Hello, and welcome to the first part of my 2010 SONAR review. As you know, last winter (and really it was 2 years in the making) I unveiled a new metric I had created to try and evaluate minor league performance across all levels, taking into account a player’s age, his league he was playing in, his home park, and measure the performance based on his true performance, not just the standard 3 slash stats. I re-worked the stat a few times, trying to get the kinks out, and then rolled out my final scores for the 2010 season. All of the data is found here, as well as a primer on the basics of what the score measures and my rationale for creating the metric. Now that we have a full year of data, I want to re-visit my lists heading into 2010 and see how good SONAR was as a predictor, based on 2009 performances used to create each player’s SONAR score. I’m going to do 6 parts, one for each division, and keep them in separate posts. So check below and we’ll get started.
We’ll start, of course, in the NL East. I’m going to post the chart for each team that lists all of the scores, and then I’ll do a bit of line per line analysis. We’ll start with the Phillies. I’m going to provide the top 30 chart for each team, which contained the top 30 SONAR scores, and then also guys who were considered top 30 prospects heading into 2010 by Baseball America. For the purpose of looking at the Phillies list, I’ll also address guys who were not in the top 30 or mentioned by BA who may have had breakout years and try to figure out why they didn’t score higher last year.
Philadelphia
Anthony Gose, 59.75 – 2010 3 slash –> .262/.332/.393 (A+) 45/77 in SB. Gose received credit for being just 18, very young for Low A, and posting an astronomical stolen base total, while also stealing bags at a high success rate. His 3 slash numbers weren’t overly impressive, but for his age, they were good enough. He was one of the 3 pieces dealt to Houston (and then Toronto) in the Roy Oswalt deal. His bat is still raw, but in the 113 PA he logged for Toronto’s A+ team in the FSL, he did show a bit more, increasing his walk rate and also flashing a bit more power. He is still very very young, and if he was dropped down to Low A next year, he will still be age appropriate. His work at the plate actually improved in 2010, but his work on the bases took a step back. Still, I think SONAR had him properly identified, when you consider his age and his raw ability.
Domonic Brown, 56.18 – 2010 3 slash –> .327/.391/.589 (AA). Brown had an impressive 2009, and from a tools perspective, everyone loved what he was doing, especially because the power had started to emerge. The one thing that may have limited his SONAR score was his performance at Reading after a midseason promotion, where he posted “just” an .801 OPS. His power fully emerged in 2010, and his three slash line represents that big uptick. He’s now considered one of the best prospects in baseball, and has shown flashes of what he can do in his brief stint with the big league club.
Tyson Gillies, 40.10 – 2010 3 slash –> .243/.302/.336 (AA). It was a busted season (no pun intended) for Gillies from the start, as he struggled with minor injuries and then ran into problems with the law. In spring training, he was receiving really lofty praise, with Kevin Goldstein actually saying scouts were viewing him favorably and even making some comparisons to Domonic Brown. SONAR liked his work in 2009, noting his high walk totals and solid speed. Of course it didn’t translate in 2010, though SONAR isn’t able to predict injuries or substance abuse issues.
Antonio Bastardo, 28.48 – 2010 line –> 20.1 IP – 1.77 ERA – 1.03 WHIP (AAA). Bastardo performed well in 2009, but dealt with persistent injury concerns. 2010 was no different, as he pitched well in the minors and struggled briefly in the majors before getting hurt. He’s back and healthy now, but he remains a major injury risk.
Michael Schwimer, 27.51 and Justin De Fratus, 27.46. Both guys continued to pitch really well in the high minors, something which bodes well for their big league futures. Schwimer continues to rack up strikeouts, with 76 in 60 IP this year between AA and AAA. De Fratus was even better, allowing just 48 hits in 65 innings, with 71 strikeouts and 16 BB between A+ and AA.
Domingo Santana, 26.23 – 2010 line –> .211/.329/.333 (A-/A). It was a major struggle for Santana this year, and the Phillies may have simply been a bit too eager to move him quickly (and I was eager too, ranking him where I did in my Top 30). The strikeouts were a big red flag, but one I thought was offset slightly by the solid walk rate in 2009. The walk rate remained very good in 2010, but the power dried up a bit, and his contact issues remain. One area where I think I may need to re-address is the impact of strikeout totals. For the most part, strikeouts at the highest level are not an issue, but in the minors, it appears that huge contact issues could be more of a stumbling block, as the competition level does not get any easier as you move up the ladder. But that’s for another time.
Jon Singleton, 23.31 – 2010 line –> .290/.393/.479 (A). What a year for Singleton, who SONAR liked, but not quite as much as Santana. The big difference here, again, was Santana being a year younger and showing more power. But Singleton’s bat control edge should have ranked him higher. A score of 23.31 in just 119 PA from 2009 was still a very strong score.
Phillippe Aumont, 22.73 – 2010 line –> 122 IP – 5.68 ERA – 1.71 WHIP – 115 K – 80 BB. Aumont was regarded as the best prospect the Phillies got in the Cliff Lee deal, but SONAR wasn’t quite as convinced, and he did struggle (to put it lightly) in 2010. The same things I’ve said previously about him still apply. Getting 122 innings under his belt this year will have greatly helped him, even if he struggled. And he still has plenty of projection. But he’s a work in progress, and will score much lower next season. This metric is, after all, performance based.
Notes: SONAR wasn’t very high on Sebastian Valle and Trevor May. In Valle’s case, it was his approach at the plate, lack of walks, and abundance of strikeouts. He showed flashes this season, and the scouting reports are positive, but he still has plenty of work to do. May struggled mightily at High A before a demotion to Lakewood, where he seemed to get a handle on his control. He faces a big test with his 2nd stint at Clearwater. Jarred Cosart and Brody Colvin had a lower score because both had a very small sample size in 2009. Josh Zied also popped up at the bottom of the top 30, and he of course posted excellent numbers at Lakewood, though at an advanced age. On the negative side of things, SONAR was pessimistic on Joe Savery (most were) and also Mike Cisco, who struggled. SONAR was also down on Vance Worley, who ended up posting a excellent year. Worley’s peripherals were poor last season, but that may well have been the result of an overagressive assignment to AA and exhaustion from his first full pro season. SONAR saw Mike Stutes in a starting role and wasn’t impressed, but he posted a nice season out of the pen, and it’s clear that is his role going forward, he’ll see a score uptick heading into 2011.
All in all, I think there are a number of interesting things in the data. Small sample sizes obviously produce low scores, and some of the big time performances in 2010 (Worley, in particular) didn’t see all that predictable from a statistical or scouting perspective. SONAR did spot the Matt Rizzotti breakout, as well as Singleton’s solid season.
I’m going to post much less detail on the other 4 teams in the division, just the highlights.
* Its no great accomplishment that SONAR blew up Jason Heyward, since everyone else did too. He had the highest SONAR score of any prospect in the minors, and he’s lived up to the massive hype in the majors this year.
* Cody Johnson’s massive score is a result of him excelling at the 2 core principles; walks and home runs. He strikes out a TON, and scouts considered him an all or nothing prospect. 2010 was a bit of a disaster for him, as he went .212/.289/.398 across 3 levels at age 21. I think that’s an important number, because though he failed the AA test, at A+ he went .264/.333/.505, and that is an age appropriate line in a very tough pitcher’s park. As I mentioned above in the Phillies writeup of Domingo Santana, I think I may need to tweak how the system views contact rates and penalize a bit more for strikeouts. I still think Johnson has a shot because of his age, his raw power, and his ability to draw walks. But he has to hit at AA this year.
* Julio Teheran threw only 81 innings last year and struggled with his control, but he shook that off this season, and has emerged as arguably the best pitching prospect in the minors. Scouts were always high on him, and for good reason.
* Two guys who have made an impact, Craig Kimbrel and Brandon Beachy, both show up on the Top 30. Kimbrel’s ranking was right in line, Beachy had gone unranked.
* Benino Pruneda was unranked by BA and ranked 8th by SONAR. He responded with a line of 64.2 IP – 55 H – 37 BB – 93 K at A+/2A as a 21 year old, and is surely going to make the Braves top 30 this year. Gerardo Rodriguez ranked 11th by SONAR and unranked by BA, posted a line of .253/.312/.447 in A+ at age 22 in 356 PA. Respectable raw power, but not the breakout I expected.
* SONAR was no big fan of Jonny Venters, but of course he struggled mightily in 2009 as a starting pitcher before being converted to a reliever and the second coming of Billy Wagner. Mike Minor posted a score of only 7.79, but it was in a very tiny sample. Freddie Freeman struggled a bit in 2009, but was very aggressively pushed. He responded with a big 2010, and looks set to be the Braves starting 1B next year. Scouts were also higher on Randall Delgado, and he responded with a very nice season in High A before struggling a bit in AA. Scouts were also high on Zeke Spruill, a young RHP, and BA ranked him 8th in the system, while SONAR ranked him much lower. He struggled this year, posting a line of 68 IP – 5.43 ERA – 87 H – 14 BB – 42 K in A+.
I’d say SONAR’s biggest “hit” here would be spotting the breakout of the obscure Benino Pruneda, as well as the struggles of Zeke Spruill. The biggest “miss” is obviously Cody Johnson, but I kind of highlighted why I think his season wasn’t as bad as it looked. SONAR also “missed” on Julio Teheran and Freddie Freeman for different reasons.
* Mike Stanton and Logan Morrison scored very very well in the system, and both have shown what they can do in the majors. Stanton’s batting average will always be on the low end because he strikes out a ton, but he might be one of the 3 or 4 strongest players in the game, and he’s just 20. Morrison lacks elite raw power, but has an incredible approach at the plate. Both will anchor the Marlins lineup for the next 4 years until they make more than minimum wage.
* Justin Jacobs was the first big divergence, ranking 4th according to SONAR and going unranked by BA. He showed an interesting blend of power and patient for a 20 year old in Low A, but for some reason, he didn’t play at all in 2010. I don’t know if it was an injury, if he quit baseball, or if he was just released. Very odd. As I mentioned in my SONAR manifesto last year, there are always going to be statistical outliers/oddities, it’s simply the result of developing a program or algorithm that spits out raw data. You take the good with the bad.
* Thomas Hickman, who scouts gave up on, had a nice year in 2009 and it appeared he was on track. He struggled in 2010, going .232/.326/.413 across A/A+. Sonar was higher on Marcell Ozuna than BA, and he had a breakout year in 2010, hitting 22 HR in just 320 PA, most of that time coming in the NYPL at age 19. He’ll see a nice uptick in his 2011 score.
* SONAR wasn’t a big fan of Ryan Tucker, ranked 6th by BA, and he struggled mightily in 2010, with a 5.72 ERA and 1.80 WHIP in 67 innings across A+/3A at age 23. Likewise, SONAR was down on Issac Galloway, ranked 8th by BA. He bombed out to the tune of .200/.259/.290 in 108 PA at A+ at age 20. Kyle Skipworth, a former first rounder, had a disastrous 2009 season, but still ranked 7th in the system by BA. He rebounded somewhat in 2010, hitting .249/.312/.426 in 436 PA at Low A, as a 20 year old. Still, I don’t know that you can consider him a top 10 prospect based solely on his draft position in 2008.
* SONAR sleeper Elih Villanueva (12th, unranked by BA) posted a nice line of 179 IP – 137 H – 34 BB – 115 K in AA as a 23 year old. His low GB% is a worry, but with his excellent control, he should be able to carve out a career. Performance at AA is a big deal. Another sleeper, Brent Keys, posted a line of .267/.360/.313 in 261 PA in the NYPL at age 19. No real power, but a good approach at the plate, and he was 11 for 15 in SB. Peter Andrelczyk, unranked by BA, had a blistering 2009 despite being old for Low A, but he put up decent numbers in A+ as a 24 year old. I don’t think the numbers were overpowering enough, considering his age, to rank him this high again, but I like him as a future bullpen arm.
* Big dissension at the top, as SONAR liked Kirk Nieuwenhuis above all others in the Mets system. He posted solid numbers at AA (.289/.337/.510) before struggling at bit after a promotion to AAA. He showed solid power (ISO of .200) which will play well if he remains in CF defensively. He struggled with contact (132 K in 566 PA), and his speed kind of tapered off a bit. That said, I still like the decision to rank him ahead of most of the guys behind him.
* Jenrry Mejia ranked 2nd, and he had a decent season in between the ML bullpen/spot starting/starting in the minors carousel he was on. Having watched him pitch, I think he’s going to end up a reliever in the long run, which would downgrade his stock for me, but he does have an excellent fastball. However, his shoulder injury has to raise some concern.
* SONAR pegged RHP Josh Stinson as a potential breakout guy, as he was unranked by BA. His overall line (138 IP, 3.90 ERA, 1.36 WHIP) doesn’t look all that great, but he did finsh the season well, starting 4 games in AAA and posting a line of 28 IP – 22 H – 8 BB – 21 K. He generates about 50% groundballs, but doesn’t miss many bats, and his control needs to be tightened up a bit. Still like him as a sleeper at this point.
* Wilmer Flores scored low in the system as a result of being pushed extremely aggressively. He “held his own” against much older players in 2009, but his stat line wasn’t a net positive. In 2010, his line was a positive; .289/.333/.424 in 597 PA between A/A+. He doesn’t walk a ton, but he also has great bat control for someone his age, and his 36 doubles indicate power may be coming. I think he’s the safe pick as the Mets top prospect at this point.
* SONAR didn’t like Brad Holt at all, despite BA ranking him 5th in the Mets system. He responded with a line of 95 IP – 8.34 ERA – 111 H – 79 BB – 87 K (and 17 HBP) at A+/2A. Josh Thole’s approach (no power, great patience) is one that has been tough to translate to the majors, and SONAR wasn’t really buying it. He had decent numbers in the minors in 2010, and he’s hit an empty .277 so far in the majors. It will be interesting to see what he does over the course of 450-500 PA’s in the majors next year, if the Mets give him that chance.
* BA jumped off the Lucas Duda train a few years ago, SONAR was a believer, and he hit .304/.398/.569 this year at 2A/3A in 495 PA. He’s struggled so far in his brief MLB debut, but I still like his bat. Another sleeper, Richard Lucas, didn’t fare nearly as well, hitting just .212/.285/.339 in 486 PA at A+. The third big breakout guy predicted was Brahiam Maldonado, and he fared much better, hitting .282/.342/.539 across A+/2A in 415 PA, with 50 XBH. He does has contact issues, so temper expectations a bit.
* Of course Stephen Strasburg has a score of 0, he didn’t pitch in 2009. Nothing else to say there, really.
* Derek Norris got plenty of scout buzz, and his numbers really only fell just behind Jason Heyward’s in 2009. 2010 was unkind to him, as he dealt with hand injuries (really the worst kind for a power hitter) and managed just a .235/.419/.419 line in 399 PA at A+ as a 21 year old. That said, you can see that the patience is still there, he drew a massive 89 walks compared to 94 K. The hand injury seemed to sap his power. With him 100% healthy in 2011, I expect his overall line to go back up.
* Brad Meyers was great in 6 starts, and I’m assuming he was injured, because he made only 6 starts all season at AA.
* Marcos Frias posted excellent numbers as a 20 year old in Low A last year, but the success didn’t carry over this season, as he struggled to the tune of a 5.68 ERA, 1.51 WHIP, and 65/35 K to BB rate in 96.2 IP at A+.
* Stephen Lombardozzi was ranked much higher by SONAR than BA (28th), and had a nice season, going .294/.371/.431 in 625 PA between A+/2A at just age 21. I think you’ll see him much higher up prospect lists next year.
* After a monster debut in 2009 in the GCL, Hendry Jimenez struggled mightily (.218/.291/.279 in 206 PA at RK), but another sleeper I liked, Josh Wilke, fared a bit better, posting a line of 69.2 IP – 2.45 ERA – 57 H – 22 BB – 62 K at 3A at age 25. He should get a shot at a bullpen gig next year, based on the number or arsonists the Nationals currently employ in their pen.
* Ian Desmond ranked 12th according to SONAR, but was more highly thought of by BA and BP. He’s had himself a decent rookie year, .280/.317/.412, going 16/21 in SB and showing decent pop for a middle infielder. From my watching him defensively, I think he’s kind of shaky there, and if I were the Nats, I’d be more inclined to play the better defender (Danny Espinosa) at SS and Desmond at 2B. But that said, he’ll provide them with a cheap middle infield option for the next 2 years, and he may yet have some upside offensively.
* Ranked 7th by BA heading into 2010, Jeff Kobernus struggled in his first full year, hitting just .279/.316/.346 in 343 PA at Low A despite being 22. SONAR wasn’t very impressed, and will likely be even less impressed this winter.
* Michael Burgess struggled in 2009, resulting in him being dropped to #9 in BA’s rankings. SONAR was still a fan, ranking him #3 in the system. SONAR may have won this round, as he hit .265/.357/.465 in 529 PA across A+/2A at age 21. More impressively, he actually showed improvement upon being promoted to AA, hitting 6 HR in just 74 AB. Small sample size and all of that, but yeah, get back on board with Burgess.
* Tom Milone, who was ignored by BA last winter, ranked 17th by SONAR, and put up great numbers in 2010; 158 IP – 161 H – 23 BB – 155 K – 2.85 ERA at AA at age 23. Milone is a soft tossing LHP, which is probably why scouts ignored him, but he has amazing control, and if he has a decent defense behind him, he’s basically an improved version of John Lannan. That shouldn’t be ignored considering the relative lack of upside in the system.
There ya go. Volume 1 is in the books. I’m going to wait until I’ve reviewed all 6 divisions before writing a wrapup/drawing big conclusions/talking about where the system goes from here. I want to go over all the results first. I’ll move on to the NL Central next. As always, if the SONAR stuff is a foreign language to you, check out the SONAR archive, located here.
I am very interested in SONAR as a kind of Sabermetric evaluation of young talent. The beauty is, sight unseen, you can predict what a prospect will do at the MLB level simply by looking at his stats. But your commentary seems to infer that this has a long way to go, and might need continuous reevaluation, with the ultimate prize being able to pick some stud on the farm–sight unseen, just based on the data–who will be an MLB standout.
One thing I think that is missing is a regression analysis that shows where your assumptions are wrong, based on what major talent evaluators say. I know that in the long term, you would like to have your analysis match MLB player performance, and I also understand that unless you undertake the bewildering task of downloading every players’ data –and accurately–available on SABR or Baseball-Reference, you are pretty much stuck evaluating players on the current year and moving forward. As Bill Clinton once said, I feel your pain. This isn’t a one person job, and what you’ve done, so far, is absolutely great.
But still, one glaring thing to me is that Gose was ranked #1 at 18 yo. I would put him in the middle top ten, as the “experts” did. (Great upside, but, look at his age.) While he is young for his level, the point you seem to be missing is that he is too young. He may never be able to correct what he does wrong now. I know, personally speaking, there are some things I will never be able to overcome in my personal life, like spill coffee every time I pour it. I might have exceptional talent for my age, have great skills that transcend every aspect of my life, and possibly be the smartest person you ever knew. (Not really!!) However, I couldn’t ever be a great waiter at a diner, because I can’t pour coffee without making a mess.
Is there any way I can help? Somehow, SONAR needs to correlate with what the experts predict, in the short term, and MLB success, in the long term.
LikeLike
“Somehow, SONAR needs to correlate with what the experts predict, in the short term, and MLB success, in the long term.”
The latter, yes. I would have thought that part of the interest of SONAR would be the cases where it *diverges* from what the experts predict, and turns out to be right. Otherwise, why not just listen to the experts instead of bothering with SONAR?
LikeLike
“But still, one glaring thing to me is that Gose was ranked #1 at 18 yo. I would put him in the middle top ten, as the “experts” did.”
Well in the original SONAR formula, Gose rated much lower. It seems he tweaked the formula to account for stolen bases and CS. Once that occurred, Gose shot up the Sonar rankings and Gillies fell some.
LikeLike
Is Terrance Warren TJ Warren? He was released and his stats were terrible.
LikeLike
Don’t know that SONAR needs to correlate to what the experts predict in the short-term but its value as a measure will certainly be dependent on the success of the players ranked highly in the long-term.
Also interested to see how the same players rank over several years. For example, what actual value does the SONAR score provide if an Anthony Gose has a down statistical year in 2011 and his score drops dramatically. Is he still the top prospect he was in 201o or is 2011 his true measure? Without taking career progression into accout it seems to simply be another measure of who’s having a good statistical season based on specific measures rather than a true prospect evaluation.
LikeLike
yes Anonymous terrence warren is TJ Warren he was cut during the offseason and then played independent baseball and was cut by the team there
LikeLike
If Worley’s success was not predictable from a scouting perspective, why was he a third round draft pick and why did he skip high A ball? SONAR accounts a bit for skipping a level by using age, but I wonder if college players are at a bit of a disadvantage to get a high SONAR score their first couple of years while they adapt to pro ball. The same might be said for players who missed significant time due to injury or visa issues. Perhaps adding a criterion for games played previously in professional baseball (or innings pitched, innings played, plate appearances, etc) would make SONAR a more useful tool and fine tune the age adjustment.
LikeLike
Your the man! I can’t believe that the Phillies haven’t hired you yet. I mean, how in the world you can provide 100 times more info about the minor league system than anything else out there, is unreal. Hiring you would only increase fan fair, and interest in the phillies as a whole. If you did this fulltime, (you probably already do) without having to work another job, you could even take this further. The options are endless. Interviews, video, etc, it would be awesome.
Great job, beyond any recognition I could ever give you, I can’t wait to read what you write about every morning I wake up. Awesome analysis, as always!
LikeLike
Regarding ken45 – the reason it might be hard to make the kind of adjustment that you suggest is that I don’t think that you could come up with a metric that would be universally applicable. For most college kids the current system I think CORRECTLY puts them at a disadvantage, because they ARE at a disadvantage (ditto your other examples). The relatively rare exceptions …. may be hard to predict. At least without resorting to more conventional scouting reports, which would sort of undercut the purpose of SONAR.
It’s going to be interesting to see what kind of scores the low A pitching staff gets this year. I wonder if maybe the scores might be a little lower than perhaps some of us hope, given (primarily) ball park adjustments. Singleton OTOH should have a sky high ranking, partly for the same reason, and also because SONAR is blind to the seasonal progression (hot start then slump) which I think (unfairly) colors most people’s subjective impression of his season.
LikeLike
I think Justin Jacobs played in the Frontier League this season for the Florence Freedom.
LikeLike
quero meu msn de vouta.
LikeLike
quero meu msn de vouta que ele pegou de mim..
LikeLike