For those sabermetrically oriented baseballing enthusiasts who enjoy moonlighting as prospect mavens, the Arizona Fall League — and the various Caribbean leagues, too, for that matter — represent a conundrum. On the one hand, it’s exciting to see so many young, talented players competing against one another; on the other, the small sample sizes prevent us from making authoritative statements about the respective players’ performances in said leagues, even though we’d like so badly to do just that.

Because of these limitations, we’re almost entirely at the mercy of the Bryans Smiths of the world. No, it’s not “mangy scoundrels” I’m talking about — although that’s certainly an appropriate description of him — but rather “authentic prospect mavens.” Certainly, Smith’s observations on the AFL are helpful, and first-hand accounts are preferable to numbers at this stage.

But if we really insist on ever using winter-league stats — even in the most offhanded fashion — allow me to propose a method by which we might use a hybrid of scouting and, uh, stats-ing.

By way of Russell Carleton’s (a.k.a. Pizza Cutter’s) often referenced, now archived post on the reliability of sample sizes in baseball, we learn that two of the three triple-slash stats — on-base and slugging percentages — don’t become reliable until around 500 plate appearances. That is, 500 plate appearances is the level at which, in Carleton’s words, the “stat can be considered to be saying something about an individual player.” As for batting average, Carleton found that it doesn’t become reliable until somewhere around 1000 PA.

Herein lies at least one of the problems with winter-league stats. Because the AFL leaders in plate appearances rarely top even the 125-PA threshold, we’re forced to regress them over two-thirds of the way back to league average. That creates little in the way of meaningful separation.

An alternative, however, is to look at those categories that (a) become reliable more quickly, but also (b) tell us the sorts of things we like to know about a prospect — namely, the quality of his tools. In this case, we can probably say at least something about contact- and power-hitting — via strikeout and home-run rate, respectively.

Per Carleton’s study, strikeout rate becomes reliable at the 150-PA threshold; home-run rate, at 300 PAs. Those numbers are more friendly than the necessary samples for the triple-slash stats. Even after regressing the remainder of the way* to league average, the resulting adjusted strikeout and home-run rates are distributed widely enough to help us make some kind of observation about player performance in Arizona.

*Which would be, using strikeout rate, for example, [[150 – PA * (Lg K-Rate)] + [PA * K-Rate]] / 150.

Note, please, that there are a number of caveats to make here. First, is that the player pool — i.e. almost exclusively young and developing players — might very well alter the reliability thresholds of the different metrics. Second, there’s the chance that, because the players are being observed by coaches and working in different roles than they might usually, that they are performing differently than they would under different circumstances. (See Chris Carpenter’s comments, noted in Bryan Smith’s recent AFL Notebook.) Third, the parks in Arizona are weird. Does that change things? I don’t know, entirely. But finally — on the subject of my fallibility — there’s the distinct possibility that I, Carson Cistulli, am the sort of person who knows just enough to be dangerous. I think I’m being responsible here, but I also think having garlic as a pizza topping counts as a serving of vegetables. So, that’s what you’re dealing with here.

With all that as preface, allow me to introduce what I’ll call SCOUT. To devise it, what I’ve done is to find the regressed strikeout and home-run rates (xK% and xHR%) for all the qualified batters in the AFL. Then, for each player, I’ve found the z-score (that is, standard deviations from the mean) in xK% and xHR%, and averaged them (i.e. the z-scores) together. SCOUT is the result of that.

By that method, here are are the current leaders in the AFL:

And the laggards, too:

What I think SCOUT is able to capture — and this is why I think it might have some value — is a couple of underlying skills that inform batting production. Like, consider the case of Conor Gillaspie. His slash-line (.231/.279/.410) is pretty bad. Still, given his strikeout rate (just 7.1% so far, unadjusted) and his pair of homers, we know that Gillaspie probably hasn’t had a horrible fall so far — at least so far as his contact and power tools go. SCOUT helps us see that, I think.

Conversely, among the laggards, we see that Kris Negron is having some success, posting a .282/.364/.538 thus far. But most of that is coming from batted-ball success that he’s unlikely to sustain. In fact, Negron has struck out in a just over a third of his plate appearances. Though one might be tempted to say that he’s off to a good start, there appears to be little reason to make that claim — besides his rather unreliable slash stats, that is. Perhaps, at the very least, we can say that Gillaspie is demonstrating greater success so far as his contact- and power-hitting tools go. That’s not a totally unhelpful remark.

What are SCOUT’s (limited) uses, ultimately? They’re two-fold, I think. First, I’ll be using the number — in addition to, maybe instead of, slash stats — in the offseason notes I’ll be providing here. Second, it can serve, I think, as the tiniest contribution to the much larger discussion over how we might use numbers to complement scouting.

Tomorrow, I’ll begin the offseason notes in earnest and begin looking at some of the leaders more closely.

Courtesy of Jeff Caplan on ESPN Dallas:

“Probably the two questions were: Can [Neftali Feliz] close, and can he start?” [Jon Daniels] said. “You knew that somewhere that arm is going to play. I think we’ve definitively answered one of them. There’s no doubt in anybody’s mind he can close. We don’t know whether he can start and I don’t know if we’re going to find out. We’ve talked about it at some point that we might. We’re not going to close that door, but we’re also not going to speculate on it any more than is necessary. He’s our closer. I would expect he’ll be our closer.”

Knowing well how quickly a development like this can change makes commenting too forcefully a risky proposition. Regardless, the expected initial response here is something like, “How wasteful.” It’s not an indefensible position either. Giving a 22-year-old righty with a fastball that moves like a livid missile a life sentence to the bullpen is not entirely wise. At the same time: what else is Daniels going to say about Feliz while the season is still going on. Announcing that his best reliever with a 2.82 FIP through 100 career relief innings is moving to the rotation right now is just as foolhardy. How about the more nuanced response?

The Rangers have encountered moving pitchers to and from the bullpen before. Look at C.J. Wilson, Joaquin Benoit, and Scott Feldman for examples. Texas is experience rich in these situations and while they are fallible, that wisdom does lend credence to the folks who will defer to their expertise. Besides, their 2011 rotation as it stands today is likely to include the aforementioned Wilson, Colby Lewis, Derek Holland, and perhaps one (or more) of Feldman, Tommy Hunter, and Cliff Lee.

Therefore, there is no lacuna of rotational talent or options available to the Rangers for at least next season. The same cannot be said of their bullpen. That does not mean Feliz remaining in relief is the only option. If somehow the Rangers were afforded the opportunity to swap Chris Davis for Tim Lincecum, they surely would not decline the offer because of their abundance. Answering whether Feliz as starter is worthy of a rotation spot is a difficult task and ultimately guess work.
Read the rest of this entry »

Since September 1st, they have allowed 2.31 runs per nine innings – truly one of the great stretches of team pitching in baseball history. They earned the championship, and their place in history. Enjoy it, Giants fans.

However and whenever it ends, one thing that the 2010 World Series has demonstrated is that even teams without significant contributions from high-priced free agents have a shot at a championship. (Yes, the Giants still have a lot of money bound up in Aaron Rowand and Barry Zito, but the former has been consigned to the bench and the latter didn’t even make the playoff roster.)

The successes of the Texas Rangers and the San Francisco Giants this season would have been impossible without good drafting and development. This has given hope to fans of other franchises such as the Kansas City Royals. While Dayton Moore has made many terrible trades and free agent signings in the past, it can’t be denied that he has done a splendid job of rebuilding the Royals’ minor league system. Featuring draftees such as Mike Moustakas, Eric Hosmer, Wil Myers, Mike Montgomery, John Lamb, and Danny Duffy, among others, the Royals’ farm system is almost universally considered to be the best in baseball at the moment. While Moore’s record in trades and free agent signings has been pretty horrible taken as a whole, the same can be said of Giants GM Brian Sabean, leading Rany Jazayerli, Dean of Royals bloggers, to tweet (following a related blog post)

If Brian Sabean gets a ring, there’s no way on God’s Green Earth you can convince me that Dayton Moore can’t do the same thing.

It is an interesting thought, and one that isn’t unique to Jazayerli. Others have seen in the Giants and/or Rangers a draft-based “plan” that other less-affluent (i.e., not the Yankees or Red Sox) franchises can embrace. No one denies, of course, that it is necessary for almost every team to draft and develop well. The question is how far doing so is sufficient for a team to compete. It would be silly to argue that a team can compete solely with players they originally drafted or signed — neither Jazayerli nor anyone else is making that claim. But just how much did the World Series teams need to “supplement” their excellent drafts by way of free agency and trades? Rather than focusing on the Royals’ situation, Rany’s discussions, specifically (see also the recent excellent posts at Royals Review), or whether the current World Series teams got “lucky” in their acquisitions or not, in lieu of a detailed study I want to take a cursory look at the provenance of some of the more significant contributors to the 2010 success of the Giants and Rangers. Today’s post will focus on the Giants; the follow-up will deal with the Rangers.

The Giants’ recent drafting prowess is obvious from a look at their World Series starting rotation. Staff ace Tim Lincecum put up 5.1 WAR in a “down” year for him; Matt Cain, 4.0; Jonathan Sanchez, 2.6; and Madison Bumgarner, 2.0 (in only 111 innings). Closer Brian Wilson is also a Giants draftee, and had 2.7 WAR this season. That’s more than 16 wins from pitchers the Giants drafted and who are all still under club control. Among the position players, catcher Buster Posey stands out as another budding superstar, accumulating 3.9 WAR in less than 500 plate appearances this season.

However, it is clear that the Giants, who just barely made the playoffs, couldn’t have done so without some surprising performances from other kinds of acquisitions. As good as Posey was in a partial season (for reasons totally not related to service time), the 2010 Giants’ two most valuable players according to FanGraphs WAR were 2009 scrap-heap signing Andres Torres (6.0 WAR) and first baseman and second-choice first-baseman Aubrey Huff (5.7 WAR, imagine if first-choice Adam LaRoche had accepted the Giants’s offer). Juan Uribe, brought back as a bench player, was worth about three wins, as was Freddy Sanchez, and Pat Burrell was as valuable as Sanchez in just over half a season in San Francisco after being fished out of Tampa Bay’s dumpster. That’s about 20 wins from veterans signed for far below what they were worth. Indeed, even if, for example, Huff had been “only” a 3 WAR player this season, that would have been a great deal for the Giants, but they wouldn’t have made the playoffs.

The point isn’t whether or not the Giants “knew” these players could do this or that they got “lucky,” but that, even with their excellent recent drafting, they required significant contributions from other players to make it into the playoffs. The next post will take a look at the Rangers and make some concluding reflections.

I’ve made some changes to the player pages:

– The change to the header allowed us to remove the sidebar and increase the width of the stat boxes. I personally think they look less cramped and now there’s room to add more.

– MLB Trade Rumors links specific to that player are now available on each page.

– Facebook like and a universal “share” button is now available on all the player pages.

– We’re now getting all the latest RotoWire news for every player and no longer just a small sampling of players.

I think that’s about it, for now. There may be some future changes depending on how these are received.

MLB has a crazy new feature that let’s you view the entire crowd of every World Series game and two games from each of the other playoff series. During the game a single camera takes hundreds of photographs of the crowd, which are spliced together into a gigantic panoramic photograph. The photograph is of such high resolution that you can zoom in to see individual faces. It is linked up with Facebook so that you can tag yourself or your friend. It is called TagOramic. This is the full picture from last night’s game.

The boxes indicate how many fans in each section have been tagged. Those tagged have a blue indicator, but only their friends can see who they are. If you are friends with a person the indicator is red, and, if it is you, yellow. Here is a picture zoomed in on a section of the crowd.

Pretty good resolution, MLB gives the details on the photograph:

Panoramic image from the third inning of Game 4 of the World Series at Rangers Ballpark in Arlington, TX. The image is made up of 360 photos (30 across by 12 down) stitched together taken over a 19-minute span. The final hi-res file is 83,287 X 19,158 pixels or 1,596 megapixels. Photos by David Bergman.

I imagine it would be fun to see myself or a friend captured at some random point in the game, but none of my friends have been to any of the games. Even so it is cool to scan around the field and find people mid-hotdog bite, cheering, talking on their phone (it is surprising how many people are talking on or checking their phone), or whatever.

It does feel strange, and a little voyeuristic, to see this one moment in time for 50 thousand-ish people. From a privacy perspective I guess going to a game leaves you open to being projected on the jumbotron, or even being on the TV broadcast, and this isn’t any more of an invasion.

I was on the lookout for a particular group of fans at the game last night and found them in the first row on the first-base line.

Anyway a pretty cool application of high-resolution photography, photo-splicing software and Facebook tagging.

During the regular season the Rangers had one of the most effective bullpens in the American League. The relief unit managed a 3.38 ERA, second in the league by a mere 0.05. They also worked more innings than any other team’s bullpen, 503.2 innings. That looked like a major advantage heading into the postseason, but it has ended up as one of the team’s vulnerabilities. In each round the bullpen has cost them games.

ALDS

The bullpen didn’t factor much into the first two games of the series, since the Rangers’ offense staked the team to a lead while its top two pitchers, Cliff Lee and C.J. Wilson, absolutely dominated Tampa Bay. Game 3 was a much closer affair, with the Rangers leading 201 heading into the eighth inning. Darren Oliver, who had pitched a scoreless seventh inning, started the frame by striking out Sean Rodriguez. From there it was all downhill.

Dan Johnson started the rally with a double and was singled home a batter later by Carlos Pena. Darren O’Day then came in to strike out one batter before Ron Washington went to his closer, Neftali Feliz, to keep the game tied. He responded by walking the first batter he faced. One more single allowed the Rays to take the lead. Feliz then allowed a home run to lead off the ninth; his replacement, Dustin Nippert, allowed another one to make the score 6-2.

In Game 4 Derek Holland allowed two runs in four innings. Under normal circumstances this might not be bad, but it did allow Tampa Bay to increase the lead from 3-0 to 5-0. Thankfully for the Rangers, Lee made the bullpen a moot point in Game 5.

ALCS

The Rangers’ bullpen melted down just once in the ALCS, but it was their gravest offense of the postseason. It actually started with the starter, Wilson, who allowed the first two runners to reach base in the eighth inning of Game 1. That made the score 5-2. Oliver then entered the game and issued walks to both hitters he faced. Washington had seen enough and went to O’Day, who gave up a two-RBI single to make the score 5-4.

With the lefty Robinson Cano coming up Washington opted for a left hander, despite Cano’s success against left-handed pitching (.368 wOBA on the season, .348 for his career). Cano responded with a game-tying single. That led Washington to make yet another change, this time bringing in Holland to face the lefty-mashing Marcus Thames. Thames broke his bat, but the ball went over the infield and allow the go-ahead run to score.

The Rangers’ WE heading into the inning was 95.9 percent. When Wilson left it was still at 86.5 percent. After Thames’s single it was just 17.9 percent, and at the end of the inning it was 32.5 percent. The bullpen did everything in giving away that game.

World Series

Lee caused the greatest damage in Game 1, allowing seven runs (six earned) in 4.2 innings. But the bullpen put this one away. O’Day started by alllowing a run in relief of Lee. Mark Lowe allowed the death blow, though, by allowing three runs in the eighth inning. The Rangers responded with three of their own in the ninth, but by that point it didn’t matter.

In Game 2 the bullpen implosion was far worse. Wilson, despite a blister issue, managed to keep his team in the game through six innings, allowing just one run. A run in the seventh made that 2-0 Giants, but the Rangers still had a chance. That is, until the eighth inning, when Holland threw 12 of his 13 pitches for balls, resulting in three straight walks. Then, for some reason, with his team down just three runs, Washington again went to Lowe, who didn’t record a single out before letting the game get away. Michael Kirkman finished it off by allowing three more runs on a triple and a double.

Take away one of a team’s strengths and chances are things are going to go wrong for them. The Rangers have suffered because of their bullpen this postseason. While their offense hasn’t been stellar, their starters have for the most part kept them in games. Without that bullpen that pitched so well during the season, they haven’t been able to compensate. And so they will once again need Cliff Lee to eat as many innings as possible. And once he’s out, it’s time Washington handed the ball to Feliz.

In September, Andres Torres hit .164/.188/.328 in 69 plate appearances, finishing with his worst month of the season. In the NLDS, Torres hit .125/.176/.125, striking out six times in 18 trips to the plate. After going 1 for 9 with five strikeouts in the first two games of the NLCS (four of those coming in a brutal Game 2), he took a seat on the bench. He looked awful at the plate, and everyone began to talk about how he had finally been exposed.

His line since then? .414/.469/.655 in 32 trips to the plate. He’s struck out just six times, and he leads the Giants in extra base hits during the World Series. His production is one of the main reasons they’re a game away from being World Champions. Two weeks ago, Torres was being written off as a guy who couldn’t hack it. Today, he’s in the running for World Series MVP.

His postseason is a classic case study in confirmation bias. Torres’ odd development path, where he went from being a scrub to a quality player at age 31, bred skepticism over his performance. When he slugged .533 last year, well, it was just 170 plate appearances. When he kept hitting this year, there were no easy explanations, but there was still an expectation that it wouldn’t last forever. He had been a Triple-A outfielder for nearly a decade, and that made it hard to accept that he may have actually become a good hitter late in his career.

So, when Torres struggled in September, then again in the first round of the playoffs, and finally capped it off by looking hopeless against Roy Halladay and Roy Oswalt to start the NLCS, we had six weeks of recent performance where the results matched our preconceived notions – Andres Torres had reverted back into being Andres Torres. We didn’t know how to explain what he had done in the couple of years prior, but the last six weeks, well, that was the real Torres, the one we had been expecting all along. The one who couldn’t hit.

The only problem is that as soon as he was written off, he caught fire. Torres began to disprove the narrative as quickly as it gained momentum. Had the Giants bought into recent performance and kept Torres on the bench, they wouldn’t have champagne on ice tonight, just in case they have some celebrating to do. They had to ignore the bias that comes at us all when something we expect actually happens and realize that a larger view of the situation needs to be taken.

Overcoming confirmation bias is a critical part of good decision making. Fulfilled expectations are often the worst thing that can happen to someone’s analytical process, as we get an overinflated sense of the value of what we believe is going to happen. Torres wasn’t exposed as a guy who couldn’t hit in the playoffs – he was simply in a slump.

If Pat Burrell launches a couple of home runs tonight, don’t be shocked. If the Rangers win three games in a row and steal this series back from the Giants, it won’t be the biggest surprise in sports. In reality, the narratives that have been shaped by recent postseason performances – Burrell can’t hit, the Rangers can’t win in San Francisco, the Phillies are unbeatable, etc… – are often just not a reflection of reality, but instead just a big collection of confirmation biases.

I’m back in chilly Chicago after a warm week in Arizona. My final games were on Wednesday and Thursday, but I’m just now able to get to publishing my notes. As far as my experience in the AFL goes, it’s something that I think all minor league fans should make an effort to experience. The atmosphere is as pure as professional baseball gets, with a level of competition higher than you’ll get at your local minor league park. The league is chock full of future middle relievers and tweener outfielders, but Major League talent nonetheless. This is the rest of what’s found from the pitchers in my notebook. Notes on hitters will be found the rest of the week, as I have more extensive thoughts on them.

Tucked in my thoughts last week on Bryce Harper was a mention about his at-bat against Cubs pitching prospect Chris Carpenter. If you were astute, you’d notice that I mentioned Carpenter touched 98 mph during that at-bat, and if you were really astute, you’d pair that with Jason Grey’s recent profile of the reliever. My favorite quote from that piece came from Carpenter himself:

“If they tell you that you have one or two innings to throw, you hold nothing back,” Carpenter said. “I’m going out trying to throw as hard as I can.”

The quote is something you aren’t going to find a pitcher often admit, and, frankly, probably not something you want to hear from a guy with a long medical injury. But I can’t help but kind of like the bulldog nature that quote suggests, and if you read his other quotes from Grey’s article, he’s not bereft of self-awareness. If the Cubs tab him a reliever, he knows it will be on the heels of his high velocity. If the Cubs return him to the starting rotation, he realizes that he’ll have to develop consistency with his breaking ball and change-up. But it seems clear to me that reliever is the only route that makes sense, and it seems clear that Carpenter is destined for success in that role.

There are plenty of negatives on Carpenter’s scouting report. His fastball command is a bit of a mess, easily blamed on a delivery ripe with flaws and inconsistency. Carpenter has a long delivery in which he turns his back to the plate, and his arm slot and landing point can be different each time. His change up, as you might guess, is a mess more often than it isn’t. But, then again, I saw him hit 96-98 mph, with one radar gun having him touch 100 mph. His 85 mph power curveball was a plus pitch against Scottsdale, and certainly projects to be that at the Major League level. It’s elite closer stuff, and it’s not that far away. As if my belief that the Cubs need to trade Carlos Marmol needed to get any stronger…

The other high talent that I saw in the Arizona Fall League was Manny Banuelos, in his not excellent last start in the AFL: October 28, allowing 7 hits and three earned runs in three innings against Peoria. Clearly, he didn’t live up to expectations, and it shouldn’t surprise that neither did his velocity. After those summer reports of the 5-foot-11 lefty hitting 97 mph, my hopes were high driving into Peoria that day. But Banuelos looks worn out, pitching 90-92 mph, touching 93 a couple times. I agree with what Keith Law wrote in his report, saying: “the only concern I’d have off this look was that hitters did square up his fastball when he came toward the middle of the zone, as the pitch has some downhill plane but not much lateral movement.”

Other than the straight, somewhat mediocre fastball, there really is a lot to like about Banuelos. Despite the results, his curveball is actually really good, better than advertised. It had good, hard break, and that day was even better than his highly acclaimed change, which was good, mind you, but more average-to-plus than the 65 pitch that has been hyped elsewhere. He also is a really strong southpaw given his slight build, and should have no problem handling big innings jumps each of the next two seasons. I’m not sure the velocity reports are true, or at least sustainable, but it doesn’t mean he’s not still the best of the Yankees pitching prospects.

Quick Notes: I wasn’t sitting in the right section to get updated velocities on Justin De Fratus, but he looked like the best of the middle relievers that I saw in Arizona. It won’t be long before he’s ready to contribute in Philadelphia; in fact, he might already be ready … Carson Cistulli favorite Daryl Thompson isn’t going to be his next Colby Lewis, as he doesn’t have a third pitch, and he tries to throw four. He’s got a small build, but he’s an aggressive 91-94 mph, with a slider that currently resides between 82-84 mph. If he’s anything, it’s a middle reliever …. I remember when Jason Stoffel was the best member of a University of Arizona bullpen that included Ryan Perry and Daniel Schlereth. He no longer is close. Stoffel is heavier than he was back then, and his velocity is a little less, at 91-93 mph. But the worst news is that his slider is below average now, 79-81 mph, and he’s prone to hanging it. At this point, he certainly doesn’t look likely to make it 3-for-3 from that Wildcats bullpen to make the Major Leagues.