SoxProspects News
|
|
|
|
Legal
Forum Ground Rules
The views expressed by the members of this Forum do not necessarily reflect the views of SoxProspects, LLC.
© 2003-2024 SoxProspects, LLC
|
|
|
|
|
Forum Home | Search | My Profile | Messages | Members | Help |
Welcome Guest. Please Login or Register.
|
Post by pedrofanforever45 on Jul 23, 2018 21:28:10 GMT -5
Truth. Can't trust the guy.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,933
|
Post by ericmvan on Jul 24, 2018 1:33:23 GMT -5
Truth. Can't trust the guy. Incredibly poorly is just plain wrong. I have his adjusted xwOBA as .304, which makes him an average bullpen arm in this age of loaded pens. There are big names who have been less good, e.g. Kelvin Herrera is .308 and Raisel Iglesias is .313. It's still a lot better than the average starter (.330 or so), but it's also true that it's the sort of number you want from the 5th guy in your pen, not the 2nd or 3rd. And right now he's either the 5th or 6th guy in the pen, depending on Thornburg.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,933
|
Post by ericmvan on Jul 24, 2018 3:24:28 GMT -5
So, all of the guys who: 1) Pitch for non-contenders 2) Have 3 years of control or less 3) Have pitched well enough this year, according to adjusted expected wOBA (jxwOBA), to be one of the top 3 setup guys on a contender. jxwOBA tells you how well they've actually pitched, based on strikeouts and walks (adjusted by me for ballpark) and launch angle / exit velocity (adjusted by me for balls hit to CF, which are good), all of it adjusted for quality of hitters faced.
The second number, WPA-, is the actual effectiveness, in Win Probability Added adjusted for game leverage, scaled so that 100 is average and the range is like that of ERA-. Those are the only two numbers that matter: pure results off the bat (or missing it) and pure results on winning. Any other number you look at is intermediate between these two, a way of explaining why they might differ. YOC is years of control. The final number, QO, is quality of opposition split, and is the career difference between OPS allowed to 3 and 4 hitters versus 7 through 9 (excluding pitchers). It's really only meaningful for veterans who have compiled a decent number of innings. More on that in the comments.
I've included the Sox for comparison. One thing that jumps out at you is how good the Sox' WPA- is relative to jxwOBA. That is the sign of a brilliantly managed pen (technically known as a non-Farrelistic bullpen). When leverage is low, you just send guys out to get work without regard as to how they match up with the opposition hitters. When leverage is high, you attend to those matchups. When leverage is low and a guy is not sharp, you can leave him in to give up some damage, but when leverage is high and he isn't sharp, you have a quick hook. If a streaky guy is going well, you move him into high leverage and when he starts to slump, you move him into a mopup role. Joe Kelly's jxwOBA of .304 is not good enough to make this list but his 63 WPA- is topped by only 4 of the trade candidates. Wow.
In contrast, the Twins have 5 guys good enough to make this list (including two guys who are pre-arb) and they average 104 WPA-, with four of them over 100. I wouldn't have thought that was possible. Someone should take away Paul Molitor's dartboard, eh? Name H Tm jxwO WPA- YOC QO Ky. Barraclough R MIA .253 106 3 Craig Kimbrel .255 11 Zach Duke L MIN .259 127 FA Kirby Yates R SDN .260 16 2 297 Tony Barnette R TEX .260 88 3 Ryan Brasier .262 Ryan Tepera R TOR .267 90 3 Tyler Clippard R TOR .269 110 FA Seunghwan Oh R TOR .270 78 3 Keone Kela R TEX .272 56 3 Craig Stammen R SDN .275 72 FA 201 Joakim Soria R CHA .275 73 opt 256 Sergio Romo R TBA .276 71 FA 115 Jared Hughes R CIN .276 56 1+O -56 Mychal Givens R BAL .276 79 3 Matt Barnes .276 64 Ryan Pressly R MIN .281 105 1 Shane Greene R DET .283 105 2 Fernando Rodney R MIN .288 101 FA Luis Avilan L CHA .290 89 1 John Axford R TOR .291 95 FA Heath Hembree .293 72 David Hernandez R CIN .293 82 1 Adam Conley L MIA .297 64 3 Jake Diekman L TEX .297 63 FA 100 Robbie Erlin L SDN .299 58 2 Kirby Yates remains the only stud worth paying a premium for. He'd give you a monster 8th inning guy and then two years of potential Kimbrel replacement. I've thrown in his QO split just to show how misleading it can be for a guy like him who's had a breakthrough, because this year it's negative. The reason he's now so desirable is that he's started getting elite hitters out.
Zach Duke and Jake Diekman are viable alternatives to Zach Britton if they get outbid for him, which seems likely, and feel they need to add a LHR. I'm not sure the latter is actually a need, given the very solid probability that either Pomeranz or Price will be in the pen for the post-season.
Sergio Romo is far and away the most desirable of the guys who are essentially Matt Barnes, and you might well want to get one of those guys if you think that Brasier's just on a hot streak and/or if Thornburg fails to continue like his last spectacular outing. His season is indistinguishable from Soria's (and Stammen's) but he does not have a career track record of feasting on hitting dregs. The better the quality of opposition, the more of an edge he has on Soria.
Jared Hughes might be better than Romo and has the extra years of control. Is he good enough to warrant the acquisition price? I'm not sure.
I think that those four or five guys and Britton constitute your shopping list. I can see them adding no one.
|
|
|
Post by pedrofanforever45 on Jul 24, 2018 5:09:52 GMT -5
Truth. Can't trust the guy. Incredibly poorly is just plain wrong. I have his adjusted xwOBA as .304, which makes him an average bullpen arm in this age of loaded pens. There are big names who have been less good, e.g. Kelvin Herrera is .308 and Raisel Iglesias is .313. It's still a lot better than the average starter (.330 or so), but it's also true that it's the sort of number you want from the 5th guy in your pen, not the 2nd or 3rd. And right now he's either the 5th or 6th guy in the pen, depending on Thornburg. Kelly just doesn't instill a lot of confidence in anybody these days, including manager Alex Cora who only used him in a 9-1 ballgame recently. Joe Kelly's best pitch is his change up, which explains his success against LHB throughout the year. Even still, the guy gets hit (usually) hard in outings he gets through (which you have mentioned yourself FWIW). Unless the guy is on a nice hot streak, you can't trust him. Even then, I wouldn't feel great about it after understanding his inconsistencies from one month until the next. The guy is so bad during his inconsistent streaks that it makes you wonder when or if he's ever turning it around. I'm not for trading the guy. He's earned enough to stay on this roster for at least this season, but I'm not thinking this guy is a viable playoff arm unless it's a blowout. Sorry, I can't do it. I've been wishy washy on Kelly myslef throughout the year (how can you not be with his inconsistencies?), but I wouldn't throw guaranteed money at this guy either past 2018. Great one advanced metric tells you otherwise and what not. The team is telling me different. Right now I have a hard time believing Kelly is even ahead of Workman in terms of the depth chart of bullpen arms to trust right now. The Sox should find another arm for the bullpen, and they will whether it's Joe Smoe or "it's hey we have something interesting here." Throw a dart, who freaking knows, maybe you land on bullseye. That's where I'm at. You don't necessarily *need it,* but it can't certainly hurt.
|
|
|
Post by umassgrad2005 on Jul 24, 2018 5:56:43 GMT -5
I do like Sergio Romo.
|
|
|
Post by rjp313jr on Jul 24, 2018 6:17:37 GMT -5
Sorry this is a subscription required article but it’s worth a read: theathletic.com/442639/2018/07/24/after-years-in-limbo-heath-hembree-finds-a-role-as-alex-coras-problem-solver/It breaks down how Hembree has become the guy who Cora goes to when guys are on base. He excelled in that role to the point where, he’s very rarely used now in a clean inning. He comes in with inherited runners. “In the month of July, Hembree has pitched seven times, five of them coming into the middle of an inning with runners on base. He’s inherited nine runners this month, four of them in scoring position, and has stranded all nine. He has not let an inherited runner score since the end of April.” “Since June 1, though, Hembree has held hitters to a .150 average with 26 strikeouts in 16 2/3 innings., allowing only four runs to bring his season ERA down to a respectable 3.73. His worst innings have been the clean innings. He’s been at his best when things are messy.”
|
|
|
Post by philsbosoxfan on Jul 24, 2018 8:14:11 GMT -5
Sergio ? (Compliments of the way back machine)
|
|
|
Post by jmei on Jul 24, 2018 8:48:47 GMT -5
Judging relievers almost exclusively based on a half-seasons worth of performance is a bad idea.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,933
|
Post by ericmvan on Jul 24, 2018 10:46:04 GMT -5
Judging relievers almost exclusively based on a half-seasons worth of non-Statcast performance data is a bad idea. FIFY.
If you look at the disparity between xwOBA and wOBA (not to mention wOBA versus whatever else you like), you understand why it's a terrible idea.
My assumption is that adjusted xwOBA by the time the season is two-thirds done is going to be hugely more predictive than anything we've had in the past. You've literally removed all of the readily measurable luck. The only deeper metric would be to have an actually accurate system of predicting batted ball results from pitch/fx data, to factor out the guys who have gotten away with bad pitches from the guys who have had good ones hit hard. But separating out the sequencing skill in that is a ferocious problem.
I'm not sure that looking at what a guy did in previous years is actually beneficial. Guys make real changes from year to year much more often than they do in the course of this season. This year's jxwOBA of a veteran is going to tell you how the guy has been feeling, on average.
|
|
|
Post by jmei on Jul 24, 2018 12:30:06 GMT -5
Performance variance is a lot more than just batted ball luck, especially in 30-40 inning samples. Adding prior year performance will almost certainly increase predictiveness.
|
|
|
Post by Oregon Norm on Jul 24, 2018 13:16:39 GMT -5
I've mentioned this before. Examining the variance in the performance of starters, say over a 30-40 game running average, would likely result in the same sort of fluctuations as the year-over-year variability of relievers. The difference is that starters can get a few hundred innings in during any given season to stabilize performance metrics. That affects the perception we all have of who's having a good season, and who isn't.
|
|
|
Post by Addam603 on Jul 24, 2018 14:00:53 GMT -5
|
|
|
Post by maxwellsdemon on Jul 24, 2018 15:23:13 GMT -5
It ain't the market value until there's a trade. Until then it's just the offer and we haven't heard the bid side yet.
|
|
|
Post by GyIantosca on Jul 24, 2018 16:56:49 GMT -5
The Sox should give a couple of prospects a chance before the deadline maybe in low leverage situations. I like to see Buttrey and Lakins. Kelly needs a dl stint. Something not right with him. I never feel 100% with Barnes but I like a couple of his pitches a lot. He does look like he is rounding into a solid piece.
I just hate trading for relievers. I don't mind as much at the deadline because your rounding your team weaknesses for the stretch drive. But the offseason moves have to stop 100%. Except for closers.
|
|
redsox04071318champs
Veteran
Always hoping to make my handle even longer...
Posts: 15,666
Member is Online
|
Post by redsox04071318champs on Jul 24, 2018 17:17:19 GMT -5
An article on WEEI.com mentions Yates and Stammen as guys the Sox are interested in as well. Not exactly earth shattering news here, but two more names to keep in mind as possibilities.
|
|
|
Post by telson13 on Jul 24, 2018 21:34:31 GMT -5
I've been all in on the Brasier hype train since day one. Now the numbers are backing it up finally. I give you credit on that one, you def were. Let’s hope that train keeps rolling and gaining steam.
|
|
|
Post by James Dunne on Jul 24, 2018 21:36:15 GMT -5
Remember when Daniel Bard threw that pitch that was 99 and also basically went sideways? That was cool.
|
|
|
Post by kingofthetrill on Jul 24, 2018 21:44:56 GMT -5
Remember when Daniel Bard threw that pitch that was 99 and also basically went sideways? That was cool. I think someone had that gif as their avatar for a while.
|
|
|
Post by telson13 on Jul 24, 2018 21:50:57 GMT -5
Judging relievers almost exclusively based on a half-seasons worth of non-Statcast performance data is a bad idea. FIFY.
If you look at the disparity between xwOBA and wOBA (not to mention wOBA versus whatever else you like), you understand why it's a terrible idea.
My assumption is that adjusted xwOBA by the time the season is two-thirds done is going to be hugely more predictive than anything we've had in the past. You've literally removed all of the readily measurable luck. The only deeper metric would be to have an actually accurate system of predicting batted ball results from pitch/fx data, to factor out the guys who have gotten away with bad pitches from the guys who have had good ones hit hard. But separating out the sequencing skill in that is a ferocious problem.
I'm not sure that looking at what a guy did in previous years is actually beneficial. Guys make real changes from year to year much more often than they do in the course of this season. This year's jxwOBA of a veteran is going to tell you how the guy has been feeling, on average.
I think it’s fair to look at past performance to identify trends though. I think it’s probably rolled into sequencing, new pitches/approaches, etc. But I have a tendency to agree that Statcast data, while not perfect, is the best and most predictive information. Obviously shifts, defensive positioning in/deep, etc. have an effect and probably lag some behind real-time performance; also the expected numbers are league averages on massive data sets and are normalized and thus exclude team defense effects. It also fails to account for the effects of a player’s own tinkering due to bad luck (or good luck), which can disrupt the “all things the same” equation. But yeah, my sense is that those effects are minor especially on a data-driven team that rapidly adjusts, and regression towards expecting is likely the major driver of future results. For example, Hembree’s elimination of his platoon splits, his sustained high SwStr rate, his SL use/location, and the positive trends of his contact% and O-swing all are suggestive of WHY one can believe in the value of his Statcast-predicted xresults ...but the Statcast data are likely to “lead” real-world actualization reflecting the sum benefit of those developments.
|
|
|
Post by telson13 on Jul 24, 2018 21:52:11 GMT -5
|
|
|
Post by telson13 on Jul 24, 2018 21:55:00 GMT -5
An article on WEEI.com mentions Yates and Stammen as guys the Sox are interested in as well. Not exactly earth shattering news here, but two more names to keep in mind as possibilities. Idk what, exactly, they hope to gain unless it’s the installation of Barnes as the closer in the offseason and Yates as the setup. Otherwise, that’s a hefty prospect price to pay. Idk how I feel about that...
|
|
redsox04071318champs
Veteran
Always hoping to make my handle even longer...
Posts: 15,666
Member is Online
|
Post by redsox04071318champs on Jul 24, 2018 22:06:18 GMT -5
An article on WEEI.com mentions Yates and Stammen as guys the Sox are interested in as well. Not exactly earth shattering news here, but two more names to keep in mind as possibilities. Idk what, exactly, they hope to gain unless it’s the installation of Barnes as the closer in the offseason and Yates as the setup. Otherwise, that’s a hefty prospect price to pay. Idk how I feel about that... I'm with you. I'd rather grow bullpen help and then rent help as needed. I don't see the need for the Red Sox to give up a crazy about for "controllable" years. Between a healthier Thornburg, Barnes, maybe Smith at some point, Hembree, Poyner, Buttrey, Brasier, and perhaps Kimbrel, I don't see why the Sox can't start 2019 with these pitchers, see if Buttrey or even Lakins can turn into something or somebody else is lightning in a bottle. Early in the season you have time to experiment. If it doesn't work, find a rental reliever to trade for. But don't give up really good prospects for relievers (unless they're elite like Kimbrel who had an amazing track record) like Yates.
|
|
|
Post by telson13 on Jul 24, 2018 22:21:27 GMT -5
Idk what, exactly, they hope to gain unless it’s the installation of Barnes as the closer in the offseason and Yates as the setup. Otherwise, that’s a hefty prospect price to pay. Idk how I feel about that... I'm with you. I'd rather grow bullpen help and then rent help as needed. I don't see the need for the Red Sox to give up a crazy about for "controllable" years. Between a healthier Thornburg, Barnes, maybe Smith at some point, Hembree, Poyner, Buttrey, Brasier, and perhaps Kimbrel, I don't see why the Sox can't start 2019 with these pitchers, see if Buttrey or even Lakins can turn into something or somebody else is lightning in a bottle. Early in the season you have time to experiment. If it doesn't work, find a rental reliever to trade for. But don't give up really good prospects for relievers (unless they're elite like Kimbrel who had an amazing track record) like Yates. The one thing I do really like about Yates is that his outstanding results this year include a jump in GB rate and what looks like a BABIP skill, exactly in time with the addition of a splitter. So I do think his results this year are more easily extrapolated than if, say, he were still leaning on the slider. I think he is clearly a superior version of himself. He’s been elite, and I think there’s a good chance he remains so moving forward. But yeah, idk...depends on who they have to give up. If he makes letting Kimbrel go possible, it saves them $15+ M a year, until he becomes a FA. He’s 31, so...idk. FB only sits 94, but Koji sat 88 with the great splitter, so... Ugh. He’d be a great get but giving up Groome or Mata, or Hernandez? Yuck. Chavis and Shawaryn and a guy like Joan Martinez I’d be OK with, if Kimbrel walked and they got under the lux tax.
|
|
|
Post by telson13 on Jul 24, 2018 22:26:27 GMT -5
I want no part of giving up extra for Stammen, who’s 34, throws 91, and while he’s on a great deal, I’m not sold that this isn’t just noise this year. Ugh...I guess the best trades hurt both parties, but boy, would losing Groome or Mata (or Houck, who’s really looking like the real deal these days) hurt. They really need to learn to develop their own high-end relievers.
|
|
|
Post by pedrofanforever45 on Jul 24, 2018 22:28:41 GMT -5
I want no part of giving up extra for Stammen, who’s 34, throws 91, and while he’s on a great deal, I’m not sold that this isn’t just noise this year. Ugh...I guess the best trades hurt both parties, but boy, would losing Groome or Mata (or Houck, who’s really looking like the real deal these days) hurt. They really need to learn to develop their own high-end relievers. The high end relievers are coming, but unfortunately they are a year away (think second half of 2019). There in lies the problem, but it makes the loss of Kimbrel more palatable.
|
|
|