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.
Hitting Prospects, Ranked by Actual Performance
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 7, 2014 10:11:57 GMT -5
Intro if you're semi-new to "Davenport Peak Translations"Clay Davenport was one of the co-founders, and the original stats guru, of Baseball Prospectus. He's the inventor of EqA (Equivalent Average), which BP later renamed True Average (TAv), one of the better run metrics around, and one scaled to make it intuitively useful (.260 is defined as MLB average). Clay has probably devoted more time to trying to measure and adjust for the difference in play between minor league levels than anyone alive, and he is the only person I know of who has tried to develop a system that takes a minor league hitter's line and projects his MLB peak from it, factoring in age and level played at, while being as smart as possible about the components. Clay's system automates what we try to do intuitively, e.g., looking at a guy like Will Middlebrooks and saying, he doesn't project all that well with that crappy plate discipline. As you can see if you look at that player card, WMB projected as just an average MLB hitter his first three years in the minors, despite putting up numbers that were much better, and he has returned to that guy (at best) after two outlier years. I've found these Peak Translations to be a tremendously useful tool for evaluating ml hitters. All the questions about a player's hitting style, whether he had BABIP luck, whether he was too old for his league -- they're all answered and boiled down into a single number. Intro if you're skeptical about minor league defensive rankingsClay is also the author of BP's original defensive metrics, based on PBP data. It's basically his version of Sean Smith's Total Zone, which is good enough for b-ref to list alongside the Fielding Bible's Defensive Runs Saved, and be used as the basis for their defensive WAR for the years before there was DRS data. All these PBP metrcs work the same way. You know which fielder handled every ball in play, so you can look at all the plays a 3B made, for instance, and all the ground ball hits handled by the left fielder, and thus start with the percentage of balls hit to the 3B that he fielded successfully. You then adjust for a bunch of known biases, and, presto, you've got a defensive rating in runs saved. The resulting numbers have the same caveats as any defensive numbers, but they are certainly good enough to toss into the grinder in the course of producing an overall player ranking. The one is the report below clearly pass the smell test. Intro for this data.
I'll just explain the columns. Rnk. The rank of the players on this list. I've included but haven't ranked the three rookie graduates. The list omits players who played this year in the GCL, because Clay hasn't yet reported those peak translations, and it omits mlfa's who are out of options (DeJesus, Brown, etc.) and players 27 and older. I'll do a separate report with those guys and with recent MLB grads who have played this year in the minors (Weeks, Herrera, Middlebrooks, Lavarnway) later this week. SP. The Sox Prospect ranking among the players on this list, for direct comparison. Only 20 players are ranked in the top 60. Pos. The best position the player appears to be able to play competently at present, not necessarily where he has been playing mostly. Stefan Welch, for instance, is still apparently a pretty good defensive 3B and only played mostly at 1B because we had prospects to play 3B. Corner outfielders are not differentiated between (hence "OF"). Ped. Short for "pedigree," it's the best (lowest number) from among BA ranking, actual draft slot, or equivalent draft slot based on bonus and that year's slot recommendations. It is missing for players acquired from other organizations (I put a "--" in Holt's box as a reminder, but after him you're on your own). Ros. Roster status. FA for this year's mlfa's, and for most others, the year they are Rule 5 eligible. M is guys in MLB and O are guys on option, and the number following is their options left after this year. 2012, 2013, 2014. The last three Davenport Peak Translations in EqA / TAv (combining split seasons by weighting by PA), adjusted for position based on actual MLB hitting by position, the last 10 years (weighted 10 to 1). That's +.014 SS, +.013 C, +.006 2B, .000 3B / CF, -.011 OF, -.017 1B. ePA. Effective PA. This is your sample size measure; it's 2014 PA, plus 2/3 2013 PA, plus 1/3 2012 PA. Why the weights? Because of ... Net. Weights the last three seasons 3-2-1 (and by PA, of course). Contrary to prior threat, there is not yet a "Swihart Exemption" for the guys who have improved continuously, but you can take the difference between 2014 and Net and add it to the final ranking if you believe that 2012 through 2014 represent a true trend (improving more, each year, than is needed simply to play at the next level) rather than random variation. I could have regressed Net to the mean based on ePA, and probably would have done so it were obvious what to regress it to. That may be included in a later revision, one that also tries to alter the weights based on trend. Def. A heavily regressed defensive ranking at the specified position, in runs per 150 games. Geekage for this is so complex that I'll put it after the data. Vazquez includes his 2014 pitch framing per 120 games. DAdj. Was in the original report, but removed for space since it's actually redundant. It translates the Def ranking into points of EqA / TAv. A defensive run is worth a a bit more than a point of EqA (1.24 points, in fact). If you disagree with a defensive ranking, it's easy to produce a revised adjustment (e.g., at least .007 more for Bradley). Just add or subtract as many points to the final ranking as you think "Def" is off by, and you're close enough. LvL. The highest level they've played at, hence their closeness to MLB. 0 is DSL,1 is GCL, 2 is Lowell, 3 Greenville, 4 Salem, 5 Portland, 6 Pawtucket, 7 Boston (or, of course other teams at those levels). Split seasons are weighted by PA. Tot. Adds Net, DAdj, and an adjustment for level (distance from MLB) to get a total ranking. The level adjustment is simply because a guy who projects to be great and is already in AAA or MLB (e.g., Betts) is universally regarded as a significantly better prospect than a guy with a similar projection based on the GCL (e.g., Chavis). I actually derived this adjustment by seeing what would need to be done to the guys in last year's report (data I never posted) to make their rankings correlate best with SP's ranking. It takes the number of levels below MLB to the 1.75th power to get the number of points to deduct, hence AAA -.001, AA -.003, A+ -.007, A -.011, A- -.017, R - .023. The rankings are in the next post, split into two chunks for ease.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 7, 2014 10:20:14 GMT -5
And here they are.
Bill James once observed that a good metric conforms to perceived reality, but with a small number of interesting exceptions. No exceptions, and it tells you nothing new; too many, and it doesn't pass the smell test. I think this set of rankings fits that prescription really well.
Scroll the list leftward and you should be able to see everything except this intro, with the Tot outside the box. If that doesn't work for everyone, I'll have to shorten the player names ... (I already had to eliminate the mostly redundant column ckowing the size of the defensive adjustment.)
Rnk SP Name Pos Ped Ros 2012 2013 2014 ePA Net Def Lvl Tot 1 1 Betts M 2B 50 M3 .257 .311 .324 1044 .313 4 5.8 .317 2 3 Vazquez C M2 .259 .278 .251 808 .261 36 6.3 .305 3 2 Swihart C 5 '14 .254 .275 .286 858 .278 14 5.2 .293 Bogaerts SS 86 M3 .314 .311 .267 1062 .291 -3 7.0 .287 4 5 Cecchini 3B 22 O2 .279 .305 .261 1007 .280 3 6.0 .283 5 16 Asuaje 2B 208 '16 .254 .293 695 .285 1 3.3 .278 6 7 Coyle 2B 23 '14 .255 .280 .296 695 .283 -3 5.0 .277 7 4 Margot CF 42 '15 .264 .273 613 .271 12 3.1 .275 8 6 Marrero SS 14 '15 .277 .252 .259 911 .259 10 5.4 .269 9 9 Rijo 2B 86 '16 .256 .279 618 .274 4 3.0 .267 10 Blair C 142 FA .242 .271 .289 524 .280 -6 4.2 .266 Holt SS -- M1 .293 .234 .281 1079 .271 -4 6.8 .266 11 12 Hassan OF O1 .258 .287 .253 759 .261 3 6.0 .263 Bradley CF 30 M2 .304 .274 .214 968 .252 9 6.8 .262 12 20 Gragnani 2B '16 .241 .277 584 .268 -3 4.0 .257 13 14 Brentz OF 28 O2 .258 .253 .261 674 .257 -1 6.0 .255 14 11 Shaw 1B '14 .285 .237 .256 538 .253 2 5.6 .254 15 17 Mars CF 194 '17 .258 235 .258 5 2.2 .249 16 Miller SS '15 .246 .288 .245 451 .249 3 4.3 .247 17 15 Butler C M1 .253 .282 .221 675 .246 0 6.0 .245 18 8 Travis 1B 56 '17 .257 289 .257 1 2.4 .244 19 Suarez C '15 .268 .253 227 .258 1 2.0 .243 Denney C 25 R .261 63 .261 4 1.0 .243 20 Gibson SS 68 FA .233 .256 .267 357 .246 0 5.2 .243 21 Betts J 3B '17 .252 270 .252 5 2.0 .241 22 Welch 3B FA .260 .249 .221 840 .238 5 5.0 .240 23 13 Ramos CF 173 '14 .243 .252 .267 719 .254 -9 5.0 .239 24 18 Dubon SS '17 .219 .262 313 .257 -2 2.0 .237 25 Lin SS 17 '16 .250 .248 .234 684 .239 8 3.0 .237 26 Dent SS 46 FA .231 .224 .236 702 .232 3 5.2 .234 27 Flores 2B 35 '15 .221 .249 343 .240 6 2.0 .230 28 Rivero 3B M0 .267 .205 .227 890 .227 3 6.0 .230
Rnk SP Name Pos Ped Ros 2012 2013 2014 ePA Net Def Lvl Tot 29 Hudson CF '17 .253 .237 329 .242 2 2.0 .228 30 Witte 1B '16 .133 .248 644 .234 3 3.5 .228 31 19 De La Cruz OF '14 .267 .224 .242 836 .239 -7 5.0 .227 32 Roof 3B FA .197 .225 .241 755 .231 1 4.5 .227 33 Tavarez CF '15 .251 .229 .242 798 .237 -4 4.0 .226 34 Lopez 2B '15 .243 .230 .252 363 .242 0 2.0 .225 35 Procyshen C '17 .241 95 .241 -2 2.3 .224 36 Meneses SS FA .238 .238 .196 821 .217 7 5.2 .223 37 Allday CF '16 .253 .225 343 .234 -2 3.0 .222 38 Romanski C '16 .224 .224 440 .224 7 3.1 .222 39 Sopilka C '14 .241 .243 .216 210 .231 5 2.0 .221 40 Peralta 3B '14 .230 .226 284 .228 4 2.6 .219 41 10 Longhi OF 117 '17 .159 .274 154 .249 -11 2.0 .218 42 Wilkerson CF '14 .226 .221 .210 1041 .216 3 5.2 .218 43 Martinez 3B FA .235 .222 .210 948 .218 4 4.0 .217 44 Hissey OF 35 FA .219 .231 .208 485 .219 0 5.0 .216 45 Johnson M CF '14 .203 .196 .234 908 .222 -1 4.3 .215 46 Perkins OF 58 '14 .250 .217 .241 485 .234 -8 3.3 .214 47 Sturgeon CF '17 .214 236 .214 9 2.9 .213 48 Heller OF '15 .233 .198 .232 550 .225 -5 4.1 .212 49 Chester 1B '14 .232 .233 .197 737 .216 -1 4.7 .211 50 Rider SS '15 .190 .227 .231 545 .227 -5 3.0 .210 51 Vinicio SS 9 '14 .249 .213 .223 512 .222 -5 4.0 .209 52 Kapstein OF '14 .211 .227 .224 267 .223 -4 3.0 .206 53 Weems C 73 '15 .243 .229 .245 555 .240 -21 3.3 .204 54 Guerrero SS '14 .231 .193 .219 374 .213 -3 4.0 .202 55 Roberson C '14 .200 .236 .208 656 .215 0 2.4 .200 56 Moore 1B '15 .237 .205 .231 438 .220 -5 2.3 .199 57 Guzman OF '16 .208 214 .208 3 2.7 .198 58 Meyers OF '16 .269 .193 .202 327 .207 1 2.0 .191 59 Hernandez C '14 .168 .195 .213 259 .199 -2 4.3 .191 60 Coste C '15 .188 .228 .217 123 .217 -15 3.0 .188 61 Brenley C FA .208 .071 .158 272 .173 10 5.0 .181 62 Mager OF '15 .153 .218 .166 660 .183 0 3.2 .173 63 Escobar C FA .203 .174 .170 138 .174 3 4.0 .171 64 Bethea C '16 .202 .170 213 .188 -10 3.0 .164 65 Tellez 1B '17 .183 138 .183 2 1.0 .162 66 Gedman 1B '14 .176 .214 .113 419 .162 -1 4.0 .154
|
|
|
Post by klostrophobic on Sept 7, 2014 10:53:29 GMT -5
That was a long nine minutes waiting for the second post. Great stuff as always. It'd be interesting to see how these numbers stack up against other seemingly elite farm systems like the Cubs.
|
|
danr
Veteran
Posts: 1,871
|
Post by danr on Sept 7, 2014 11:15:29 GMT -5
I copied the numbers as unicode text into Excel and they filled out the spreadsheet. There are some missing cells, and they have to be entered to get the columns lined up. Once done, it is a very interesting document.
If you addressed it, I missed it, but isn't there an issue with projecting players who only have one or two years of minor league experience? How can you rank JBetts, STravis or DMars on the same scale with Bogaerts, Vazquez, etc?
|
|
|
Post by klostrophobic on Sept 7, 2014 12:04:31 GMT -5
RE: Vazquez' defense rating involving his pitch-framing numbers. How would he look relative to Swihart if we removed the pitch framing numbers? Just trying to grasp how excited to get about swihart if he is remotely similar in terms of additional value from pitch-framing.
Looking at these numbers, why isn't everyone in the general Boston area running around with a mookie in their pants?
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 7, 2014 12:17:29 GMT -5
I copied the numbers as unicode text into Excel and they filled out the spreadsheet. There are some missing cells, and they have to be entered to get the columns lined up. Once done, it is a very interesting document. If you addressed it, I missed it, but isn't there an issue with projecting players who only have one or two years of minor league experience? How can you rank JBetts, STravis or DMars on the same scale with Bogaerts, Vazquez, etc? Well, I do have an adjustment for distance from the majors that does most of that work. The rest of that would be an adjustment for small sample size, and as I mentioned, that's something that I may cook up, for next year. (It involves copy and pasting maybe 10 full years of Clay's data, integrating it with data from FG or BP, and then looking at it in as much detail as possible.) BTW, I'm perfectly willing to share the original spreadsheet (it's an .xlsx file) if folks can point me to a favorite way of doing so. The "Pedigree" data is drawn from a really great spreadsheet I've been maintaining for years, that I really ought to just send to Mike and see if he wants to put in the Wiki or something. I'm going to make some comments on these results, probably this evening, explaining some of the unexpected results, etc. And I'll start a thread in the meta-forum for discussing the implications of these results for prospect rankings. Because I am fairly certain that in cases where there is a big disparity between the two, the SP rankings are at least a little off. BTW, I think Clay's pitcher ratings are much less helpful, but still quite interesting, and I plan to do a similar thing with those. I've got the 2014 numbers already are just need to grab the previous two years and get an average.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 7, 2014 12:26:42 GMT -5
RE: Vazquez' defense rating involving his pitch-framing numbers. How would he look relative to Swihart if we removed the pitch framing numbers? Just trying to grasp how excited to get about swihart if he is remotely similar in terms of additional value from pitch-framing. Looking at these numbers, why isn't everyone in the general Boston area running around with a mookie in their pants? That's precisely the sort of question I was going to address in the comments that were going to wait until after the Pats and Sox games get watched! Good to know that at least one person wants to know ASAP. Vazquez was a regressed 15 runs per 150 without pitch-framing, one more than Swihart. Which is, BTW, just about exactly what he's been doing in MLB in terms of controlling the running game and picking guys off (13.5, last I looked). So he's getting a regressed 21 runs per 150 games for framing, when he's actually been framing 33 per 120 -- so that's a conservative defensive ranking for him. Really, from what we now understand about pitch-framing and what we have seen so far, he is looking like not just a GG winner but one of the historically great defensive players. Imagine if he ever gets good at blocking pitches! Oh, I'll explain the defensive metrics this evening, too. Re Mookie: having declared Xander to be best Red Sox prospect since Tony C, it kind of blows my mind that Mookie's 2014 season at the plate has been even better, according to this methodology, than either of Xander's last two years. It's better, in fact, even if you've moved him to 3B or CF. And Mookie is a better defensive 2B than Xander is a SS. I'm not quite ready to say that Mookie is the better prospect, but it's becoming evident that he, like Xander, is a once-in-a-generation talent. Who are nevertheless happening simultaneously! (And including defense, so is Vazquez. Golly.)
|
|
|
Post by thelavarnwayguy on Sept 7, 2014 13:30:20 GMT -5
I just don't buy the pitch framing issue as anywhere near as important as it has been given credit for. No way Vasquez is a plus DEF of 36 as compared to Swihart's plus 14.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 7, 2014 14:15:20 GMT -5
I just don't buy the pitch framing issue as anywhere near as important as it has been given credit for. No way Vasquez is a plus DEF of 36 as compared to Swihart's plus 14. He's not. We have no idea what Swihart's pitch-framing skill is, so I just assumed it was average rather than guessing. But he's reported to be good a receiver and almost certainly will close some of that gap. In the Swihart vs. Vazquez thread, there's discussion of how good a framer Swihart has to be to be the better player. Your general complaint, that pitch framing isn't as important as studies indicate it is, is pretty much like asserting that you don't buy that the Earth is more than 6,000 years old, and -- short of a rationale for your objection, which I note hasn't appeared yet in the Vazquez pitch-framing thread or Vazquez vs. Swihart thread -- is about as interesting to the rest of us.
|
|
|
Post by thelavarnwayguy on Sept 7, 2014 14:40:33 GMT -5
Possibly I am misinterpreting the data from what little I have had time to look at it today, but it appears you are claiming that Vasquez's defensive ability is of comparable impact to Andrelton's Simmons overall defensive contributions. That seems unlikely in itself.
And Swihart who has been a solid defender so far is 20 runs less impact (Edit: 22 runs less impact)? Swihart has a lot fewer passed balls and has thrown out runners at about the same pace. That's a lot of impact from pitch framing (mainly). More impact than all the rest of Swihart's overall defensive abilities put together.
I don't think that is likely to be accurate.
|
|
|
Post by thelavarnwayguy on Sept 7, 2014 14:42:36 GMT -5
Let me also say I appreciate the data overall a lot and don't mean to nitpick.
|
|
|
Post by philsbosoxfan on Sept 7, 2014 19:59:42 GMT -5
I'm not sure how catcher defense is calculated but I'm pretty sure there's no way Weems is anywhere near a minus 21 defensive catcher. Also no way KDLC is a better defender than Ramos.
I do want to add that I appreciate the data as well and understand that minor league defensive metrics need to be digested with a large grain of salt.
ADD: I realize that you don't take Shandler's data manipulation approach but looking at the defensive numbers and just pure amateur observation, most numbers are within reason but there are 4 that I can't see.
Weems, relative to other catchers numbers, I'd have had him slightly on the positive side.
Ramos around neutral, he's somewhat raw but has plus defensive tools, should be closer to 0
KDLC I think is the worst defender relative to level in the minors, no defensive tools and makes a ton of bad decisions. Think Manny Ramirez with less speed and less arm.
Blair, I'd think lower, he has significant problems with pitches in the dirt. Maybe -12 to -15. If there's a stat which is pitchers wild pitches in games caught, I'm pretty sure you'd find that to be significantly higher than and of our other catchers.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 7, 2014 20:42:15 GMT -5
I'm not sure how catcher defense is calculated but I'm pretty sure there's no way Weems is anywhere near a minus 21 defensive catcher. Also no way KDLC is a better defender than Ramos. I do want to add that I appreciate the data as well and understand that minor league defensive metrics need to be digested with a large grain of salt. 1) KDLC is indeed four runs worse than Ramos as a RF, but Ramos is being ranked as a CF. 2) Ramos the last three years was -13, -23, -5. If you've only watched him this year, you're seeing a probably tangibly improved guy. Of course, KDLC has gone -16, -19, -2 ... but that still makes him a worse defender when you deduct the 6 run difference between CF and RF. 3) The catching metrics are among the most reliable, since they're based on SB, CS, WP, and PB. You may have been lucky and only seen him in games where they didn't run on him at will and/or balls didn't go through him like a colander. But if you're implying that he's a much better receiver than that rating (-42 / 150 unregressed), that's good to know. And indeed these ml fielding numbers are designed a) as a starting point for your own adjustments, and b) to make us question scouting reports a bit, on the rare occasions when they disagree. Some guys make a lot of plays without looking all that good at it, or vice versa ( a la Jeter).
|
|
|
Post by philsbosoxfan on Sept 7, 2014 21:06:24 GMT -5
I really find the Weems metric to be odd. He's athletic, moves well behind the plate and has a cannon for an arm with decent but not exceptional throwing mechanics (his mechanics are nowhere near Vazquez fast). From limited observation, he was also pretty accurate. I pretty much have taken the view that his defense will carry the bat rather than the other way around in spite of the rapid improvement in batting stats.
Ramos should be ranked as a RF, that's where he played all year and likely where his future lies. It's unfortunate that Ramos had the stress fracture when he did, his game was coming together nicely. Hopefully the trainers are familiar with micro-calcium products for 3rd world country athletes. Yes, this is the only year I've watched Ramos. This year, Pawtucket, Portland and Greenville broadcasted. Most of the CAR league including Salem doesn't.
|
|
|
Post by philsbosoxfan on Sept 7, 2014 21:34:04 GMT -5
With Carson Blair I'd have to give the SSS benefit of the doubt because Salem doesn't broadcast and Portland's camera angle is terrible for evaluating catchers. I think though, in his first 3 or 4 games with Portland there were 3 strikeout/wild pitches where the runner advanced to first. You can't see what actually happened but from the back it appeared than Blair goes down and blocks the pitch but it bounces in front of him and too far to avoid a wild pitch. I don't think I've seen 3 strikeout/wild pitch combos all year otherwise.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 8, 2014 0:23:25 GMT -5
Possibly I am misinterpreting the data from what little I have had time to look at it today, but it appears you are claiming that Vasquez's defensive ability is of comparable impact to Andrelton's Simmons overall defensive contributions. That seems unlikely in itself.And Swihart who has been a solid defender so far is 20 runs less impact (Edit: 22 runs less impact)? Swihart has a lot fewer passed balls and has thrown out runners at about the same pace. That's a lot of impact from pitch framing (mainly). More impact than all the rest of Swihart's overall defensive abilities put together. I don't think that is likely to be accurate. I'm guessing you haven't read any of the stuff about pitch-framing, then. (Hint: when offering a skeptical or contrary opinion in a forum where folks often assume obsessive interest in the subject matter, briefly mentioning that you haven't been keeping up with the topic will prevent people like me from assuming you have rejected the conventional wisdom -- and hence getting snarky at you, for which I apologize! -- when in fact you just haven't encountered it.) The conclusion people have reached is that it pitch-framing has an enormous impact, making it easily the biggest defensive component in all of baseball. That Vazquez might be the best defender in the game because he's he best pitch-framer (or second to Jose Molina) and the best at controlling the running game is not a surprise to anyone. I think I can quickly convince you that the bolded statement is obviously wrong, and plain backwards. According to Inside Edge, Andrelton Simmons has had 73 non-routine fielding chances this season, in 1135 innings, or 128 games. That sounds right, doesn't it? One chance to make a good or great play a bit more often than every other game. Christian Vazquez has caught 2966 close pitches in just 337 innings. That's 9989 per 1135 innings. That's a rather larger number. (It also sounds right, too.) This year, the Sox staff has thrown 1289 IP, so I'll multiply their staff numbers by 1135/1289 = .88 to continue the Simmons comparison. Per 1135 innings, 197 PA (224 * .88) have ended in a full-count walk. 144 have ended in a full-count strikeout. 70.7% of all strikeouts have been looking, but full-count hitters may be a bit more free-swinging, so we'll use the 65% of all strikes swinging or looking that are swinging to say that 93 PA (per 1135 innings) have ended in a called full-count strike three. Now, some of the 197 ball fours are well outside of the zone, but figure that 120 - 150 are framing chances ... that's something like 210 - 240 fielding chances that have an out versus runner-reaches-base outcome, where different pitch-framing would have reversed the outcome. Three times what Simmons is getting, or more. And then you have to add in all of the 2 strike and 3 ball framing chances on non-full counts, where effective framing can grab a strikeout versus letting the PA continue, or keep it alive versus allowing a walk. All of those are the equivalent of half of one of Simmon's fielding chances. And then add the impact of framing, say, the first pitch, which creates a 250-ish OPS point shift in the likely outcome of the PA. I mean, the guy handles every pitch of the game, many of which are borderline and which he can influence. Obviously his defensive impact will be the largest on the field. Finally, Vazquez's current framing -- 35.3 runs saved per 1050 innings -- amounts to cutting 0.30 off the team's RA. Well, folks have always thought that a great defensive catcher might be able to take 0.30 points off a team's ERA, based on looking at catcher's ERA's. We just were unsure how they might be doing it. Now we know.
|
|
|
Post by thelavarnwayguy on Sept 8, 2014 5:18:20 GMT -5
It may well be that I do not have the level of information needed to make an informed decision regarding things like pitch framing and catcher ERA but what little I have studied it leaves me unconvinced it has the impact some give it. For example, if pitch framing is so important and Jose Molina is the best in the game at it how come he has never made more than $2,125,000 in any of his many years in mlb? No GM in the league thought it was so important that they were willing to pay much for it.
I readily admit that I haven't read a single word of the Vasquez pitch framing thread. I'm sure it has impact. I'm just not sure it has the impact people give it. I apologize for my ignorance if that is the case in this instance. Sincerely.
|
|
|
Post by jimed14 on Sept 8, 2014 9:17:26 GMT -5
It may well be that I do not have the level of information needed to make an informed decision regarding things like pitch framing and catcher ERA but what little I have studied it leaves me unconvinced it has the impact some give it. For example, if pitch framing is so important and Jose Molina is the best in the game at it how come he has never made more than $2,125,000 in any of his many years in mlb? No GM in the league thought it was so important that they were willing to pay much for it. I readily admit that I haven't read a single word of the Vasquez pitch framing thread. I'm sure it has impact. I'm just not sure it has the impact people give it. I apologize for my ignorance if that is the case in this instance. Sincerely. Because all of this stuff is in its infancy.
|
|
|
Post by raftsox on Sept 8, 2014 9:24:23 GMT -5
I enjoy these types of things, but there should really be a minimum # of PAs (perhaps, per season) ascribed. When I see people like Blair, Gragnani and Denney show up high on these lists it reinforces that context is very important.
Aside from this year Blair has been on the short end of a catching duo since day 1; now he's in his age 24 season and played most of the year in High A. Gragnani is in a similar boat in that he's much older than the majority of "prospects" at his level.
I don't necessarily have any real complaints with the results of your work, but I find prospect status really needs to look beyond the boxscore to include tools and context. Any rankings/analysis that doesn't consider those aspects of the player is flawed.
|
|
sarcasmo
Rookie
Formerly known as mtomeo
Posts: 91
|
Post by sarcasmo on Sept 8, 2014 12:37:11 GMT -5
The conclusion people have reached is that it pitch-framing has an enormous impact, making it easily the biggest defensive component in all of baseball. That Vazquez might be the best defender in the game because he's he best pitch-framer (or second to Jose Molina) and the best at controlling the running game is not a surprise to anyone. I think I can quickly convince you that the bolded statement is obviously wrong, and plain backwards.......[EDITED] I get what you are saying, but out of those actual chances a catcher has to frame a pitch, how often do they actually affect the umpire's decision? I would imagine it's a very small portion of the time. Like 1 in 10 frame-able pitches. Just pulling that out of nowhere, but based on my layman's opinion while watching baseball games, it rarely impacts the call. I suppose it could be as high as 1 in 5. If I'm not mistaken, the ump is trained to watch the ball, not the catcher's mitt. On top of all that, most infielder's chances are rather binary. Only two outcomes: out or no out. With the catcher's total chances, there are several outcomes: ball, strike, PB, WP (, etc?). If a SS doesn't make a play on one of their chances, it costs the team an out. If the catcher doesn't frame a pitch, who knows what it will lead to. The framing could come into play on any pitch count, not just one that leads to an out/walk. Plus, the pitcher could overcome bad framing by pitching to contact or simply striking the guy out of 4 pitches instead of 3 (though I definitely see a benefit here as it keeps the pitch count down). Love the data, just not sold on pitch framing having the largest impact of any defensive position.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 8, 2014 19:05:29 GMT -5
I enjoy these types of things, but there should really be a minimum # of PAs (perhaps, per season) ascribed. When I see people like Blair, Gragnani and Denney show up high on these lists it reinforces that context is very important. Aside from this year Blair has been on the short end of a catching duo since day 1; now he's in his age 24 season and played most of the year in High A. Gragnani is in a similar boat in that he's much older than the majority of "prospects" at his level. There definitely should be a SSS adjustment. I'm just hesitant to do one until I feel it's correct. The whole point of these translations is that they do attempt to adjust for age relative to league. They do so less severely than most anecdotal, seat-of-pants adjustments. That's based on a wealth of data. I've had my own attempt at ARL adjustments going back at least as far as when Lew Ford hit .315 / .390 / .479 in low-A at age 23 (1990, Augusta), which my system dismissed. Incorrectly! I suspect our memories filter out the Lew Fords as well as the guys who shot up the ladder and debuted young in MLB, but were nothing special or even busts. At age 23, Delmon Young had 1851 MLB PA. Now, what I think may be missing from the ARL adjustments is makeup. I think that if guys who didn't have the makeup to survive being rushed had been held back, you'd see a steeper ARL adjustment. IOW, David Clyde and his ilk are making it look, unfairly, like being real young for your level is no big deal. So guys like Xander and Mookie perhaps are being underestimated, and perhaps the entire ARL element of these rankings is a bit understated. But it's important to remember that the ARL adjustment is already included, and that it often does reject guys who are putting up good numbers. I have no problem moving a Gragnani down quite a bit because he has no scouting pedigree, but I'm not going to do that for Blair or Asauje, who do (they do get moved down some because they are surrounded by guys who were very high daft picks or got huge bonuses). Not flawed, but obviously incomplete, like half a meal. I use these rankings as a starting point. First, I see if there's a hitting trend that means the "Net" number is too high or low; sometimes you want to substitute 2014 for Net. Next, you can adjust the defensive rankings if scouting reports disagree to a great degree. And then comes the big step: moving guys up or down based on pedigree and scouting reports. That's a pure art form. But by starting from the objective performance, you can see where a guy like Longhi or Asauje belongs. The numbers provide a starting point for the subsequent adjustment and prevent the final ranking of a guy being too high or too low. As I note in the meta forum, out of 23 top prospects I think the data indicates that just 4 need to be modified.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,911
|
Post by ericmvan on Sept 8, 2014 20:17:55 GMT -5
It may well be that I do not have the level of information needed to make an informed decision regarding things like pitch framing and catcher ERA but what little I have studied it leaves me unconvinced it has the impact some give it. For example, if pitch framing is so important and Jose Molina is the best in the game at it how come he has never made more than $2,125,000 in any of his many years in mlb? No GM in the league thought it was so important that they were willing to pay much for it. I readily admit that I haven't read a single word of the Vasquez pitch framing thread. I'm sure it has impact. I'm just not sure it has the impact people give it. I apologize for my ignorance if that is the case in this instance. Sincerely. Because all of this stuff is in its infancy. And until there are more GMs who value pitch framing than there are elite pitch framers, they won't get paid an extra penny for their skill. Although as I'm about to note, they will keep jobs they otherwise would not have. Here's a better question: why have the Rays given Jose Molina 599 innings behind the plate when he has a -1.9 WAR per 120 games (1050 innings), based on offense, DRS SB Runs rating, and FG Blocked Pitches Runs Saved? Could it be that when you include pitch framing, he's +1.8 WAR, which is borderline starter quality? Or even better, how could it be that the Orioles barely missed a beat when they lost 4.9 WAR / 120 catcher Matt Wieters and replaced him with 2.2 WAR catcher Caleb Joseph? Could it be that when you include pitch framing, it's 5.2 and 4.4? You could go on like this. Why does Ross still have an MLB job when he's 0.3 WAR / 120? Because he's 3.0 when you include pitch-framing. (Note all the pitch framing values here are actual runs saved by count, which includes a little clutch luck, but it's real and has happened in 2014, and was much simpler to calculate.) Now that I understand where FG's defensive WAR for catchers comes from (the aforementioned two components plus .0086 runs per inning), I can create an fWAR leaderboard that includes BP pitch-framing. That will be interesting, and I'll put it in the Vazquez framing thread.
|
|
|
Post by raftsox on Sept 8, 2014 20:55:29 GMT -5
Makes sense Eric. I understand your logic now.
Edit: Follow-ups.
If I had to make any sort of cut-offs, I would probably remove any and all stats from the DSL and GCL.
It's interesting you're working on this; a few years ago I attempted to do something similar and much less sophisticated. However, I lacked the time, ability and ambition. One manner I considered adding in scouting aspects was to incorporate the 20-80 scouting grades when available.
|
|
|
Post by philsbosoxfan on Sept 8, 2014 22:13:10 GMT -5
I'd almost be willing to place a bet that Swihart ends up being a plus framer as well (but not likely as good as Vazquez because of his pre-draft background). From what I gather (no specific source so, feel free to correct if it's not the case), the entire minor league catching corps is overseen by Chad Epperson, former Salem manager and former MLB defensive oriented catcher.
My timeline might be off but it seems that at about the same time that catching guru Gary Tuck mysteriously retired then was mysteriously coaxed out of retirement by the MFY the the Sox have beefed up their minor league emphasis on catching, both in the subsequent drafts and by using Tek for special projects (and Gedman as hitting coach in Portland). I'd also guess that Ross will also eventually be involved.
|
|
|
Post by Chris Hatfield on Sept 13, 2014 9:59:57 GMT -5
Have to disagree with their "increased emphasis" on the position. Tek happened because he retired. Gedman being the hitting coach doesn't really have a ton to do with him being a former catcher.
As for acquisition of amateur catchers (Draft/UDFA + IFA), disagree that there's any particular emphasis there either.
2014: 4 (they're converting Winterburn to pitching, or so we've heard) 2013: 3 + 2 2012: 1 + 3 (I'm not counting Watkins) 2011: 4 + 2 2010: 4 + 3 2009: 3 + 1 2008: 3 + 3 (Lava, FedEx, Vazquez in rounds 6, 7, 9) 2007: 5 + 3
|
|
|