters between ages 25-30 have skills to play the field too. Looking at the AL top 10 OPS+ Nelson Cruz is a DH and Jose Abreu a first baseman who could be a DH. The rest all have significant value in the field. Same with Donaldson and Trout, the recent MVPs who miss the OPS+ leaderboard right now due to missed injury time.

Then you've got the Pujols types - players with big contracts whose skills have deteriorated to the point they can't play the field anymore. You've got to do something with them (until the skills are so bad you cut the sunk costs) so they get the DH spot. See also Hanley Ramirez, Victor Martinez, and soon Miguel Cabrera will join the list.
   55. jacksone (AKA It's OK...) Posted: September 19, 2017 at 09:27 AM (#5534223)
Walt - your post #48 is great, I will second the big thanks from Sunday. It really helps to have these micro discussions on individual stats/measurements to really understand correctly what the numbers are saying and your breakdowns are very helpful.
   56. Sunday silence Posted: September 19, 2017 at 09:41 AM (#5534232)
yeah Im sure I wasnt breaking new ground there but I didnt know where else to look for an explanation of all that. Not sure what the pt of OPS is after all that. Rbat + Rbase would be a better measure.
   57. Walt Davis Posted: September 19, 2017 at 06:11 PM (#5534704)
#56: We probably would be better off with something other than OPS+. But tradition is hard to overcome and exposure is difficult. The interesting question might be why none of the alternatives (EQA, wRC+, wOBP, etc.) have ever caught on while OPS+ has. I suspect it's just because OPS+ is sitting right there in the main stats table on b-r. But WAR isn't in the main stats table and it's caught on well. But given the immediate rebellion (including me) when Sean changed the ERA+ formula to improve it, I can understand he has to tread carefully in messing with b-r stats and tables.

Rbat doesn't work well as a replacement because it's a counting stat not an adjusted rate stat like OPS+. I'm sure you could come up with a sensible rate version but wRC+ probably is the closest to OPS+. Anyway, we have the technology!

Why can't they find DHs? #54 nails most of this -- good hitters are also usually decent fielders and, by the time they're not, they're usually just OK, declining hitters. Also, outside of this year, the break-even point for a true full-time DH seems to be around a 125 OPS+ (based on historical usage) and there just aren't that many hitters that good. But it is pretty easy to find guys who can field but not hit. It's interesting that apparently teams would rather cycle a bunch of guys who'll cobble together a 105 OPS+ (or even employ Kendrys Morales) rather than, say, add a really good fielder somewhere, bump other players down 1 defensive step and shove Cabrera into DH even though he's a capable 1B. Is Morales really a better option than finding, say, a CF who flied but posts a 65 OPS+, move your CF to LF, move your LF to 1B (or DH) and your 1B to DH? Or an awesome SS, move your SS to 3B, your 3B to 1B and your 1B to DH?

And thanks for the compliments. And I didn't mean the "OPS+ undervalues OBP is well-known" to be dismissive -- everybody discovers this when they discover it. There were some posts by Tango ... god, probably something like 7 years ago now ... where he goes pretty thoroughly into a proper valuation of OBP and how OPS+ could be fixed. Those might just have been posts here (check one of the old specialty blogs maybe) or at his site. My memory is he came up with the 1.6-1.8 range I mentioned but memories are fuzzy.
   58. Fancy Crazy Town Banana Pants Handle Posted: September 19, 2017 at 06:27 PM (#5534711)
#56: We probably would be better off with something other than OPS+. But tradition is hard to overcome and exposure is difficult. The interesting question might be why none of the alternatives (EQA, wRC+, wOBP, etc.) have ever caught on while OPS+ has. I suspect it's just because OPS+ is sitting right there in the main stats table on b-r. But WAR isn't in the main stats table and it's caught on well. But given the immediate rebellion (including me) when Sean changed the ERA+ formula to improve it, I can understand he has to tread carefully in messing with b-r stats and tables.

Bref is significantly more user-friendly than fangraphs. I am not going to go to fangraphs if I can avoid it. OPS+ and WAR are both easy to find on bref. WAR is the center piece of the value table, the fact that you have to look at the second table is no big deal. It's the first thing you will see when you go to leaderboards, the default for sorting player tables etc. It's not really that difficult to see why it caught on.

rather than, say, add a really good fielder somewhere, bump other players down 1 defensive step and shove Cabrera into DH even though he's a capable 1B.

Cabrera from all I have heard, is adamant about playing the field. Before this season, it seems pretty reasonable to just keep him happy, and let him rake. He is your most important and most valuable player... You also still need to find something to do with VMart then, who can't play the field at all at this point.

Now? Well he has probably reached the point where you push him to DH, if you think it will help him stay/get healthy, and hopefully rebound somewhat. Still leaves you having to cut bait with VMart, though he is hitting poorly enough that it is probably the right call no matter what.
   59. Sunday silence Posted: September 19, 2017 at 08:10 PM (#5534734)
Rbat doesn't work well as a replacement because it's a counting stat not an adjusted rate stat like OPS+


I think you mentioned this in the previous post as well. Why cant you use WAR per 650 AB or something? Or off WAR per 650 AB.

And I didn't mean the "OPS+ undervalues OBP is well-known" to be dismissive -


I didnt take it that way I was just discovering that on my own as you alluded to. It came as a shock because I was actually expecting the .200/.250/.500 guy to be better than the .250/.420/.260 guy but the OBP is significantly better.
   60. Fancy Crazy Town Banana Pants Handle Posted: September 19, 2017 at 10:10 PM (#5534814)
I think you mentioned this in the previous post as well. Why cant you use WAR per 650 AB or something? Or off WAR per 650 AB.

Well WAR would not be a purely hitting stat, but would include defense and baserunning. You could use something like rbat/PA, but the main issue is that it just isn't easily available. Nobody wants to mess around working those things out every time on their own. They want to look up the number and not do any extra work if necessary.
   61. Sunday silence Posted: September 20, 2017 at 06:12 AM (#5534904)
well I was suggesting the offensive component of WAR, whether its WARbat or whatever. But it leads to the obvious question: Is WAR adjusted for park effects? Walt suggests wRC+ which I cant recall if that is a different measure or what? Presumably it is park adjusted.
   62. Fancy Crazy Town Banana Pants Handle Posted: September 20, 2017 at 06:23 AM (#5534906)
well I was suggesting the offensive component of WAR, whether its WARbat or whatever.

rBat is the batting component of WAR. It can be found in the same table on bref as WAR, as can all the other components of WAR. But as noted, it is a counting stat, and not a rate stat.

Is WAR adjusted for park effects?

Yes.

Walt suggests wRC+ which I cant recall if that is a different measure or what? Presumably it is park adjusted.

It is essentially the same as rbat, but expressed as a rate (i.e. on a per PA basis), and scaled like OPS+, so that 100 is league average. But as noted can only be found on fangraphs.
   63. jacksone (AKA It's OK...) Posted: September 20, 2017 at 06:54 PM (#5535406)

Cabrera from all I have heard, is adamant about playing the field.


I've argued this point when discussing Ortiz. Some players just do not want to be a DH. It's been drilled into them that it's equivalent to the kicker in football - not a real position. Adam Dunn and Gary Sheffield being two prominent, #### defensive, great bat players both publicly expressed an unwillingness to be a DH.

move your CF to LF, move your LF to 1B (or DH) and your 1B to DH? Or an awesome SS, move your SS to 3B, your 3B to 1B and your 1B to DH?


This I suspect is MUCH easier said than done. There is prestige in being the SS/CF, just ask Derek Jeter how he felt about moving to a different position...

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Support BBTF

donate

Thanks to
Guts
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOTP 16 October 2017: Sorry, Yankee fans: Trump’s claim that he can ensure victory simply isn’t true
(1334 - 4:11pm, Oct 19)
Last: Joe Bivens Will Take a Steaming Dump

NewsblogDallas Keuchel embraces the ‘calmness’ of being Yankees killer | New York Post
(59 - 4:11pm, Oct 19)
Last: Bote Man

NewsblogNLCS Game 5 OMNICHATTER, for October 19, 2017
(7 - 4:09pm, Oct 19)
Last: Sunday silence

NewsblogPrimer Dugout (and link of the day) 10-19-2017
(9 - 4:08pm, Oct 19)
Last: Batman

NewsblogOT: New Season August 2017 Soccer Thread
(1168 - 4:04pm, Oct 19)
Last: I am going to be Frank

NewsblogLCS OMNICHATTER for October 18, 2017
(464 - 4:04pm, Oct 19)
Last: Pat Rapper's Delight (as quoted on MLB Network)

NewsblogOT - NBA 2017-2018 Tip-off Thread
(356 - 3:50pm, Oct 19)
Last: jmurph

NewsblogTV Simpsons documentary about 'Homer at the Bat' episode to air Sunday
(17 - 3:46pm, Oct 19)
Last: Bote Man

NewsblogALCS Notebook: Cashman on Yankee Analytics, Luhnow on Hiring Hinch | FanGraphs Baseball
(2 - 3:22pm, Oct 19)
Last: The Yankee Clapper

NewsblogCubs hold off Dodgers in Game 4 of NLCS | MLB.com
(33 - 3:21pm, Oct 19)
Last: Bote Man

NewsblogYankees, Masahiro Tanaka take 3-2 ALCS lead | MLB.com
(3 - 2:53pm, Oct 19)
Last: ERROR---Jolly Old St. Nick

NewsblogHow the mob, politics, a burlesque dancer and The Babe shaped the '32 World Series
(5 - 2:41pm, Oct 19)
Last: There are no words... (Met Fan Charlie)

Gonfalon CubsFive minute Los Angeles Dodgers Preview
(72 - 2:18pm, Oct 19)
Last: Barry`s_Lazy_Boy

NewsblogMemes roast Astros after unexpected loss to New York Yankees - Houston Chronicle
(13 - 12:32pm, Oct 19)
Last: bfan

NewsblogAngell: Yanks Get Even
(12 - 12:26pm, Oct 19)
Last: dlf

Page rendered in 2.7536 seconds
47 querie(s) executed