markos

I think this is a great post. Thanks for sharing your research.

Walker's AA BB% may be above-average compared to the MLB walk rate, but it is still below-average relative to the Southern League. Almost everyone walks more in the minors than in the majors because the pitchers generally don't have a good control, so it is best to try to avoid making direct minor-to-major stat comparisons. Context matters. Regarding the question for the thread, I honestly don’t know how to rate Walker right now, which makes it hard to judge if he is ‘overrated’ or ‘underrated’. So much of his value depends on how his hit tool will end up playing in the majors, and that is really hard to know without actually facing major league pitchers. The example I keep thinking about in situations like this is Oswaldo Arcia. Across the board in the minors, Arcia raked. And overall, his rate stats were also solid - certainly better than Walker’s. Back in 2013 I would have bet a lot of money that Arcia was going to hit. And most of the experts agreed. There were a few quiet questions about perhaps some contact issues, but most of the agreement was that he would hit. But as we all saw, after a few months in the majors it became very apparent that he had massive holes in his swing, a poor approach, and the inability to change either. Personally, I don’t think the odds of Walker becoming an average or better major-leaguer are good. But I’m certainly not very confident about that, since so much of his future value is dependant on how he is able to respond and adjust to major-league pitching. And projecting the hit tool is notoriously difficult to do. In Walker’s favor, he will have a wide margin for error if he is able to maintain a .200+ ISO.

Meyer's BABIP was high, but this is now three years in a row that his BABIP has been at least .317. Maybe he has been unlucky these past three seasons. Or maybe he has a tendency to groove a lot of hit-able pitches because he lacks command. Personally, I'm not terribly optimistic about Meyer's chances of being an above-average starter. 1) He hasn't demonstrated any ability to throw a lot of innings in a season. He has maxed out at 130, and had shoulder issues in both 2013 and 2014. 2) His release point has dropped A LOT - roughly 12 inches since 2012 - and now his release point is almost a foot lower than Kyle Gibson's and basically the same Tyler Duffey's. For all the talk of Meyer being able to generate extreme "downhill plane" based on his height, that isn't actually true any more. 3) He doesn't get a lot of swinging strikes relative to what his "stuff" indicates he should. Last year at AAA, his swinging-strike rate was 11%, which would put him roughly 70th among MLB relievers. And Berrios had a higher swinging-strike rate. He may still have the highest potential of any pitcher in the organization, but I think the likelihood that he reaches that potential is very, very low right now. Certainly much lower than when they acquired him.

Apologies if you already know this... Fangraphs calculates each minor leaguer's wRC+ relative to the league they are in. So Vielma's wRC+ of 93 puts his offense about 7% worst than average in the Florida State League. It is a nice feature since the various minor league levels have such different run environments. http://www.fangraphs.com/statss.aspx?playerid=sa657416&position=SS

What do Buxton and Berrios need to show during spring training for them to be on the MLB roster on opening day? Do the questions have to be specific to offseason planning? I have other questions: What were your favorite moments of 2015? What other organizations do you admire? Who is your favorite TwinsDaily blogger?

I threw out the general population stat to indicate that at some level height matters. If the vast majority of people are under 6'2", but the vast majority of elite MLB pitchers are over, then height provides at least a very crude tool to differentiate players. The question is where is it useful and where is it not. I agree that there does not appear to be a height bias at the MLB level. But does it exist in the with prospects? Or drafting? Or with signing 16-year-old international kids? Hopefully this example will demonstrate what I mean (and oh my god this is so far off topic...) "Short" defined as 6'1" or less, and all numbers made up... (X% - Y to be read as "X% of Y are 'short'") Scenario 1: 95% - the general population 80% - all baseball pitchers 50% - pitchers for competitive hs traveling teams, college, etc 25% - pitchers good enough to be drafted/signed 25% - pitchers who reach AA 25% - pitchers who make the majors 25% - pitchers who qualify for the ERA title 20% - pitchers in the top 10 in league ERA In this scenario, height is clearly not a good indicator for any player acquisition. If a pitcher is good enough to be drafted/signed, beyond that height doesn't really matter at all. Scenario 2: 95% - the general population 80% - all baseball pitchers 50% - pitchers for competitive hs traveling teams, college, etc 40% - pitchers good enough to be drafted 35% - pitchers who reach AA 30% - pitchers who make the majors 25% - pitchers who qualify for the ERA title 20% - pitchers in the top 10 in league ERA Here, height does matter, as at every step the population of pitchers gets progressively taller. This scenario would indicate to me that height does matter to some degree, at least in the fact that it correlates with success. I don't have any data to know what the actual numbers are. But my theory is that at a certain age/level (maybe 22/AA), height stops being meaningful. But prior to that it is meaningful, primarily because it is a proxy for identifying which young pitchers will develop the stuff necessary to succeed in the majors. Obviously height isn't the only thing that matters, or even one of the main things. But when dealing with draft/signing/trade decisions, when all other things (stuff/results/health/etc) are equal I would go with the tall pitcher over the short pitcher.

The "total population" I was referring to was the population of all baseball players worldwide, not just the much, much, much smaller group of MLB pitchers who qualify for the ERA title. Only 5% of the men in the world are at least 6' 2", yet they make up the vast majority of MLB pitchers.

78 pitchers qualified for the ERA title last year. 56 were 6' 2" or taller. Of the top 10 by ERA+, only 2 (Greinke and Gray) were short. My own pet theory is that taller pitchers are more likely to be able to generate the "stuff" (velocity, movement, etc) necessary to be a successful major league starter - the population of "short" players is much bigger than the population of tall players, but tall players make the larger portion of successful pitchers. That's probably why height is used as a proxy when drafting HS-ers and signing Latin American players. It is probably less of an distinguishing factor when drafting college players, as they have already gone through several years of winnowing. And once players reach (or are on the cusp of) the major, height doesn't really matter too much.

This was a fun little exercise, and more difficult than I expected. Here are my tiers: Tier 0 - Sano Tier 1 - Gibson, Dozier, Berrios, Buxton Tier 2 - Kepler, Plouffe, Escobar, Gordon Tier 3 - Duffey, May, Rosario, Hughes, Jay, Polanco Tier 4 - Milone, Gonsalves, Stewart, Hicks, Perkins Tier 5 - Walker, Meyer, Jepsen, E. Santana, Vargas, Rogers Tier 6 - Burdi, Nolasco, D Santana, Arcia, Pinto, any of the other college relievers

I think Wade should be higher. As for my reasoning, one common thread for over-achieving later-round college draft picks is an elite ability to control the strikezone in the minors, ideally with a very low strikeout rate and more walks than strikeouts. Some names that fit that profile: Dozier, Matt Carpenter, Matt Duffy, Brett Gardner, Brock Holt, Ben Zobrist and Sam Fuld. And since it sounds like Wade is athletic and can play a legit centerfield, he seems well on his way to being at least a 4th-outfielder. Granted, almost all of his success was in Rookie ball, so we will have to see how his plate discipline maintains as he is promoted. But right I would be tempted to rank him ahead of the two high school kids they drafted in the 3rd and 4th rounds. As for Michael, I do think there is a legitimate chance he is drafted in the Rule 5 draft if left exposed. As Thrylos has mentioned, when he is healthy he has been able to hit pretty well at AA. He is a switch-hitter, and if he can adequately play all the infield positions, he may be a decent and cheap bench option for a lot of teams. And in some ways his injury-prone nature may make him more inviting, as teams can probably expect him to spend a decent amount of time on the DL, which will make it even easier to keep him stashed all season. So I also hope they protect him.

While it is rather obvious that high velocity is neither necessary nor sufficient for MLB pitching success, I don't think there is any question that it has some valuable as a predictor of success. Since 2011, there are 677 starters that threw at least 100 innings in a season. Running a quick correlation with ERA- and FIP-, there is a correlation between velocity and success. It isn't terribly strong (.35 for FIP-, .23 for ERA-), but it is stronger than GB% and similar in strength to BB%. It lags rather far behind K% and other swing/chase metrics (O-Swing%, Z-Contact%, Contact%, SwStr%). As other people have already pointed out, getting strikeouts (or at least swinging strikes) is probably the more important skill than pure velocity. Of the above sample, ~20% averaged at least 93 MPH with their fastball. As a group, the 93+ club was significantly above average (90 ERA-) and made up 27% of all the above-average pitchers (ERA- < 100). Basically, that group was 67% more likely to be above-average than the <93 group. Looking at elite seasons (ERA- < 80), the difference is even more pronounced. The 93+ group had 38% of the elite seasons, which means that a pitcher throwing 93+ is 245% more likely (assuming I did my math correctly) to have an ERA- of <80. Even though we can all come up with examples of great pitchers without great velocity, in all actuality they are rare and difficult to predict ahead of time. Dallas Keuchel has been amazing while barely cracking 90 MPH, but two years ago no one was predicting he would be good, much less one of the best pitchers in baseball. And at this point nobody knows who the next Keuchel will be. That kind of ends up being the crux of the problem - it is very hard, perhaps impossible, to look at a high school or college kid sitting 90-91 and know that they will have the right combination of stuff, control, command, pitch-ability, deception, projection and moxie to consistently miss bats against major league batters. Sure, that kid might become an ace, but the probability is much, much lower than a kid throwing 95.

I think Wade is promising as well, even though he was kind of old for the level. As a differentiation between him and Max Murphy, note that Wade ended up with a super-low strikeout rate (12%) and more walks than strikeouts. I think that bodes well for him going forward. If he plays a legit centerfield and keeps his command of the strikezone, he is well on his way to being a least a 4th or 5th outfielder. That's pretty good for a 9th round pick.

I hope this link works.

This report really worries me. My biggest fear about this coming offseason is that the success will allow them to decide that the best course of action is to try to bring everyone back. So a lot of positions that should be upgraded (either internally or from outside the organization) will instead by given to the incumbent. Hunter, Nunez, Robinson, Fien, Duensing, Boyer and Herrmann all fall in this category.

The argument, at least how I see it, isn't whether or not Berrios is ready for a major league look. I think he is clearly ready. However, the argument is about whether or not the Twins are in a position to give him that look knowing there is a real possibly that he will be terrible. He actually might not pitch well enough to help the team right now. Lots of really good pitchers have struggled on their initial callups. Carlos Martinez had a 5.08 ERA during his initial callup. Chris Archer had a 4.60 ERA in his callup. Archie Bradley had a 5.80 ERA. Julio Teheran hada 5.19 ERA. Heck, even Jose Fernandez had a 4.50 ERA after the first 5 starts of his career. Those are hardly solid numbers, and they were all just as well regarded as Berrios. And just in our recent experience we've seen Alex Meyer, Trevor May and Tyler Duffy all get bombed in their first appearances. Right now, in the middle of a tight pennant race, each future win/loss has a huge impact on the Twins playoff chances. There is real risk in giving Berrios a few starts or a bunch of high-leverage relief innings if he needs any kind of adjustment time. The risk could be lowered by giving him low-leverage innings, but I have zero interest in blowing a year of control so he can do mop-up work. Maybe that risk is worth it for the potential benefit of having a great arm if Berrios hits the ground running. I think we all balance that risk-reward differently. But for what it's worth, I haven't seen anyone present thorough evidence (other than a handful of cherry-picked examples) that top prospects are always or even predominately above-average during their first 10-20 innings in the big leagues. If someone has seen or done that research, I'd love to see it, and I will change my mind accordingly. But right now I think think that Berrios, like many, many other pitchers, will most likely need an adjustment period in the big leagues where he will be pretty terrible, and the Twins are not in a position to afford that right now. Next spring is different. Part of his adjustment period will be facing big league hitters in spring training, where there is zero risk if he bombs. And then any starts in April/May just don't have the same impact.

I never said it wasn't. I said that after his 0-12 start to the season, he had a .680 OPS through May 15th. "The Twins did demote him... But then they inexplicably called him back up and played him almost every day for five weeks and June and July." He was the starting shortstop 16 times in the 30 games during that stretch, plus 3 games at CF. That is hardly every day. "Defensively, Santana has a wince-inducing -14 DRS in just 568 innings this season. UZR is equally dismissive of his play in the field." From above, he only played ~144 innings at SS during that June-July stretch. Pro-rate the DRS, that is -3.5 runs. Again, the improvement of moving from Santana to Escobar for 16 games might, maybe, in the best of circumstances, be 1 win. Maybe. Probably way less than that. So I just don't believe that calling up Santana again in June was some massive mistake that caused the Twins to fall out of contention. Mistake? Yes. Season altering mistake? No.

Your argument depends a lot on the benefit of hindsight. Santana started the season 0-12, but then for the next 30 game through May 15th, he had a .680 OPS. Not great, but certainly not terrible. Escobar, over that exact same timeframe, had a .600 OPS. Escobar is a better bet defensively, But I think it was close enough that it was NOT completely apparent on May 15th that Escobar was the right choice and Santana needed a reboot. After May 15th, Santana did turn into a dumpster fire, and had a meager .305 OPS for the next three weeks, at which point the Twins sent him to the minors. Maybe they should have sent him down a week sooner, but the team was winning during that stretch, and it seems rather impulsive to send someone down after a bad week. So I'm arguing that May 15th only makes sense if you know the future. With hindsight, yes, that is obviously the day that Santana should have been sent down. However, I don't blame the Twins for using him through June 1st. And therefore I don't think his limited playing time after that date has been a big mistake. You are talking about 24 starts scattered over 2 months, during which he had a .524 OPS. Using Escobar instead for those games (assuming he is a ~.700 OPS player) is maybe one win improvement. Maybe? Probably less. Of all the bad things that have happened to this team since June 1, starting Danny Santana 24 times is pretty low on the list.

If you strip away the A-Rod contract (and remove the Roger Clemons one year deals), the broader upward trend holds. 1) Highest AAV in 2000 was Kevin Brown at $15MM. Highest AAV in 2009 (again, non A-Rod) was CC Sabathia at $23MM. Increase of 53%, which is ~4.5% increase per year 2) Highest AAV in 2009 (again, non A-Rod) was CC Sabathia at $23MM. Highest AAV in 2015 is Cabrera at $31MM Increase of 38%, or 5.1% per year. 3) Cabrera signed his first big deal in 2008 with a $19MM AAV. He just signed an extension that is a $31MM AAV. 63% increase in only 7 years (7.2% per year). 4) Only 6 players in 2010 had ever reached an AAV over $20MM. Now there are 39 players. I will concede that 8% salary inflation is probably too much, but I definitely think that at least 5% is what we are seeing right now. If trends continue, I don't see why good-but-not great players won't be getting $30MM AAV in 6 years, just like how $20MM AAV contracts are given today to players like Choo, Porcello and Sandoval.

Here are a couple of articles on Fangraphs about recent extensions. http://www.fangraphs.com/blogs/the-freddie-freeman-deal-as-a-market-correction/ http://www.fangraphs.com/blogs/the-perfectly-reasonable-christian-yelich-extension/ In the first article, Cameron points out maybe 6 total deals in the last few years (out of 50 total) that could be considered fails. Also, 8/$73MM to someone with less that 2 years of service time would be really close to other deals. Looking over the list of extensions, the biggest contract for someone with less that 2-years is Andrelton Simmons with 7/$58MM. Other deals for the likes of Yelich or Rizzo are more in the range of 7/$40-$50 with at least one additional option year. Sano's deal is a little more expensive, but given baseball salary inflation and the extra year, it isn't way out of line.

I think the deal that Seth laid out has a strong probability of saving the Twins a ton of money. Just looking at the arbitration years, Chris Davis just make $25MM over his three seasons, and he only had one superstar year and was actually pretty bad during his pre-arb seasons. Davis's arb seasons were in 2013-2015, whereas Sano's arb seasons will come in 2019-2021. Baseball salary inflation is pretty fast (8%-ish), so it wouldn't shock me that $25MM in 2013-2015 is $40MM in 2019-2021. And that is the baseline of a decent slugger with one great season. Imagine if Sano is actually really good, and then the arb numbers are probably $50MM-$60MM. Seth's deal limits his arb years to $25MM, which I think is easily a $15MM saving, and likely a $30MM or more.

Using wins instead of runs in useful because the run environment changes from season to season, so it is hard to compare players from different eras if you use RAR or VORP. So just think of it as a handy conversion factor, just like how people adjust for inflation when comparing economic data.

Yes and no. From a straight WAR perspective, he is basically replacement-level, which would indicate that he isn't providing any value. However, the Twins are 20th in the majors for production at first-base. The position is really weak this season, and there are a lot of teams (Cle, Oak, StL, Col, Mia, Sea, TB, Pit, Hou, Was, Phi) who would be better off is Mauer had been their everyday firstbaseman for the season. So it is by no means great, but there is value in being better than many of one's peers at a position. Contrast the Mauer at first with Hunter in RF. One can argue that Mauer and Hunter have been equally valuable (or not valuable) this year, with maybe Hunter being slightly better by certain metrics. However, only two teams have been worse than the Twins at RF this season - KC and the Cubs.

But most high 1st-round picks do move fast. If you look back at the previous 6 years (2007-2012), there were 13 HS pitchers that avoided losing a full season due to an injury. Four of them appeared in the majors within their first two seasons (Bumgarner, Jacob Turner, Bundy, Fernandez). Another five at least made it to AA by the end of second season (Jarrod Parker, Shelby Miller, Jameson Taillon, Archie Bradley, Robert Stephenson).

Even assuming the best (zero sliders the past two years) his stats aren't completely irrelevant. 1) They provide a good idea of the quality of the rest of his arsenal. At this point, I think it is pretty clear that Stewart is not going to end up with four average-to-plus pitches, despite what was said prior to the draft. 2) They provide a baseline for potential improvement. Once he adds his slider back into his repertoire, how much will his strikeout rate increase? Will it double? Will increase by 50%? Or less? I don't have any firm data on this, but my intuition is that a doubling his strikeout rate is very unlikely (has any pitcher ever doubled his strikeout rate by adding a single pitch? Maybe Santana pre- and post- changeup?), and even a 50% increase (essentially going from Phil Hughes to Trevor May) isn't a given. But at his current 14% K%, a 50% increase will put him at 21%, which is good (maybe slightly above-average) but not elite. So even with all the unknowns about his development plan and pitch usage, I still think it is reasonable to look at what he has done and compare it to the historical context. Sure, there is still a lot of uncertainty, but we can adjust for some of that uncertainty, especially by giving Stewart the benefit of the doubt in certain cases. And even when giving him the benefit of the doubt in all cases, it still is the case (in my opinion) that his best case scenario is league-average, mid-rotation starter.

It is more than shouting in the dark. Every single person who has created a minor league projection system has recognized that either the most or second-most important factor for predicting a pitcher's future success is strikeout rate (the other being age-relative-to-level). And all the projection systems that I could find are in agreement that Stewart is going to be below-average at best, and career minor leaguer at worst. This isn't just a random opinion based on zero evidence. This is a comparison to the hundreds and hundreds of other HS pitchers who have worked their way through the minors. On my own, I've looked back at the 40-odd HS pitchers taken in the first round of the draft in the past 15 years. Exactly one (Chris Volstad) made it to the majors and had any semblance of success with a strikeout percentage in the low minors under 20%. And Volstad was barely above replacement level. Stewart's K% is currently at ~15%. Maybe Stewart is completely unique, and all of the historical trends are completely wrong with him. Or maybe the Twins development program for him is radically different than every other team in the league. Player development is certainly random, especially with pitchers, so it isn't like there is zero chance that he becomes an Ace or even an above-average pitcher. But I would feel a lot better if there was ANY historical precedent of a pitcher like Stewart turning into something better than a back-end starter or bullpen arm.