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 iakovos11 on Jul 15, 2014 18:24:54 GMT -5
Yet this site projects him to be in Lowell next year Touche, good sir. FWIW, I disagree with the site's projection in that case. Only situation in which Devers isn't in Greenville, imo, is if they move Chavis to third and don't want them timesharing. Salem would be a big jump for Devers in that case.
|
|
|
Post by jmei on Jul 18, 2014 12:15:44 GMT -5
I moved some Devers ranked at #4 discussion to that thread in the Meta forum. It's a good discussion, and one worth having, just not on the main boards.
|
|
|
Post by amfox1 on Aug 1, 2014 16:55:46 GMT -5
From today's BA chat:
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,882
|
Post by ericmvan on Aug 23, 2014 4:41:07 GMT -5
Since Devers was bumped to #4 (on July 18), he's hit .253 / .314 / .379 in 105 PA. He kept on raking for a week, and then has hit .203 / .267 / .348 over his last 75 PA.
I argued (in various game threads, I think) that some of Devers' great numbers were the result of his having hot zones where most LHH have cold ones, with the result that he absolutely destroyed pitchers when he was playing against a team for the first time, and perhaps for the second time as well. Folks made sarcastic arguments about the lack of advance scouting in the DSL and GCL ... but, obviously, teammates watch the games and if they're not stupid, they notice things that obvious.
Here are the updated splits by game # versus a given team. HOC, Hardness of Contact, is a weighted average of the last three columns (the weights being their contribution to run scoring).
Game PA BA OBP SA EqA K% BB% HOC HRC BABIP XB% 1 54 .444 .519 .800 .434 .093 .130 .432 .073 .447 .294 2 54 .356 .444 .556 .345 .185 .148 .360 .028 .429 .267 3-4 65 .304 .409 .446 .304 .169 .123 .312 .022 .364 .313 5-8 51 .295 .373 .409 .279 .235 .098 .305 .000 .394 .308 9-14 50 .191 .240 .340 .202 .200 .060 .221 .054 .200 .143
Now, a lot of that has to be an artifact of his going into the aforementioned slump. All but one of the games in his 1-4 splits happened before the slump started, and all of the 9-14 games are in the slump. Here's his games 5-8 breakdown, before and after the slump:
5-8 PA BA OBP SA EqA K% BB% HOC HRC BABIP XB% Hot 30 .346 .400 .423 .295 .267 .100 .352 .000 .474 .222 Cold 21 .222 .333 .389 .256 .190 .095 .249 .000 .286 .500 So let's just do these splits for before the slump started.
Game PA BA OBP SA EqA K% BB% HOC HRC BABIP XB% 1 54 .444 .519 .800 .434 .093 .130 .432 .073 .447 .294 2 54 .356 .444 .556 .345 .185 .148 .360 .028 .429 .267 3-4 61 .308 .419 .462 .312 .148 .131 .311 .023 .357 .333 5-8 30 .346 .400 .423 .295 .267 .100 .352 .000 .474 .222 Now the last line is a smaller sample, so it's likely that the apparent stylistic change (much more aggressive at the plate, with harder contact) is in part a SSS fluke. But the difference between his game 1 line and his games 3-4 or 3-8 line line is so large as to suggest that his true peak talent level, the sustained performance that got us all so excited, is best represented by something like his 3-8 line plus his game 2 line with half of the difference removed, which is .328 / .420 / .473.
|
|
|
Post by burythehammer on Aug 23, 2014 5:58:54 GMT -5
I argued (in various game threads, I think) that some of Devers' great numbers were the result of his having hot zones where most LHH have cold ones, with the result that he absolutely destroyed pitchers when he was playing against a team for the first time, and perhaps for the second time as well. Folks made sarcastic arguments about the lack of advance scouting in the DSL and GCL ... but, obviously, teammates watch the games and if they're not stupid, they notice things that obvious. The problem really isn't the scouting thing. I personally doubt that players at that level are able to or are even worried about making such adjustments when they're literally just learning how to play pro ball, but that's just an opinion. Fine. This is going to come off as combative and I don't mean it to be, I respect you, but the problem is that your initial premise, talking about a guy's "hot zones" based on literally nothing more than a slash line, is ludicrous. And it would be ludicrous at any level, MLB included. It's like saying you can tell me the color of someone's eyes based on their Social Security Number. And you can analyze it down to the most granular level you want, the information we have in a box score just cannot provide that kind of analysis. Maybe he did so well in the first match-ups because "typical" hot zones are even hotter for him! Who knows? I'll tell you who: people that have actually watched his at-bats with their eyes. Or seen Pitch F/X or Hit F/X data on them (if it even exists). I don't understand why it can't be enough to just say that he's a 17 year old having a great season in rookie ball, for whatever that's worth, and then take into account the scouting reports, whatever they may be. The sample size and relevance of rookie ball stats is just not enough to draw any grander conclusions than that.
|
|
|
Post by joshv02 on Aug 23, 2014 6:36:26 GMT -5
Since Devers was bumped to #4 (on July 18), he's hit .253 / .314 / .379 in 105 PA. He kept on raking for a week, and then has hit .203 / .267 / .348 over his last 75 Are the chances that is random 1:2500?
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,882
|
Post by ericmvan on Aug 23, 2014 8:40:10 GMT -5
I argued (in various game threads, I think) that some of Devers' great numbers were the result of his having hot zones where most LHH have cold ones, with the result that he absolutely destroyed pitchers when he was playing against a team for the first time, and perhaps for the second time as well. Folks made sarcastic arguments about the lack of advance scouting in the DSL and GCL ... but, obviously, teammates watch the games and if they're not stupid, they notice things that obvious. The problem really isn't the scouting thing. I personally doubt that players at that level are able to or are even worried about making such adjustments when they're literally just learning how to play pro ball, but that's just an opinion. Fine. This is going to come off as combative and I don't mean it to be, I respect you, but the problem is that your initial premise, talking about a guy's "hot zones" based on literally nothing more than a slash line, is ludicrous. And it would be ludicrous at any level, MLB included. It's like saying you can tell me the color of someone's eyes based on their Social Security Number. And you can analyze it down to the most granular level you want, the information we have in a box score just cannot provide that kind of analysis. Maybe he did so well in the first match-ups because "typical" hot zones are even hotter for him! Who knows? I'll tell you who: people that have actually watched his at-bats with their eyes. Or seen Pitch F/X or Hit F/X data on them (if it even exists). I don't understand why it can't be enough to just say that he's a 17 year old having a great season in rookie ball, for whatever that's worth, and then take into account the scouting reports, whatever they may be. The sample size and relevance of rookie ball stats is just not enough to draw any grander conclusions than that. Not combative at all -- a reasonable objection. Let me see if I can explain. First, there's no premise here. I'm always looking at box scores and numbers. Sometimes there's a pattern in the box scores that I notice that's so strange that I then look at the numbers. So there's no data fishing. This started when I noticed that Devers had gotten off to preposterously great, laugh-out-loud starts in his first handful of games in both the DSL and GCL, and then cooled down to mere excellence. Next, I'm asking, is there any simple explanation as to what would cause that? A look at his game log showed that the damage was being done in the first game against each team. My sense (which I'll test in a moment statistically) was that what he did in game 1 against a new team was unlikely to be random. Well, there is in fact a simple explanation. Almost ever LHH has trouble hitting the ball low and inside. I think that 17 year-old pitchers know that. You absolutely try that spot with a fastball at some point in your first game against a LHH who is hitting 3 or 4 in the opposing lineup. Well, what if, in fact, Devers loves that pitch, and that's one of the reasons he's so good? His swing is so nice that he doesn't have that hole.
I think you have to agree that it's perfectly reasonable that Devers doesn't have that hole, as almost all young LHH do, and that's one reason he's a stud.
And I think you have to agree that it's perfectly reasonable that even 17 year-old pitchers in the GCL know that most LHH have trouble there, and like to attack them there. And I think you have to agree that it's perfectly reasonable that the kid pitching the next game (who may well be charting the pitches in this game; I don't know why they wouldn't start doing that in the DSL) sees that, and goes, holy crap, that was a perfectly located FB and the kid hit it out, and since I hope to make it to MLB and be a multi-millionaire, I certainly won't try that pitch tomorrow. (In fact, in this little story, some of the kids say, well, he won't hit my FB out, and try it, and also fail, so by the 3rd game the starter knows not to even go there.) Oh, and when you look at the much lower K rate and much higher HR rate in game 1, versus games 2 through 8 before the slump, the odds against it being random are about 70 to 1. Note that K rate and HR rate are usually positively correlated, so that really stood out for me (and would make the odds longer if factored in). This doesn't make him less of a prospect; it simply helps explain why, for a while, he had insanely great numbers for a 17 y/o, rather than the merely great ones we dreamed about. And, yes, there's so much noise in rookie ball stats that they are really just reinforcers of scouting reports. My motivation here is simply that I enjoy figuring out weird patterns in data that seem, to me, to be unlikely to be random. And I think the conclusion that Devers doesn't have the obvious hole in his swing that most LHH do is actually useful.
|
|
|
Post by ramireja on Aug 23, 2014 10:15:51 GMT -5
I'm always looking at box scores and numbers. Sometimes there's a pattern in the box scores that I notice that's so strange that I then look at the numbers. So there's no data fishing. This started when I noticed that Devers had gotten off to preposterously great, laugh-out-loud starts in his first handful of games in both the DSL and GCL, and then cooled down to mere excellence. My motivation here is simply that I enjoy figuring out weird patterns in data that seem, to me, to be unlikely to be random. I think this is the entire problem that some of us have with your data analyses though. Any statistician would absolutely consider this to be data fishing. You're not generating hypotheses first and then looking at the data to test your hypotheses. You're doing quite the opposite. You're looking at the data first, locating the statistical anomalies, and then applying hypotheses to explain the data. That is indeed data fishing. If you take any normally distributed dataset with a series of numbers in truly random order, subsets of that data will not 'appear' random to the human eye. There are a number of social psychology experiments supporting the notion that humans have a really difficult time distinguishing truly random data from data manipulated to appear random. The classic example goes something like: If I flip a coin 6 times in a row, which is the more likely outcome (H = heads, T = tails): HHHHHH or HTTHTH? Now I know I'm not going to pull a fast one on you, of course they are equally likely, but the larger point applies to what you're doing. If I flipped a coin 500 times in a row, you might find a chain of say 5 or 6 heads in a row. If my coin flipping is completely unbiased, then we know that occurrence is random. You can't just identify it after its happen and then seek a cause for its occurrence. Now all that said, baseball is far more complex and less random than flipping a coin. Of course there are a number of factors that could actually affect the data. So its not that I think my example is completely analogous to what you do, but rather I do think there are fair critiques to your means of post-hoc analyses without any type of correction for Type I error.
|
|
|
Post by burythehammer on Aug 23, 2014 10:21:00 GMT -5
So, why are these pitchers aware enough to make an adjustment after a game or two but not an at-bat or two? Surely they wouldn't want to give up multiple rockets in the same game, they want to make the big leauges! And is every single pitcher he faced going to have the same strategy and the same ensuing adjustments? Like the poster above me said, you're bending the narrative to fit the numbers.
We can't prove that a single one of his hits in the games you're talking about came off a pitch down and in, so no, I don't think it's reasonable to think that many or most of them did. Possible? Sure. It's also possible that they came off of poorly located pitches from inexperienced high school kids or Latin guys.
|
|
|
Post by rjp313jr on Aug 24, 2014 9:51:48 GMT -5
A players performance is not random. There are factors at play that are very much either very much in control of a batter (stance, swing, grip on the bat etc) or not as much in his control (mood, focus, weather, lighting, energy levels, health, etc) that all have an effect on the outcome. It makes sense to notice patterns and try to find out why. And just because it's hard for most people to do something, in this case noticing patterns, doesn't mean an extraordinarily talented individual can't. If "they" did a study on the likelihood that people,could do what our top athletes or top minds can accomplish and we accepted that as how everyone was then no advancements would be made. The ability of the human brain is incredible, unfortunately very little of its power is used. Some people have abilities (gifts) that most do not. Eric bothers me as much as the next guy with his arrogance (sorry buddy couldn't keep it all roses , but he sees and notices things the rest of us don't a lot of the time. Beyond that, learn to read what he's saying. He didn't call this fact. He just shared his thought process and pointed out something that could MIGHT be true and gave his evidence that bears WATCHING. He does this ALL the time. He's very really calling anything out as fact, because he understands that'd be foolish. Once you stop reading it that way and look at it objectively you might get more out of it. Also, I'm not saying not to question or challenge or think it's not likely to be true.
|
|
|
Post by ramireja on Aug 24, 2014 11:07:25 GMT -5
A players performance is not random. There are factors at play that are very much either very much in control of a batter (stance, swing, grip on the bat etc) or not as much in his control (mood, focus, weather, lighting, energy levels, health, etc) that all have an effect on the outcome. It makes sense to notice patterns and try to find out why. And just because it's hard for most people to do something, in this case noticing patterns, doesn't mean an extraordinarily talented individual can't. If "they" did a study on the likelihood that people,could do what our top athletes or top minds can accomplish and we accepted that as how everyone was then no advancements would be made. The ability of the human brain is incredible, unfortunately very little of its power is used. Some people have abilities (gifts) that most do not. Eric bothers me as much as the next guy with his arrogance (sorry buddy couldn't keep it all roses , but he sees and notices things the rest of us don't a lot of the time. Beyond that, learn to read what he's saying. He didn't call this fact. He just shared his thought process and pointed out something that could MIGHT be true and gave his evidence that bears WATCHING. He does this ALL the time. He's very really calling anything out as fact, because he understands that'd be foolish. Once you stop reading it that way and look at it objectively you might get more out of it. Also, I'm not saying not to question or challenge or think it's not likely to be true. I realize I'm mostly responsible for shifting this conversation to the realms of a statistics throw down, I apologize.....but just a few counterpoints because I'm not sure I'm making myself clear. First, yes you're right, a player's performance is not random. However, that is exactly what is being tested when Eric is running these analyses. Specifically when Eric is testing two samples side by side and stating the odds that the two samples would occur in such proximity by chance. Eric undoubtedly is a gifted baseball fan/thinker/stats guy. I'm not trying to take away from that and I appreciate and encourage his analysis. My gripe is with his use of presenting p-values and the post-hoc nature of his analyses without accounting for Type 1 error inflation. Its purely a statistical critique. Theoretically, I enjoy his hypotheses and they make for good discussion. Regarding one's extraordinary ability to notice patterns....ummm....he's simply comparing hot streaks to cold streaks. He does this all the time. You can do that too. My point was moreso that when humans do see statistical anomalies, the natural bias is NOT to treat them as random and thus try to fit hypotheses to them. He is doing just that. All in all, keep up the good work Eric, I'm not trying to discount TOO much. Just be careful with the whole "the odds of this being due to chance are .000003"
|
|
|
Post by amfox1 on Sept 2, 2014 9:32:50 GMT -5
|
|
|
Post by okin15 on Sept 2, 2014 14:42:37 GMT -5
The part about the acceptable defense and the plus arm excites me. Not that he's going to be playing RF in Fenway, but he could be an asset in LF if he has to move from 3B... or they could convert him to pitcher, har har har.
|
|
jimoh
Veteran
Posts: 3,948
|
Post by jimoh on Sept 2, 2014 15:13:02 GMT -5
Can I ask what it means to "get rotational"?
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,882
|
Post by ericmvan on Sept 4, 2014 1:31:53 GMT -5
A players performance is not random. There are factors at play that are very much either very much in control of a batter (stance, swing, grip on the bat etc) or not as much in his control (mood, focus, weather, lighting, energy levels, health, etc) that all have an effect on the outcome. It makes sense to notice patterns and try to find out why. And just because it's hard for most people to do something, in this case noticing patterns, doesn't mean an extraordinarily talented individual can't. If "they" did a study on the likelihood that people,could do what our top athletes or top minds can accomplish and we accepted that as how everyone was then no advancements would be made. The ability of the human brain is incredible, unfortunately very little of its power is used. Some people have abilities (gifts) that most do not. Eric bothers me as much as the next guy with his arrogance (sorry buddy couldn't keep it all roses , but he sees and notices things the rest of us don't a lot of the time. Beyond that, learn to read what he's saying. He didn't call this fact. He just shared his thought process and pointed out something that could MIGHT be true and gave his evidence that bears WATCHING. He does this ALL the time. He's very really calling anything out as fact, because he understands that'd be foolish. Once you stop reading it that way and look at it objectively you might get more out of it. Also, I'm not saying not to question or challenge or think it's not likely to be true. I realize I'm mostly responsible for shifting this conversation to the realms of a statistics throw down, I apologize.....but just a few counterpoints because I'm not sure I'm making myself clear. First, yes you're right, a player's performance is not random. However, that is exactly what is being tested when Eric is running these analyses. Specifically when Eric is testing two samples side by side and stating the odds that the two samples would occur in such proximity by chance. Eric undoubtedly is a gifted baseball fan/thinker/stats guy. I'm not trying to take away from that and I appreciate and encourage his analysis. My gripe is with his use of presenting p-values and the post-hoc nature of his analyses without accounting for Type 1 error inflation. Its purely a statistical critique. Theoretically, I enjoy his hypotheses and they make for good discussion. Regarding one's extraordinary ability to notice patterns....ummm....he's simply comparing hot streaks to cold streaks. He does this all the time. You can do that too. My point was moreso that when humans do see statistical anomalies, the natural bias is NOT to treat them as random and thus try to fit hypotheses to them. He is doing just that. All in all, keep up the good work Eric, I'm not trying to discount TOO much. Just be careful with the whole "the odds of this being due to chance are .000003" I actually don't have ANY problem with this critique; I don't even regard it as such, and regard it instead as providing the important context which I habitually omit because my brain is always including it as a default. When I quote chi-square or t-test odds, they are simply to inform one and all of the size of the statistical anomaly, because that is very helpful to know: given that this may well be a fluke, is it a 1 in 100 fluke, or a 1 in 10,000 fluke? But that doesn't negate the truths you point out: statistical anomalies happen all the time, and we are really good at noticing them. So, yes indeed: the p values I cite and [however much Starbucks actually charges these days] will get you a cup of coffee. I'm really good at noticing statistical anomalies. Probably too good for my own good. I'm even better at inventing credible narratives to explain them. Certainly too good for my own good. The point of the combination, however, is very rarely to argue for any given reality, but almost always to simply point at a possibility worth considering. When I do believe there is a reality that folks are missing, BTW, I try to make that clear.
|
|
|
Post by soxfanatic on Sept 4, 2014 4:09:03 GMT -5
|
|
|
Post by moonstone2 on Sept 4, 2014 4:29:13 GMT -5
I think one thing needs to be clear. A statistical anomoly isn't necessarilly meaningful. It's not unusual for a player to show greatly improved performance over a period of time.. These periods of performance are NOT flukes because the player is.actually. playing better during this period. But they may not be meaningful as sustaining the improved performance is a whole different matter. Performing at your best or even at a much higher level than you normally do is extremely difficult to sustain.and usually is associated with an obvious explanation. Most players go back.to playing as they normally do. In a prospect's case, often even sustained improvement is not enough to change their major league outlook.
Many players can play like Mike Trout for a couple of weeks. But Mike Trout is Mike Trout all the time.
|
|
jimoh
Veteran
Posts: 3,948
|
Post by jimoh on Sept 4, 2014 10:53:25 GMT -5
Thanks. But that site makes "rotational" sound like the only popular and effective mlb swing today. But you're suggesting that "get rotational" in that analysis = "get [too] rotational"?
|
|
|
Post by soxfanatic on Sept 4, 2014 11:11:18 GMT -5
Thanks. But that site makes "rotational" sound like the only popular and effective mlb swing today. But you're suggesting that "get rotational" in that analysis = "get [too] rotational"? Yeah I'm saying that. But I could be wrong.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,882
|
Post by ericmvan on Sept 4, 2014 11:22:22 GMT -5
I think one thing needs to be clear. A statistical anomoly isn't necessarlly meaningful. It's not unusual for a player to show greatly improved performance over a period of time.. These periods of performance are NOT flukes because the player is.actually. playing better during this period. But they may not be meaningful as sustaining the improved performance is a whole different matter. Performing at your best or even at a much higher level than you normally do is extremely difficult to sustain.and usually is associated with an obvious explanation. Most players go back.to playing as they normally do. In a prospect's case, often even sustained improvement is not enough to change their major league outlook. Many players can play like Mike Trout for a couple of weeks. But Mike Trout is Mike Trout all the time. Excellent point. But it is has an important corollary: guys who have shown the ability to play at an elevated level for a decent stretch have upside that guys who have never done so lack. Given two players with identical, so-so, stat lines, the guy who had a ten-week torrid spell that was offset by a horrendous slump is a much more interesting prospect than the guy who was consistently mediocre all season. (He's also likely to be the guy with the better scouting report.) This is to be distinguished from a really well-established consistent level of play, and than an anomalous month. That can happen when a guy, for instance, changes his stance radically, turning cold zones into hot ones. Folks might remember Kevin Millar opening his stance hugely and going crazy for a month, which is how long it took the opposition to get a new book on him. Once they adapted, he went back to struggling, and then he went back to his normal stance. Carlos Pena was a guy who had long, great, hot streaks and long, terrible slumps; when he finally broke through in TB, the only difference was that the slumps were shorter. And shortening your slumps is a skill that can be learned - in fact, I believe Pena early in is career was prolonging his slumps, once they started, by changing his approach.
|
|
|
Post by ramireja on Sept 4, 2014 11:50:04 GMT -5
I realize I'm mostly responsible for shifting this conversation to the realms of a statistics throw down, I apologize.....but just a few counterpoints because I'm not sure I'm making myself clear. First, yes you're right, a player's performance is not random. However, that is exactly what is being tested when Eric is running these analyses. Specifically when Eric is testing two samples side by side and stating the odds that the two samples would occur in such proximity by chance. Eric undoubtedly is a gifted baseball fan/thinker/stats guy. I'm not trying to take away from that and I appreciate and encourage his analysis. My gripe is with his use of presenting p-values and the post-hoc nature of his analyses without accounting for Type 1 error inflation. Its purely a statistical critique. Theoretically, I enjoy his hypotheses and they make for good discussion. Regarding one's extraordinary ability to notice patterns....ummm....he's simply comparing hot streaks to cold streaks. He does this all the time. You can do that too. My point was moreso that when humans do see statistical anomalies, the natural bias is NOT to treat them as random and thus try to fit hypotheses to them. He is doing just that. All in all, keep up the good work Eric, I'm not trying to discount TOO much. Just be careful with the whole "the odds of this being due to chance are .000003" I actually don't have ANY problem with this critique; I don't even regard it as such, and regard it instead as providing the important context which I habitually omit because my brain is always including it as a default. When I quote chi-square or t-test odds, they are simply to inform one and all of the size of the statistical anomaly, because that is very helpful to know: given that this may well be a fluke, is it a 1 in 100 fluke, or a 1 in 10,000 fluke? But that doesn't negate the truths you point out: statistical anomalies happen all the time, and we are really good at noticing them. So, yes indeed: the p values I cite and [however much Starbucks actually charges these days] will get you a cup of coffee. I'm really good at noticing statistical anomalies. Probably too good for my own good. I'm even better at inventing credible narratives to explain them. Certainly too good for my own good. The point of the combination, however, is very rarely to argue for any given reality, but almost always to simply point at a possibility worth considering. When I do believe there is a reality that folks are missing, BTW, I try to make that clear. Cheers. Also, I tend to warn people not to read too much into p-values as they're not only influenced by your actual effect size magnitude, but also the size of your sample. That said, I'm an experimental psychologist, so my sample sizes generally refer the number study participants. In your case, sample sizes often to a number of at bats or innings pitched. In that sense, it seems more meaningful but its always nice to accompany a p-value with some measure of effect size (e.g., cohen's d).
|
|
|
Post by pedroelgrande on Sept 5, 2014 13:54:18 GMT -5
Nick (MA) I haven't heard much about Rafael Devers. What are the reasons for his being rated so highly? Klaw 2:31 PM ET Huge raw power, great swing.
|
|
|
Post by brianthetaoist on Sept 5, 2014 15:06:21 GMT -5
I think one thing needs to be clear. A statistical anomoly isn't necessarilly meaningful. It's not unusual for a player to show greatly improved performance over a period of time.. These periods of performance are NOT flukes because the player is.actually. playing better during this period. But they may not be meaningful as sustaining the improved performance is a whole different matter. Performing at your best or even at a much higher level than you normally do is extremely difficult to sustain.and usually is associated with an obvious explanation. Most players go back.to playing as they normally do. In a prospect's case, often even sustained improvement is not enough to change their major league outlook. Many players can play like Mike Trout for a couple of weeks. But Mike Trout is Mike Trout all the time. The issue, though, is defining a player's "normal" level of play. Obviously, this is much more true of prospects. Clearly, the bias is toward overreacting to hot streaks; people react to more recent information and impressions more strongly than older information. So an attitude of overall skepticism is a useful corrective. BUT, guys change and develop, and occasionally quite dramatically. Mookie's a great (extreme, really) example of a guy who recently completely exploded his previous expectations. But for every Mookie, having a hot stretch and falling back to earth is so unremarkable that I can't think of a specific example right now. No wait! Chih Hsien-Chang ... that's a good example. While I'm skeptical of Eric's conclusions fairly often, I think it's really interesting to try to identify what's an anomalous stretch and what's establishing a new baseline (which is, in itself, a pretty amorphous concept). edit to add: it's probably unfair to Eric to call them "conclusions" ... but you get my point.
|
|
|
Post by redsox4242 on Sept 25, 2014 13:21:16 GMT -5
Kiley McDaniel @kileymcd Won't be the last time you hear this (just in instructs for now): Rafael Devers hits a line drive single thru a Rays infield shift. Retweeted by James M. Dunne
|
|
|
Post by zil on Sept 26, 2014 2:03:50 GMT -5
It's weird that they're shifting on him. Reports were saying he sprays the ball to all fields.
|
|
|