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.
Discussion of 2014 and 2015 pitching rotations
|
Post by James Dunne on Oct 12, 2014 8:21:21 GMT -5
Most of the damage that 3/4 hitters did against Cueto this season were in the form of home runs - he allowed 13 to those two batting positions, and only 9 to the other lineup spots combined. His strikeout and walk numbers against those spots remain excellent - less excellent than against other lineup spots, sure, but not out of line with a traditional split. Fortunately, he was so dominant against the other lineup spots that it mostly resulted mostly in solo home runs.
His lines are also inflated because he pitched poorly against 3/4 hitters in 2008 and 2009. That was, of course, before he was even a good pitcher. He was excellent against #3 hitters from 2010 to 2013. He has, apparently had intermittent problems with cleanup hitters. It's a concern, but I'm going to attempt to assuage it for you.
This profile strikes me as very, very similar to Curt Schilling. Looking back at his splits, Schilling allowed a .279/.331/.489 line to #3 hitters over his career. From 2000 to 2002, the most dominant stretch of his career, Schilling allowed a .580(!) slugging percentage to #3 hitters. Now granted, offensive numbers were different in that era, but still - those numbers came from an approach to pitching. Schilling was often pitching to the #3 hitter with two outs and the bases empty. Always wanting to be efficient with his pitch count and wanting to avoid walks, particularly in the early innings, he would pitch to that #3 hitter aggressively. That would result in often missing onto, rather than off of the plate. Missing against #3 hitters in 2001 led to home runs. Home runs are bad, but solo home runs on the second pitch of an at-bat in the first inning are definitely the best kind of home runs to allow.
If - and this would be a big if - Cueto was underperforming his fantastic overall peripherals, then struggling against those middle-order bats would be a concern because it would show a problem not just with good hitters, but with sequencing around those best hitters. I find it helpful to think of it as the Javier Vazquez effect. Cueto's performance record is much more similar to Schilling than Vazquez - he's a pitcher who works aggressively against the best hitters, particularly in the early innings. It results in some short-term agony in the form of the solo home run, but it also results in good K rates, low walk rates, and him getting through early innings efficiently.
I'll admit bias on this though - I really like Cueto and have always enjoyed watching him pitch.
As a tangentially-related question not necessarily directly related to Cueto: is it worse to be bad against the two best hitters in the lineup and dominant against the rest, or generally more consistent but less good all the way through? I would almost think the latter would be worse, right? The chances of piecing together something like a four-run inning seem like they would be much higher when you have several .310 OBP/.350 SLG guys than it is when you have two .350 OBP/.480 SLG players and the rest are all punchless .250 OBP/.300 OBP types. I'm legitimately asking this, though. I'm guessing there has been research into this, but I don't really know of it and I'd be interested to see it.
|
|
|
Post by jclmontana on Oct 12, 2014 10:09:51 GMT -5
I'll be surprised if Lester doesn't wind up on the Cubs. Fit makes too much sense. I, too, would be stunned if Lester doesn't wind up with the Cubs. The Cubs have a great young offense coming along. They need an innings guy and Lester doesn't cost them a draft pick or young talent, just a lot of money, and with all those young kids coming along there's plenty of room in the payroll for Theo to splurge. Where the Sox are hesitant, Theo won't be. Lester will wind up getting a huge contract with the Cubs. Eh, the safe bet is always to take the field. There are logical reasons for about half the teams to shell out for Lester, as well as logical reasons to stay away. This kind of certainty about where free agents will end up is pretty silly. Remember when Crawford was a near lock to go to the Angels? And Matt Holliday was going to be the next example of Yankee free agent hegemony?
|
|
|
Post by jmei on Oct 12, 2014 10:38:29 GMT -5
Well, if we're going to drill down to this level of detail, the first thing we want to do is remove Wright's April 2013 roster-emergency MLB debut, because he didn't become the pitcher that has a few of us excited until the middle of that year. It's completely irrelevant to his projection. So he's a guy with a 66 ERA-, 78 FIP-, 82 xFIP-, and 2.97 SIERA in 30.2 IP. There are 272 MLB pitchers who have thrown 25 or more innings and at least 3 IP per G (Wright is 3.4), 2013-2014. Wright-minus-debut ranks tied for 8th, tied for 17th, tied for 16th, and (after adjusting for league) 12th in the four metrics. It has been a spectacularly good 30 innings for a guy who has more often that not faced guys a second time in a game. So, what about that? I happen to have just looked into the relief benefit in a lot of detail, because I wanted to adjust the Davenport translations of piggyback starters and relievers who routinely pitch 2-3 innings, which doesn't happen much in MLB (the average relief appearance this year was 1.0 innings long). Tango et al in The Book came up with an 0.80 relief benefit (I'm not sure where you're getting 1.00). It turns out that 0.40 of that is the times-around-the-order effect. The other 0.40 has to derive from the fact that relievers can air it out more when they're not expecting to pitch more than an inning; guys in relief routinely throw harder than they do as starters. Well, there's no reason to expect a knuckleball pitcher to have either one of these things happening. They're throwing the same pitch mix from the get-go, and hitters seeing them a second or third time in the game can't get used to the break or spin of their primary pitch, since it's random. And obviously they're not exerting extra effort in shorter stints. Even ignoring the cherry-picked removal of his worst game, you still have to adjust for the fact that he's thrown almost all those innings in relief. Using his IP/G as a filter is misleading-- there are pretty much no more classic long-men in the game anymore, and there are basically no qualified relievers who average even 2 IP/G, so you're still making the same mistake by comparing Wright only to starters. And Wright absolutely still benefits from making relief appearances. For one thing, the times-through-the-order penalty escalates as pitchers face a hitter the second, third, and fourth times through the order, which means comparing Wright to starters who almost always go through the lineup at least three times is a skewed comparison. Hitters would absolutely benefit from seeing Wright a second and third time, actual results in a tiny sample notwithstanding. The sheer unconventional nature of the pitch (it has abnormally slow speed and unusual movement) as well as its scarcity (two pitchers in the league throw it) means that if anything, knuckleballers might have a steeper times-through-the-order effect than most pitchers, especially in relief appearances when hitters aren't expecting it and might not have read their scouting reports. You're right that airing it out matters less for Wright, but it still matters a little-- for instance, he might not have to go to his change-of-pace pitch as often (Wright also throws a fastball), which might benefit him. As for the leverage thing, that's a silly argument. MLB hitters are trying to do their best at all times, even in a blowout, aren't they? Heck, doing just that once won ARod an MVP that Ortiz completely deserved instead. You can make a better argument that hitters in a low leverage situation would be more relaxed about facing the dreaded knuckleball, which can make the best hitters look silly, and that therefore pitching in low leverage hurt him. I don't buy that either, but if you had a gun to your head and had to pick a barely credible argument about leverage, you'd pick that one and not yours. Hitters absolutely perform differently in lower-leverage situations, and it's not because they stop trying. Despite the fact that the pitchers who pitch in low-leverage situations are almost uniformly worse than those who pitch in high-leverage situations, the walk rate in low leverage situations is substantially lower (7.3% in LL this year, 9.2% in HL), and the strikeout rate is slightly lower as well (20.7% LL, 21.2% HL). That's because in blowouts, hitters swing more often and are less likely to work the count. This obviously benefits a knuckleballer a great deal, as one of their intrinsic flaws is that the unpredictable break of the pitch means they can struggle to find the strike zone. My broader point is just this: we cannot take Steven Wright's career-to-date major league stats at face value. We can quibble about the precise adjustment to make, but in the end, it doesn't really matter because the sample is almost uselessly small. It is a risk to assume that Wright has a substantial chance of equaling the performance of a Latos or a Leake, and the 2015 roster is a rather high-risk one already. Plus, based on their usage of Wright, the current front office almost certainly leans more in my direction than in yours. If they thought he could be a good second/third starter, he almost certainly would have been getting starts over, say, Ranaudo.
|
|
|
Post by mgoetze on Oct 12, 2014 10:51:11 GMT -5
And of course not every 3 and 4 caliber hitter hits there. This split is just a proxy for a more refined one by quality of opposition. Indeed, and your regular use of these proxy stats always leaves me scratching my head why noone has yet come up with a respectable opponent-adjusted stat (for pitchers or for hitters) and published it somewhere on the internet. I mean even total trash stats like OPS+ get published all over the place...
|
|
|
Post by FenwayFanatic on Oct 12, 2014 11:42:09 GMT -5
Trading Mookie or Xander doesn't make a lot of sense to me. Call me crazy but if we're going to overpay on the free agent market why not overpay for the best pitcher, max scherzer?
|
|
jimoh
Veteran
Posts: 3,966
|
Post by jimoh on Oct 12, 2014 14:44:44 GMT -5
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,923
|
Post by ericmvan on Oct 13, 2014 4:16:57 GMT -5
Hitters would absolutely benefit from seeing Wright a second and third time, actual results in a tiny sample notwithstanding. The sheer unconventional nature of the pitch (it has abnormally slow speed and unusual movement) as well as its scarcity (two pitchers in the league throw it) means that if anything, knuckleballers might have a steeper times-through-the-order effect than most pitchers, especially in relief appearances when hitters aren't expecting it and might not have read their scouting reports. The points you make are credible ones about hitting mistake knuckleballs: a knuckleball pitcher may be likelier to get away with a mistake the first time around the order than subsequently. That may factor into Wakefield's time around the order splits. But as I've pointed out before, when the knuckler is thrown correctly, there is actually no hitting it at all except by luck. Really and truly. So there's no times-through-order effect for well-thrown knuckleballs. The better the knuckleballer, the smaller you'd expect his splits to be. Phil Niekro's career splits: 656, 675, 697, 700. And the faced-4th number is a huge sample (56% the size of the faced-third). Charlie Hough career: 686, 695, 702, 687. Wilbur Wood career: 688, 665, 664, 711. Knuckleballers were more common in his day, which would reduce your hypothesized familiarity split, which I'm conceding may apply to mistakes. I think these numbers rather back up my take on things. If you can keep your feel for the knuckler all game long, there's almost no times-through-order split at all. Interestingly, Dickey had a big split in 2012 and a totally flat split in 2013. So his CY season came from being pretty much perfect with an outrageously good knuckler at the start of games, and then usually losing the feel for it as the game progressed. Whereas in his off year he pitched at nearly his previous 3rd-time-through level the whole time. In retrospect, the 2012 splits were a warning sign for season-to-season inconsistency; if he'd had a 630, 640, 650 split in 2012 instead of his 509, 651, 783, you would have been much more confident about his repeating the season. But he had demonstrated that he didn't always have the feel for his pitch. My memory of reading Wright's minor league game logs is that he did not have any times-through-order split (if I have time this winter I'll compile the numbers). If that's true, it may well indicate a more consistent career. (Hey, congrats, you've prompted me into being even higher on him!) Those differences seem to me to be much more likelier driven by changes in the pitching approach. In very low leverage with the team ahead, pitchers famously just throw the ball over the plate and let the other team hit it. I don't see a reason why hitters would be inherently less patient in low-leverage situations; they're reacting to the pitchers being less careful. Of course it isn't. It's very small, but it's terrific. You can look at other pitchers and see how their frequency of throwing 30 consecutive innings that good correlated to their overall quality; if you had the data and the statistical chops, you could get an excellent estimate for 90% and 95% confidence intervals for his metrics. And since we know he was throwing an excellent knuckleball, and since we know there's almost no times-through-order split for guys who do that (both in theory and in practice), and that he had in fact a sizable reverse split, then comparisons to starters would be absolutely fair. Now, all I've done with this approach is browse Latos's game logs, but it's clear that stretches of 30 innings that good are rare for him. That gives you some confidence that Wright is good and perhaps very good. And, because of the nature of the pitch, I would argue that you have more confidence in that assessment than you would for a conventional pitcher with the same SSS excellence. There are just so fewer variables involved in his performance. This is sort of a paradox: the knuckleball is itself inconsistent, but if you get a consistent excellent stretch, you are more confident in ascribing that excellence to the inherent performance as opposed to random and extraneous factors. Because the latter have been reduced. I think that was true until he went out there and demonstrated something that I've been arguing, that they may not have been aware of analytically: that the MLE multipliers for knuckleball pitchers are close to 1.0. Ranaudo and Webster had lower ERAs, higher K rates, and better or (for Webster) the same FIP, and if you don't realize that the AAA numbers will translate differently, with Wright losing much less, he's obviously the third guy you give a shot to. I don't think they were planning to start him until he pitched so well in relief, and after the start there were some media suggestions that he's now being taken much more seriously. I'm guessing that he and Webster are the two current options for fifth starter, should they acquire just one pitcher. I'm actually perfectly fine with the idea of acquiring a second guy -- Kazmir might be my choice -- if the price is right, because I very much agree with you that it reduces risk. We may disagree on what we'd be willing to pay to reduce risk, however. (BTW, I just adore this sort of debate.)
|
|
|
Post by jrffam05 on Oct 13, 2014 12:16:51 GMT -5
I'll be surprised if Lester doesn't wind up on the Cubs. Fit makes too much sense. I, too, would be stunned if Lester doesn't wind up with the Cubs. The Cubs have a great young offense coming along. They need an innings guy and Lester doesn't cost them a draft pick or young talent, just a lot of money, and with all those young kids coming along there's plenty of room in the payroll for Theo to splurge. Where the Sox are hesitant, Theo won't be. Lester will wind up getting a huge contract with the Cubs. I was thinking Lester was a fit for the Cubs even before he was traded. Lester is my pick for #1 pitching target, I think he is the best combination of value, cost, and risk. Plus I would think it would be pretty cool to trade him and resign him, along with the goodwill it would build up in the FO. It get's me thinking though, what is the real market for Lester, and Shields + Scherzer for that matter. Just eyeball the standings, I would think these teams would have some level of involvement in the bidding. Yankees Blue Jays Red Sox Tigers Angels Rangers Mariners Cardinals Cubs Dodgers Giants Nationals Mets Marlins Phillies Of those teams I'm not sure there will be a ton realistically in on the top 2 free agents Lester + Scherzer. I think the Yankees and Dodgers will stay away from those megadeals in the free agent market this offseason. Angels, Nationals, Rangers, and Blue Jays don't have a ton of salary coming off the books. Phillies really shouldn't be in the bidding, but you can't rule them out. Marlins, Mets, and Cubs are three teams with losing records with the payroll space and need for a pitcher, I think they can be in it. Giants, Seattle, and Cardinals all could have the space, but might be better off spending the money elsewhere. Tigers I think will pick between Scherzer and Price this offseason. That's really how I see the market in a quick breakdown. Thoughts?
|
|
|
Post by jimed14 on Oct 13, 2014 15:09:15 GMT -5
Lester is a good fit for the Red Sox too.
|
|
|
Post by jmei on Oct 13, 2014 19:21:30 GMT -5
So there's no times-through-order effect for well-thrown knuckleballs. The better the knuckleballer, the smaller you'd expect his splits to be. Phil Niekro's career splits: 656, 675, 697, 700. And the faced-4th number is a huge sample (56% the size of the faced-third). Charlie Hough career: 686, 695, 702, 687. Wilbur Wood career: 688, 665, 664, 711. Knuckleballers were more common in his day, which would reduce your hypothesized familiarity split, which I'm conceding may apply to mistakes. I think these numbers rather back up my take on things. If you can keep your feel for the knuckler all game long, there's almost no times-through-order split at all. Interestingly, Dickey had a big split in 2012 and a totally flat split in 2013. This data seems to be the opposite of your conclusion. Four of the five knuckleballers that you've looked at have a pretty clear times through the order penalty. I decided to do a quick slightly more systemic look. Here are the major knuckleballers in recent history (min. 100 games) and their career tOPS+ splits, then their collective average compared to the league times-through-the-order split in 2014: Dickey: 96 87 106 131 Sparks: 97 104 103 102 Springer: 89 106 124 62 Wakefield: 92 102 112 99 Candiotti: 98 101 98 104 Hough: 100 102 104 100 J. Niekro: 93 104 107 103 P. Niekro: 94 99 106 107 Bouton: 99 91 102 130 Wood: 105 98 98 112 Average, knuckleballers: 96.3 99.4 106 105 Average, 2014 league: 94 102 113 100 ...so you can argue that knuckleballers have a smaller times-through-the-order penalty (likely because they don't lose as much velo as most pitchers), but it stretches credulity to argue that they lack one altogether. Those differences seem to me to be much more likelier driven by changes in the pitching approach. In very low leverage with the team ahead, pitchers famously just throw the ball over the plate and let the other team hit it. I don't see a reason why hitters would be inherently less patient in low-leverage situations; they're reacting to the pitchers being less careful. That's a fair point-- the effect is definitely because of changes in both hitters' and pitchers' approaches-- but there are plenty of reasons why hitters would be less patient in low-leverage situations. With the game all but decided, hitters are more likely to elevate personal success over team success, and even in this SABR-friendly environment, extra-base hits are sexier than working the count for walks. There's also the fact that guys just want to go home after a long night and so are disincentivized to extend the game by working counts and drawing walks. And, because of the nature of the pitch, I would argue that you have more confidence in that assessment than you would for a conventional pitcher with the same SSS excellence. There are just so fewer variables involved in his performance. This is sort of a paradox: the knuckleball is itself inconsistent, but if you get a consistent excellent stretch, you are more confident in ascribing that excellence to the inherent performance as opposed to random and extraneous factors. Because the latter have been reduced. [...] I think that was true until he went out there and demonstrated something that I've been arguing, that they may not have been aware of analytically: that the MLE multipliers for knuckleball pitchers are close to 1.0. This is basically just ascribing quasi-magical qualities to the knuckleball, right? Absent more systemic analysis actually proving these propositions, I'm inclined to reject them. I mean, I can come up with folksy, unproven rationales that make the opposite argument. Maybe the knuckleball's inherently random break means knuckleballer performances are more non-predictive than most pitchers, or the MLE is much higher because minor league clubs don't have scouting reports preparing hitters for how to combat the knuckleball.
|
|
|
Post by larrycook on Oct 13, 2014 22:44:22 GMT -5
Could we interest the mets in some spare outfielders for one of their middle rotation starters like a niese?
|
|
|
Post by xanderbogaerts2 on Oct 14, 2014 4:41:05 GMT -5
Could we interest the mets in some spare outfielders for one of their middle rotation starters like a niese? I don't know the world on advanced sabermetrics on pitching yet, but his FIP, RA9, are league average. He's a prototypical ground ball pitcher so he tends to give up more hit than innings pitched. Besides his one good year he averages 1.6 WAR in a pitchers park. Cespedes has been linked to the Mets but they aren't getting Cesp for Niese at least I hope not.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,923
|
Post by ericmvan on Oct 14, 2014 6:39:46 GMT -5
So there's no times-through-order effect for well-thrown knuckleballs. The better the knuckleballer, the smaller you'd expect his splits to be. I decided to do a quick slightly more systemic look. Here are the major knuckleballers in recent history (min. 100 games) and their career tOPS+ splits, then their collective average compared to the league times-through-the-order split in 2014: Dickey: 96 87 106 131 Sparks: 97 104 103 102 Springer: 89 106 124 62 Wakefield: 92 102 112 99 Candiotti: 98 101 98 104 Hough: 100 102 104 100 J. Niekro: 93 104 107 103 P. Niekro: 94 99 106 107 Bouton: 99 91 102 130 Wood: 105 98 98 112 Average, knuckleballers: 96.3 99.4 106 105 Average, 2014 league: 94 102 113 100 ...so you can argue that knuckleballers have a smaller times-through-the-order penalty (likely because they don't lose as much velo as most pitchers), but it stretches credulity to argue that they lack one altogether. You'll note that I didn't in fact argue that; I was saying that well-thrown knuckleballs have no split, and mistake knuckleballs probably have one. That suggests that the better knuckleballers should have smaller splits. Here's 3rd time around less the smaller of 1st or second time around, and career ERA-: Wood: 0, 88 Candiotti: 0, 92 Hough: 4, 95 Sparks: 6, 106 P. Niekro: 12, 86 J. Niekro: 14, 102 Dickey: 19, 96 Wakefield: 20, 96 Springer: 35, 113 (Bouton really only threw the knuckler as a reliever.) The correlation of size of split and ERA- is .63, which just misses statistical significance at p =.07. Even better: you can get a great prediction of career ERA- from the times-through-order split, which we take as largely driven by mistake percentage, and BB rate, which is the other thing that would effect quality. Here's actual and predicted career ERA-: Wood: 88, 94 Candiotti: 92, 89 Hough: 95, 92 Sparks: 106, 100 P. Niekro: 86, 95 J. Niekro: 102, 102 Dickey: 96, 93 Wakefield: 96, 96 Springer: 113, 113 R = .83, p = .03. Another way of looking at this: as expected, there is a strong correlation between career ERA- and BB% (r = .65, p = .057). The residuals of that correlation, which is to say the part of ERA- not explained by BB%, have an even better (fractionally) correlation to the times around the order split, r = .66, p = .053. The more I think about this, the more sense it makes. Mistake knuckleballs probably have a pretty big times-through-order split because the pitch is indeed so unfamiliar. There may be a relatively small variation in non-mistake knuckleball effectiveness (P. Niekro the best, Sparks the worst), which may well have to do with average number of rotations and/or ability to change speeds; and/or that last residual may be ability to mix in other pitches effectively. But the data is certainly consistent with the idea that a well-thrown knuckleball has no times-through-order split because you can't learn to hit it by practice, period. Hitting the knckleball is entirely about not missing the miss-thrown ones. That will indeed be likelier to happen as the game progresses. That you tossed that out indicates that you're not getting the essential argument. The pitch literally breaks, and truly breaks, in a more or less random direction, when thrown correctly. If you had a machine that threw perfect knuckleballs and used it to try to teach guys how to hit the knuckleball, you'd accomplish nothing. They could practice hours a day and not get better; it would be like practicing to predict radioactive decay. And, because of the nature of the pitch, I would argue that you have more confidence in that assessment than you would for a conventional pitcher with the same SSS excellence. There are just so fewer variables involved in his performance. This is sort of a paradox: the knuckleball is itself inconsistent, but if you get a consistent excellent stretch, you are more confident in ascribing that excellence to the inherent performance as opposed to random and extraneous factors. Because the latter have been reduced. [...] I think that was true until he went out there and demonstrated something that I've been arguing, that they may not have been aware of analytically: that the MLE multipliers for knuckleball pitchers are close to 1.0. This is basically just ascribing quasi-magical qualities to the knuckleball, right? Absent more systemic analysis actually proving these propositions, I'm inclined to reject them. I mean, I can come up with folksy, unproven rationales that make the opposite argument. Maybe the knuckleball's inherently random break means knuckleballer performances are more non-predictive than most pitchers, or the MLE is much higher because minor league clubs don't have scouting reports preparing hitters for how to combat the knuckleball. No, there's no woo-woo or handwaving here. It all follows from the science of the pitch. The MLE ratio for a great knuckleball pitcher is going to be close to 1.0, and it will get higher as he gets less good, because the MLE ratios of a properly thrown knuckleball are 1.0 going from Little League all the way up to robotic superhuman hitters, because the pitch breaks at random and cannot be hit except by luck. The MLE ratio for mistake knuckleballs is properly larger than for conventional pitching, because a knuckleball that doesn't break is a cripple among cripples. The other point follows as well, although it may be less dramatic than I was making it sound. Most crucially, thirty excellent innings of conventional pitching is often the result of having all three or four pitches working right, which we can take as a possibly rare confluence; but a knuckleball pitcher who has done that simply had his one pitch working right. There's little possibility that the pitcher was helped by facing weak opponents (since even poor MLB hitters can hit mistake knuckleballs, and good ones don't hit good knuckleballs any better), too. I might look at start-to-start variation for the best knuckleball pitchers versus comparable conventional ones. That might be eye-opening.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,923
|
Post by ericmvan on Oct 14, 2014 7:09:54 GMT -5
Steven Wright is 30 years old As far as Wright goes, how many knuckleballers in the modern era have been the best P on their staff year after year? Definitely Wilbur Wood with the White Sox (WAR numbers from 1971 to 1974: 11.7, 10.7, 7.5 and 5.6 with IP of 334, 376, 359 and 320!!) and P. Niekro with the Braves for a long stretch. Wright's upside is to be another Wake: an innings eater who pitches to a little better than a league average year after year. Nothing wrong with that. Among good career knuckleballers, Wood and Niekro were, on average, #2 starters, Candiotti was a solid #3 and Hough, Dickey, and Wakefield were low-end #3 starters. I think it's too early to rule out that Wright is in the former group; anyone who has shown an excellent knuckler that he can throw for sustained stretches almost by definition has Niekro upside. Of course, Wakefield-like is the more likely positive outcome. And of course his age is doubly unimportant. He's essentially in his fourth year of pro pitching, and there's no reason why he can't pitch another dozen years if he's good, which makes him the same virtual age as the guys he's competing with for a 25-man spot. In fact, he's probably "younger" than Webster. Oh, and of the two pitchers we're talking about, the one that is tougher to get a read on and project is not Wright, but Kelly.
|
|
|
Post by Chris Hatfield on Oct 14, 2014 7:19:34 GMT -5
anyone who has shown an excellent knuckler that he can throw for sustained stretches almost by definition has Niekro upside. Of course, Wakefield-like is the more likely positive outcome. Eric, why isn't this comment just another way of saying "knuckleballers who were good pitchers were good pitchers"?
|
|
|
Post by jmei on Oct 14, 2014 8:27:35 GMT -5
The mistake knuckleball versus non-mistake knuckleball is a distinction without a difference. Whatever you want to chalk it up to, it's clear that knuckleballers as a whole perform worse as the game goes on. Remember, we don't know Wright's mistake/non-mistake ratio, and I hope you'll concede that not even the best knuckleballers throw a perfect knuckleball every time. As such, comparing Wright (who almost never faced hitters a third time) to starters is a flawed comparison.
The velo comment is meant to suggest that part of the times through the order penalty is because other pitchers' stuff declines throughout the game, while knuckleballers' don't really.
The MLE question is an empirical one, and your thesis appears to be wrong at first glance. Many knuckleballers with great minor league stats have busted in the majors-- see the Wright thread in the Meta forum. Remember, we're not talking about the perfect knuckleball here (a perfect fastball might be as unhittable as well). We're talking about the one thrown by Wright, which even he'll admit is imperfect much of the time. We can't assume he'll be Niekro and not Haegar.
|
|
|
Post by Chris Hatfield on Oct 14, 2014 9:42:45 GMT -5
I moved a bunch of posts discussing Kelly, Wright, etc. from the Cespedes/Lester trade thread into this one.
|
|
|
Post by moonstone2 on Oct 14, 2014 22:09:57 GMT -5
This whole discussion about Wright is pretty much moot if he doesn't pitch. I don't think anyone in their right mind is planning on Wright being the #1 or #2 they need, but realistically they have three slots open and probably can't afford to go and get three veteran starters. From Alex Spier's chat today we learn that at the very least, Wright is a realistic option for one of those spots and maybe more realistic than Webster, DelaRosa, Workman or Ranaudo.
Seems to me that Eric is going to get his wish and we can either laugh at him or he can declare himself a genius (again), in ten months time.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,923
|
Post by ericmvan on Oct 31, 2014 10:22:24 GMT -5
We all agree that the Sox need to, and will, acquire at least one front-line pitcher (a #2 starter, because there are likely no true aces available) .
There is widespread agreement that signing two frontline pitchers to long deals is not a good idea, because of all the depth we have.
There is disagreement about whether we should try to get a second guy who is in his walk year. Personally, I'm comfortable either way: my Plan A is to trade Cespedes for Heyward, and the plan B is to trade him for a pitcher. In Plan A, we roll with what we've got plus one new guy.
But here's a new thought: why can't that sole new guy be a rental? Forget Lester, forget Hamels. The whole obsession with finding an "anchor" for the staff could be rendered moot in a year if, if Buchholz has the year he clearly is capable of, or if one of the kids makes another step forward.
Let's say they like the odds of a Latos comeback (you can run this hypothetical with your favorite equivalent, i.e., Cueto, Samardzija, Iwakuma, etc.). Could you win a WS by starting the season with Buchholz, Latos, Kelly, De La Rosa, and Wright, with the historically good collection of Owens, Rodriguez, Webster, Barnes, and Johnson at AAA, all battling to seize a spot (probably pushing Rubby to the pen)? Before you simply quote me and answer "No.", remind yourself what other post-season rotations looked like. Tillman, Chen, Norris, Gonzalez, and Jimenez were not exactly the 1995 Braves. After Bumgarner, the Giants had very little.
I know what you're thinking. Isn't it a huge gamble to count on Buchholz being a top-of-the-rotation starter? Well, yes. How is that less of a gamble than paying Lester a huge sum to be a top-of-the-rotation starter for five, six, or seven more years?
I think the Sox should be be focused on trying to trade for both Heyward (this winter) and Stanton (at the deadline), and that means not exhausting resources (prospects or dollars) extravagantly. (That's also why I like Nava for Valbuena as a 3B solution.)
Thoughts? I'm not convinced this is the way to go -- just that it's an option that should be on the table should, say, the Lester bidding get out of hand, and the asking price for Hamels remains too high (since they're still deluded they can contend).
Edit: in this scenario, if you're worried about depth and/or think Wright is useless, then you can sign someone like Masterson as a second acquisition.
|
|
|
Post by jmei on Nov 5, 2014 16:41:26 GMT -5
I moved the proposed Cespedes/Cueto trade to its own thread in the trade proposal subforum.
|
|
|
Post by larrycook on Nov 5, 2014 21:45:29 GMT -5
This whole discussion about Wright is pretty much moot if he doesn't pitch. I don't think anyone in their right mind is planning on Wright being the #1 or #2 they need, but realistically they have three slots open and probably can't afford to go and get three veteran starters. From Alex Spier's chat today we learn that at the very least, Wright is a realistic option for one of those spots and maybe more realistic than Webster, DelaRosa, Workman or Ranaudo. Seems to me that Eric is going to get his wish and we can either laugh at him or he can declare himself a genius (again), in ten months time. Which rookie pitcher was the most impressive in 2014?
|
|
|
Post by jimed14 on Nov 6, 2014 14:28:14 GMT -5
Another example at how off fWAR is for pitchers. Buchholz was worth 2.2 fWAR in 2014. 49th best in the majors. That's a solid #2.
|
|
|
Post by wcsoxfan on Nov 6, 2014 17:45:15 GMT -5
Another example at how off fWAR is for pitchers. Buchholz was worth 2.2 fWAR in 2014. 49th best in the majors. That's a solid #2. Well, if there are only 12 aces in a given year then what would be considered a '#2 pitcher' performance-wise should probably be limited to the top 42. Which would make him a #3. Then you have to account for superior pitchers who were injured. The number of qualified pitchers last year was '88' so this would place Buchholz into the 44th percentile of qualified pitchers, making him slightly below average. If you account for Buchholz having a very high BABIP and a very low LOB% (each the second worse of his career) then this label seems to fit as his 5.34 ERA seems a bit inflated (4.01 FIP /4.04 xFIP). He had a disappointing year, but the above seems fair to me. Expecting Buchholz to rebound next year to at least 'above average #3' status if not better.
|
|
|
Post by jimed14 on Nov 6, 2014 17:54:30 GMT -5
Another example at how off fWAR is for pitchers. Buchholz was worth 2.2 fWAR in 2014. 49th best in the majors. That's a solid #2. Well, if there are only 12 aces in a given year then what would be considered a '#2 pitcher' performance-wise should probably be limited to the top 42. Which would make him a #3. Then you have to account for superior pitchers who were injured. The number of qualified pitchers last year was '88' so this would place Buchholz into the 44th percentile of qualified pitchers, making him slightly below average. If you account for Buchholz having a very high BABIP and a very low LOB% (each the second worse of his career) then this label seems to fit as his 5.34 ERA seems a bit inflated (4.01 FIP /4.04 xFIP). He had a disappointing year, but the above seems fair to me. Expecting Buchholz to rebound next year to at least 'above average #3' status if not better. Why are you only using qualified pitchers? Buchholz was bad, and that's probably why his BABIP was high. He was getting nailed. John Lackey had 1.6 fWAR in 2011.
|
|
|
Post by James Dunne on Nov 6, 2014 18:43:48 GMT -5
So you know something weird? Clay Buchholz set a career high in strikeouts this season.
|
|
|