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.
Report: Red Sox have agreement for Wade Miley
danr
Veteran
Posts: 1,871
|
Post by danr on Dec 11, 2014 15:41:00 GMT -5
django, did you read the fangraphs piece? MAsterson lost quite a bit from his fastball, possibly from his injury, but whether he recovers won't be known until next season. Miley is a solid 200 IP guy, and the Sox have him for several more years.
BTW, Miley's style of pitching seems very similar to Brian Johnson's, which may mean that Johnson has been underrated.
|
|
|
Post by soxfan06 on Dec 11, 2014 15:41:51 GMT -5
Hardly. Masterson is a gamble. Miley is solid. Really? I feel like we bought high on Miley. Something about this deal feels like when we traded for Mark Melancon. I hope, as others have speculated, that we flip Miley for Hamels. Pitching in Fenway, with Xander at short, I can't see Miley's value going anywhere but down. Bought high? Really we gave up two fringe pitchers who are both likely bullpen arms. If that is buying high, boy we should be able to get Cole Hamels for Devin Marrero and Garin Cecchini.
|
|
|
Post by thursty on Dec 11, 2014 15:47:11 GMT -5
At first I was a little bit skeptical about this trade, with Rubby being the part I didn't like giving up. I don't think Webster will be anything more than a bullpen guy really. But I'm coming around on the deal and this piece on fangraphs is pretty interesting: www.fangraphs.com/blogs/wade-miley-who-is-better-than-you-think/If it's been posted here before, that's my bad lol. There is a table which compares Miley to Homer Bailey and Jeff Samardzija over the past 3 years, and you really can't tell the difference. Plus, Miley was pitching in Arizona which has got some short fences, and he still didn't give up that many homers. Interesting piece to be sure, but one telling item that seemed to be omitted is his spike in BB% in 2014 (from 7.8 to 8.7, 3.35/9), which went hand-in-hand with his spike in K% (which was focussed on) I still think he profiles as a #4, and simply isn't worth 3 prospects with any amount of upside. If the 3rd player is a cipher, then I guess it's OK - anything more than that and I don't like it. Miley is something of a salary dump for Arizona, they have no leverage
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,924
|
Post by ericmvan on Dec 11, 2014 15:57:22 GMT -5
In the trade forum I talked about how to get Cueto, and step one was getting a groundball pitcher who would be a good fit for that park with its crazy HR PAF. I talked about getting Ross or Cashner from SD. Well, Miley fits that bill to a T.
If you were the Reds, would you take 3 years of Miley and 6 of Marrero for 1 of Cueto? I would, especially since I've just traded away Latos and Simon as well. If not, what if the Sox added Ranaudo (whom the Reds could flip to a team with a big stadium for a comparable prospect with GB tendencies)? Or the Sox could trade Ranaudo and Coyle for a very decent young player and include him. And / or sign Badenhop and include Workman.
IOW, flipping Miley for Cueto is 100% doable. The Sox have enough excess talent to go Godfather on them (although of course you hope to get him for as little as possible).
Of course, all this works with Kelly as well.
Cueto Buchholz Porcello Kelly or Miley Masterson Wright [Workman]
Barnes Rodriguez Owens Johnson Escobar
That's a great rotation and great depth.
|
|
danr
Veteran
Posts: 1,871
|
Post by danr on Dec 11, 2014 16:01:52 GMT -5
That is a very sound approach and it might work. I suspect that the chances the Reds will trade Cueto have been reduced by their other deals. It's the same approach I suggested for Hamels. However, we do have a big disagreement. My first choice would be to keep Miley and flip Buchholz. I know you don't like that idea!
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,924
|
Post by ericmvan on Dec 11, 2014 17:11:20 GMT -5
That is a very sound approach and it might work. I suspect that the chances the Reds will trade Cueto have been reduced by their other deals. It's the same approach I suggested for Hamels. However, we do have a big disagreement. My first choice would be to keep Miley and flip Buchholz. I know you don't like that idea! All of their seasons, 10 or more GS, by bWAR / 30 starts: 8.1 Buchholz (2013) 6.0 Buchholz 4.1 Buchholz 3.8 Buchholz 3.6 Miley (2012) 1.2 Miley (2013) 0.9 Buchholz (2012) 0.7 Miley (2014) -1.7 Buchholz (2014) -2.4 Buchholz (rookie season)
So, defining peak as average of 3 best seasons (since that's all Miley has), you want to trade the 6.0 WAR talent and keep the 1.8.
Admittedly, Clay's been hurt a lot and Miley's a workhorse. So, straight WAR: Buchholz 5.6 Buchholz 4.3 Miley 3.5 Buchholz 2.0 Buchholz 1.9 Miley 1.3 Buchholz 0.9 Miley 0.8 Buchholz -1.2 Buchholz -1.6 Now you want to trade the 4.0 peak value guy and keep the 1.9. Which is to say, trade the guy whose 3 best seasons (in the span of the last five) are at All-Star level and keep the guy who has been average. Recency? Clay is just a year removed from the second best bWAR/GS season in all of MLB over the last five or ten years (I forget how far back I went). Note this pattern in Clay's last 5 seasons (all bWAR is per 30 starts): 6.0 bWAR, 28 GS 4.1 bWAR, 14 GS -> hurt 0.9 bWAR, 29 GS, strong finish 8.1 bWAR, 16 GS -> hurt -1.7 bWAR, 28 GS, strong finish He's had two bad years, each following an injury season and characterized by a terrible start and strong finish. The last time, he bounced back to compile 4.3 WAR in half a season. Now, you present your rational argument.
|
|
|
Post by jimed14 on Dec 11, 2014 17:17:46 GMT -5
Eric, perhaps look at the downside risk rather than the possible upside. Miley is as close to a sure thing as you can get. Buchholz is as likely to be hurt or phantom DLed so he can work on his mechanics as he is to be great. If Cherington is looking at the baseline worst likely outcome more than he is dreaming about the best possible outcome, he could prefer Miley over Buchholz because he doesn't want to roll the dice on a completely unreliable player with unbelievable upside anymore.
It probably depends a lot more on which player teams value more if they are trading one of them.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,924
|
Post by ericmvan on Dec 11, 2014 17:34:52 GMT -5
Eric, perhaps look at the downside risk rather than the possible upside. Miley is as close to a sure thing as you can get. Buchholz is as likely to be hurt or phantom DLed so he can work on his mechanics as he is to be great. If Cherington is looking at the baseline worst likely outcome more than he is dreaming about the best possible outcome, he could prefer Miley over Buchholz because he doesn't want to roll the dice on a completely unreliable player with unbelievable upside anymore. It probably depends a lot more on which player teams value more if they are trading one of them. They have the depth this year (which they absolutely did not have last year) to be aggressive if Clay is struggling. Clay's never been bad two years in a row. His struggles have not come at random; he gets hurt mid-year, and then has a bad start and strong finish the next year, then is great again (and gets hurt). Best season of each two-year pair starting in 2009: 5.6, 5.6, 1.9, 4.3, 4.3. Given how much time has passed since we was last physically unhealthy, and his strong finish, I have no hesitation counting on 2.0 to 4.5 WAR from him. Since his upside is beyond that, I don't see a scenario where the downside risk comes close to outweighing the upside.
|
|
|
Post by JackieWilsonsaid on Dec 11, 2014 17:59:22 GMT -5
The thing about Buck long term is the body type issue.
He is Porcello light er.
Instead of getting thicker as he ages and stronger he seems lighter than ever which is a part of his durability issue.
Coming out of Arizona, the seemed to be a strong buzz around Marrerro.
I won't be surprised if he and a lesser starter gets more than you think.
if Cinci is trying to build in place I would think buck and a lesser piece would yield Cueto. Btw I don't know if anyone caught the live presser announcing the Detroit Cincinnati deal, but it was really cool with both gms together and each came off really well with great mutual respect.
|
|
|
Post by umassgrad2005 on Dec 11, 2014 18:01:32 GMT -5
Eric, perhaps look at the downside risk rather than the possible upside. Miley is as close to a sure thing as you can get. Buchholz is as likely to be hurt or phantom DLed so he can work on his mechanics as he is to be great. If Cherington is looking at the baseline worst likely outcome more than he is dreaming about the best possible outcome, he could prefer Miley over Buchholz because he doesn't want to roll the dice on a completely unreliable player with unbelievable upside anymore. It probably depends a lot more on which player teams value more if they are trading one of them. They have the depth this year (which they absolutely did not have last year) to be aggressive if Clay is struggling. Clay's never been bad two years in a row. His struggles have not come at random; he gets hurt mid-year, and then has a bad start and strong finish the next year, then is great again (and gets hurt). Best season of each two-year pair starting in 2009: 5.6, 5.6, 1.9, 4.3, 4.3. Given how much time has passed since we was last physically unhealthy, and his strong finish, I have no hesitation counting on 2.0 to 4.5 WAR from him. Since his upside is beyond that, I don't see a scenario where the downside risk comes close to outweighing the upside. If patterns stay the same then look at this 2010 28 starts, 2011 14 starts, 2012 29 starts, 2013 16 starts, and 2014 28 starts. So in 2015 we should get 15 starts. I don't care about WAR if he's hurt at the end of the year and can't help us win a championship.
Using rWAR Buchholz WAR since 2010 5.4, 1.8, .8, 4.3 and -1.6. In his second best season based on WAR 2013 he only made 16 starts. So only one year was he healthy and pitching very well at the end of the year and that was 2010. We can't count on Buchholz for anything.
|
|
|
Post by jmei on Dec 11, 2014 18:32:50 GMT -5
Eric, perhaps look at the downside risk rather than the possible upside. Miley is as close to a sure thing as you can get. Buchholz is as likely to be hurt or phantom DLed so he can work on his mechanics as he is to be great. If Cherington is looking at the baseline worst likely outcome more than he is dreaming about the best possible outcome, he could prefer Miley over Buchholz because he doesn't want to roll the dice on a completely unreliable player with unbelievable upside anymore. It probably depends a lot more on which player teams value more if they are trading one of them. They have the depth this year (which they absolutely did not have last year) to be aggressive if Clay is struggling. Clay's never been bad two years in a row. His struggles have not come at random; he gets hurt mid-year, and then has a bad start and strong finish the next year, then is great again (and gets hurt). Best season of each two-year pair starting in 2009: 5.6, 5.6, 1.9, 4.3, 4.3. Given how much time has passed since we was last physically unhealthy, and his strong finish, I have no hesitation counting on 2.0 to 4.5 WAR from him. Since his upside is beyond that, I don't see a scenario where the downside risk comes close to outweighing the upside. Do they? Their best MLB-ready starters are now Workman (who is likely to start the season in the bullpen) and Wright, with Barnes maybe ready as well. That's not really that deep, and the team last year (with Workman, Webster, and RDLR) had at least comparable, if not greater, depth. Sure, Owens/Rodriguez/Johnson will be in the mix, but they had Barnes/Ranaudo in a similar position last year, and it's tough to count on guys with a combined couple months of AAA time to be averagish major league starters by July. I think you're dreaming a little much on upside here. The second part of your analysis is basically numerology, right? "Might as well give up in 2016, cause the Giants are going to win the WS again"-type stuff. Seeing patterns in randomness. The kind of stuff I'm constantly pushing back on. I'm happy to keep Buchholz because he does have that tantalizing upside, but I see no reason to believe that he's any more likely to be that guy in 2015 just because his risk profile has manifested itself in alternating good/bad seasons in the recent past.
|
|
|
Post by jmei on Dec 11, 2014 19:39:04 GMT -5
Something to note re: Miley: he has one of the worst IFFB% in baseball over the last few years, which is why his SIERA is consistently worse than his xFIP. So while he gets a lot of ground balls, he isn't getting many infield fly balls, which are pretty much as good as strikeouts.
|
|
|
Post by Sammy on Dec 11, 2014 20:34:14 GMT -5
Longtime reader finally getting around to posting. I've read this site since the Chad Spann and Jeff Natale days. I had Spann penciled in for several years manning third - his major league comp per this site was Joe Randa, after all, not too shabby. And there were a few new Aaron Sele and Jeff Suppan comps every year. It goes to show why, I assume, this site has abandoned those comps - because evaluating minor leaguers disproportionately by their ceilings is pointless. MLB comps for guys who statistically have a slim chance of carving out a MLB career only serve to overstate their value. Performance trumped best-case projection with all three pitchers involved in this deal. We are talking about 2 ex-Sox prospects with #3 upside who did not seize their opportunities. Maybe Miley has less best case projectability, but he made the most out of it so far. I think RDLR is being overrated by many people here who view him as a near lock to reach pretty close to his ceiling. He might, but there's a greater chance he won't.
The value of a prospect to a team, imo, is not in daydreams like some people are having about Rubby, it is in cost controlled years. Sox only have 1 fewer of Miley's than they would have of RDLR. And they have Miley for his peak age 28-30 seasons, at a time, 3 years into the league, when many starting pitchers finally put it all together. People are looking at this like just because Miley broke into MLB, his development has flatlined. Who's to say he doesn't still have untapped potential too? All because he doesn't have as many mph on the radar gun? Otoh, the DBacks will now have RDLR for his age 25-28 seasons, so maybe for the start of his peak *if* he harnesses his potential, but lucky him, he'll be back on the FA market right in the middle of it. So if AZ molds Rubby into a star, they will have the pleasure of watching him pitch some of his best baseball for someone else, though they will have a comp pick to show for it. Webster they have for age 25-30 seasons so if they can turn him around, more power to them.
Maybe pitcher durability is this year's version of moneyball OBP and the Sox have a new way to project or quantify it. It's long been almost aximoatic in baseball that if you can get around 1,000 innings from your opening day starting rotation, you will have a pretty successful season. Yeah, yeah, all pitchers are durable... until they're not, but the Sox trend this offseason has been big bodied-starting pitchers who have a history of durability, clean bill of health on their arms, don't require commitments past age 31, have low-stress deliveries, and are already established in the majors.
|
|
jimoh
Veteran
Posts: 3,972
|
Post by jimoh on Dec 11, 2014 20:45:27 GMT -5
The second part of your analysis is basically numerology, right? "Might as well give up in 2016, cause the Giants are going to win the WS again"-type stuff. Seeing patterns in randomness. The kind of stuff I'm constantly pushing back on…. Yeah, this line is hilarious: "Clay's never been bad two years in a row. His struggles have not come at random."
|
|
|
Post by redsox1534 on Dec 11, 2014 20:54:18 GMT -5
Not a bad firt post Sammy
I like Miley hes got the ability to strike guys out, and get grounders. Hes got a good delivery and arm actions when pitching. Hes big and durable. RDLR has a fastball and a change up and untill he improves the breakingball and command he isnt gonna live to the hype. Plus theres the problmen of worrying he can stay healthy. I no he can be a great RP if not elite back end RP one day. I love what his upside could be and think hes closer to it and its higher then some other SP we had in 13 like Webster and Ranuado and Workman. But hes a question mark and if he doesnt answer answer the call in 15 hes getting shipped to bullpen and I would be ok with that he could be the answer to our future in the back of the bullpen. Webster deff has big upside like RDLR but he is further away he has questionable control and not so good command. He is more up and down then any one we had last year. His stuff all around is top notch sure he can use some sharpening here and there on some a couple of his pitches but thats not his problem. Its the command, placement of his pitches. He too seems like he has to take a big step in 15 or be destined to the bullpen. And that isnt a bad thing either because man he could be nasty going to the bullpen. Imagine RBLR or AW adding a couple more miles to that fastball out of the pen wow. Were talking potentiel CL at some point in there careers. I am ok with losing both but id rather have kept one thrown him in the pen and traded Workman in his place. I doubt the minor leaguer is any one great but I think the hold up could be the player we are suppose to get from the Cubs is gonna go to Arizona. Thats just one idea I had any way.
|
|
|
Post by deepjohn on Dec 11, 2014 22:18:09 GMT -5
The second part of your analysis is basically numerology, right? "Might as well give up in 2016, cause the Giants are going to win the WS again"-type stuff. Seeing patterns in randomness. The kind of stuff I'm constantly pushing back on…. Yeah, this line is hilarious: "Clay's never been bad two years in a row. His struggles have not come at random." Eric was the first I saw to write about baseball patterns this way. I'm sorry, but to me it's nothing like numerology or the Giants. The reasoning goes something like this: Every player has his own reasons as to why he plays the way he does, but we just don't know those reasons, and in fact, the player typically may not know the reasons consciously, either. If you were to interview Clay and study what he has done differently from year to year I think you could reasonably expect to find there is a non-random pattern in his life that explains this apparently non-random pattern of being injured. Rather than push back, let's hope the team reads this forum and tries to correct whatever Clay may be doing so as not to be injured every other year.
|
|
|
Post by grandsalami on Dec 12, 2014 1:24:30 GMT -5
The Red Sox need an ace to headline their rebuilt rotation, Nick Cafardo of the Boston Globe opines. From that same piece, Cafardo talks to an NL scout who believes the Sox got the better of the Wade Miley trade. The scout calls the young left-hander an “unrefined [Jon] Lester right now, but he’s on his way to being a No. 2 [starter]. Not sure what the D’Backs are thinking on this one, except to get more bodies.”
MLBTR
|
|
|
Post by Oregon Norm on Dec 12, 2014 1:45:55 GMT -5
The Red Sox need an ace to headline their rebuilt rotation, Nick Cafardo of the Boston Globe opines. From that same piece, Cafardo talks to an NL scout who believes the Sox got the better of the Wade Miley trade. The scout calls the young left-hander an “unrefined [Jon] Lester right now, but he’s on his way to being a No. 2 [starter]. Not sure what the D’Backs are thinking on this one, except to get more bodies.” MLBTR One of the real problems I have with off-the-cuff analysis is that there's no real definition for "ace", and none for #1, #2,..., #5. Was Lester an ace? If so, why? Is it all based on hearsay, on intuition, feelings? I need something I can grab on to. The way these terms often are used is near meaningless to me. To get real about it, in 2012 Jon Lester himself was an "unrefined" Jon Lester.
|
|
nomar
Veteran
Posts: 10,790
|
Post by nomar on Dec 12, 2014 1:58:38 GMT -5
The Red Sox need an ace to headline their rebuilt rotation, Nick Cafardo of the Boston Globe opines. From that same piece, Cafardo talks to an NL scout who believes the Sox got the better of the Wade Miley trade. The scout calls the young left-hander an “unrefined [Jon] Lester right now, but he’s on his way to being a No. 2 [starter]. Not sure what the D’Backs are thinking on this one, except to get more bodies.” MLBTR One of the real problems I have with off-the-cuff analysis is that there's no real definition for "ace", and none for #1, #2,..., #5. Was Lester an ace? If so, why? Is it all based on hearsay, on intuition, feelings? I need something I can grab on to. The way these terms often are used is near meaningless to me. To get real about it, in 2012 Jon Lester himself was an "unrefined" Jon Lester. Agreed. I guess you could calculate what percentile a pitcher is in for a given category.
|
|
|
Post by mgoetze on Dec 12, 2014 2:14:38 GMT -5
One of the real problems I have with off-the-cuff analysis is that there's no real definition for "ace", and none for #1, #2,..., #5. Was Lester an ace? If so, why? Is it all based on hearsay, on intuition, feelings? I need something I can grab on to. The way these terms often are used is near meaningless to me. To get real about it, in 2012 Jon Lester himself was an "unrefined" Jon Lester. Obviously neither Cafardo nor "an NL scout" will buy in to this, but sticking to the semiridiculous definition that only half the teams have a #1 pitcher, I go to the FanGraphs leaderboards and derive: #1: 4+ WAR #2: 2.5-4.0 WAR #3: 1.0-2.5 WAR #4: Oh... wait... there aren't even 90 qualified pitchers, much less 105... Let's try that again with all Pitchers FG deems "Starters", 0 IP minimum #1: 4.1+ WAR #2: 2.8-4.1 WAR #3: 1.9-2.8 WAR #4: 1.3-1.9 WAR #5: 0.7-1.3 WAR By FanGraphs Projections that makes Porcello, Miley, Masterson and Buchholz #3s and Kelly a #5.
|
|
|
Post by charliezink16 on Dec 12, 2014 3:09:41 GMT -5
Oh god, not this discussion again.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,924
|
Post by ericmvan on Dec 12, 2014 5:42:07 GMT -5
The second part of your analysis is basically numerology, right? "Might as well give up in 2016, cause the Giants are going to win the WS again"-type stuff. Seeing patterns in randomness. The kind of stuff I'm constantly pushing back on. I'm happy to keep Buchholz because he does have that tantalizing upside, but I see no reason to believe that he's any more likely to be that guy in 2015 just because his risk profile has manifested itself in alternating good/bad seasons in the recent past. What you're saying is: Every time he finishes a season injured, he pitches poorly at the start of the next season, but that's not causative, since he had to spend the off-season recovering from the injury, and couldn't do his regular workout ... it's random! Every time he finishes a season strongly he starts the next season great, but that's not causative, because he was healthy in the off-season and could do his regular routine for working out and staying strong ... it's random! Every time he starts a season poorly, at some point he starts to pitch dramatically better, but that's not causative, because he's a hugely talented guy who just takes longer than a single off-season to recover physically from an injury and then re-assemble his mechanics ... it's random! Seriously? The one thing that seems to be random is getting hurt. BTW, it's not unusual for a pitcher to have a demonstrable alternating-year pattern derived on how much off-season work they did. Josh Beckett used to be consistent every year through pitch 25, and in alternate years he would either sustain that impressively through 100 pitches or drop off progressively and dramatically as the game went on. Gee, you don't think he worked harder after having a bad year than a good year? Nah, why use basic human nature to explain things in an obvious fashion, when you can just declare them random? It's a bit risky to make predictions based on an effort-based pattern, because at some point the guy might notice it and wise up. But Esteban "No One Will Be" Loaiza had an amazing alternate year pattern 2002-2005, and I dove into it and found the same sort of deeper patterns you'd expect if he was in shape one year and out of shape the next. I told the Sox to stay away, but Billy Beane bit, and got burned in '06 as I'd predicted. Now, you're not wrong on principle -- a lot of narrative does get constructed around random variation (the point of a classic xkcd cartoon). But sometimes either the narrative is so inherently likely that you can trust it, as with Clay, or it's reasonably likely and makes predictions about data you haven't looked at yet, which are consistent with the narrative to a degree that they essentially verify it. To correct a tale Peter Keating got a bit wrong in the ESPN story, after the 2003 season I concocted a narrative about Jason Varitek being an extreme early-bird based on his day-night splits (hugely better at day); it was backed up by finding other data patterns that were predicted by that hypothesis. I was mocked at SoSH. John Henry apparently read it and passed it on to 'Tek, who proceeded in 2004 to have the opposite split (which of course is much better, since there are so many more night games), engendering further mockery. Then Henry hired me, because I'd been right; as Theo put it "he is the first guy to go to bed on the road." Having been made aware of the splits, he'd found a way to deal with them (Theo and Jed felt that having had a kid had contributed motivation-wise). I think it was in '06 that I noticed he'd started the season being hugely better at day again, and I e-mailed and said "tell 'Tek to do whatever he does to play better at night." And the splits turned around dramatically beginning the next game (which suggests that the compensation was something fairly simple, like doing caffeine). Now, maybe that was random, but I suspect not. Telling the real from the random is an art form, IMHO. You need both a sense of the reasonableness of the narrative and the statistical unlikeliness of the pattern. You need to distinguish between patterns you found while searching for all possible patterns, and ones that jumped up and shouted at you. But the best tool is the question: if this narrative were true, what else would we expect to see? If you see it, you've probably got something real.
|
|
jimoh
Veteran
Posts: 3,972
|
Post by jimoh on Dec 12, 2014 5:42:41 GMT -5
Yeah, this line is hilarious: "Clay's never been bad two years in a row. His struggles have not come at random." Eric was the first I saw to write about baseball patterns this way. I'm sorry, but to me it's nothing like numerology or the Giants. The reasoning goes something like this: Every player has his own reasons as to why he plays the way he does, but we just don't know those reasons, and in fact, the player typically may not know the reasons consciously, either. If you were to interview Clay and study what he has done differently from year to year I think you could reasonably expect to find there is a non-random pattern in his life that explains this apparently non-random pattern of being injured. Rather than push back, let's hope the team reads this forum and tries to correct whatever Clay may be doing so as not to be injured every other year. If I flip a coin six times and it alternates heads and tails, do I have a magical power to make coin flips alternate?
|
|
jimoh
Veteran
Posts: 3,972
|
Post by jimoh on Dec 12, 2014 5:48:23 GMT -5
The second part of your analysis is basically numerology, right? "Might as well give up in 2016, cause the Giants are going to win the WS again"-type stuff. Seeing patterns in randomness. The kind of stuff I'm constantly pushing back on. I'm happy to keep Buchholz because he does have that tantalizing upside, but I see no reason to believe that he's any more likely to be that guy in 2015 just because his risk profile has manifested itself in alternating good/bad seasons in the recent past. What you're saying is: Every time he finishes a season injured, he pitches poorly at the start of the next season, but that's not causative, since he had to spend the off-season recovering from the injury, and couldn't do his regular workout ... it's random! Every time he finishes a season strongly he starts the next season great, but that's not causative, because he was healthy in the off-season and could do his regular routine for working out and staying strong ... it's random! Every time he starts a season poorly, at some point he starts to pitch dramatically better, but that's not causative, because he's a hugely talented guy who just takes longer than a single off-season to recover physically from an injury and then re-assemble his mechanics ... it's random! Seriously? The one thing that seems to be random is getting hurt. BTW, it's not unusual for a pitcher to have a demonstrable alternating-year pattern derived on how much off-season work they did. Josh Beckett used to be consistent every year through pitch 25, and in alternate years he would either sustain that impressively through 100 pitches or drop off progressively and dramatically as the game went on. Gee, you don't think he worked harder after having a bad year than a good year? Nah, why use basic human nature to explain things in an obvious fashion, when you can just declare them random? It's a bit risky to make predictions based on an effort-based pattern, because at some point the guy might notice it and wise up. But Esteban "No One Will Be" Loaiza had an amazing alternate year pattern 2002-2005, and I dove into it and found the same sort of deeper patterns you'd expect if he was in shape one year and out of shape the next. I told the Sox to stay away, but Billy Beane bit, and got burned in '06 as I'd predicted. Now, you're not wrong on principle -- a lot of narrative does get constructed around random variation (the point of a classic xkcd cartoon). But sometimes either the narrative is so inherently likely that you can trust it, as with Clay, or it's reasonably likely and makes predictions about data you haven't looked at yet, which are consistent with the narrative to a degree that they essentially verify it. To correct a tale Peter Keating got a bit wrong in the ESPN story, after the 2003 season I concocted a narrative about Jason Varitek being an extreme early-bird based on his day-night splits (hugely better at day); it was backed up by finding other data patterns that were predicted by that hypothesis. I was mocked at SoSH. John Henry apparently read it and passed it on to 'Tek, who proceeded in 2004 to have the opposite split (which of course is much better, since there are so many more night games), engendering further mockery. Then Henry hired me, because I'd been right; as Theo put it "he is the first guy to go to bed on the road." Having been made aware of the splits, he'd found a way to deal with them (Theo and Jed felt that having had a kid had contributed motivation-wise). I think it was in '06 that I noticed he'd started the season being hugely better at day again, and I e-mailed and said "tell 'Tek to do whatever he does to play better at night." And the splits turned around dramatically beginning the next game. Now, maybe that was random, but I suspect not. Telling the real from the random is an art form, IMHO. You need both a sense of the reasonableness of the narrative and the statistical unlikeliness of the pattern. You need to distinguish between patterns you found while searching for all possible patterns, and ones that jumped up and shouted at you. But the best tool is the question: if this narrative were true, what else would we expect to see? If you see it, you've probably got something real. The Varitek story is a good one, but the pattern seems based on a lot more data than with Buchholz. In the 70s I remember Fisk was told he had terrible stats late in games and he stopped sprinting to back up the 1b on ground balls and his late-game hitting improved b/c of lack of fatigue.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,924
|
Post by ericmvan on Dec 12, 2014 5:57:05 GMT -5
If I flip a coin six times and it alternates heads and tails, do I have a magical power to make coin flips alternate? And if you do that sixty times, I'm going to guess that your "flip" consists of picking up the coin as is, and giving it a half-rotation. News flash: the world contains causation as well as randomness.
|
|
|