Not sure it's so simple. PITCHf/x is based on a model, and there's a chance it may miss something here. Let's say Sanabia achieved the same movement but with different spin/tumbling action?

I suppose the sample size could make age statistically significant, but it's just not practically significant here, which you've all resolved. One additional point about how the variance is important: The population may be large on the whole, but teams have to care much more about applying it back to an individual. If you were to choose between two pitchers based on expected differences (of five points of BABIP), wouldn't you expect a ton of overlap in observed BABIP? And I would venture that there'd still be a ton of overlap looking at population means for an entire staff of young vs. old. So that's why I don't see a meaningful trend here, along with Ben. That said, it could be worth testing whether controlling for other factors make age a more significant predictor. (When is it not?)

At most five points of BABIP, and with enough variance to look like there's no meaningful trend in this basic look.

Thanks, Tom. Do you mind explaining why the properties would be different? Would it be because of how TAv is rescaled? And I do see that there is a slight bit more skewness to ERA, though the scale is also much larger, enhancing very small differences. Either way, if the distributions are barely skewed, I'd still love mean + s.d. as a more compact way to deliver the info in a spreadsheet.

Colin, where would we see this? I just checked four players: Stanton, Posey, Castro, Votto, and Kemp, and each seems to show a symmetric distribution (comparing TAv for 10-50-90). Not that I'm bothered by a symmetric distribution—though I do understand the Bayesian reasons you wouldn't have them. With how they look now, I'd definitely use the s.d.'s if you can generate them.

Super 2 eligibility only has anything to do with service time. You must be thinking of the old Elias free-agent arb rankings.

Absolutely love that you went out to the ballpark to grab those photos. A+.

One question I haven't solved: We'd all agree that judging the process instead of the results is most important. And a good manager hires people "smarter" than him, including, but not limited to, a position like an analyst. But if that manager has to resist judging the results, and he's unable to fully grasp the methods being used, how is he to evaluate the researcher? Of course, I'm really limiting things to the quality of research itself, not communication or interpersonal abilities.

More Harry in 2013!

Plus plus article (including format).

Thanks, Joe. Really appreciate the answer. I bet it's not that often and still love the work. I can't imagine Jason can read many prospect comments, and I wouldn't expect him to have to take on more than he does. I guess in this whole discussion, I'd like to say that for me, the quality/accuracy of player comments is most important. More so than PECOTA or the essays. Interesting to note that an MLB.com article says Martin *had* moved away from his curve but that the Phillies had him mix it back in last August. So perhaps it was just a case of working off old information here.

Good book as always, but a few of the comments have gotten me a bit concerned about how much I can trust them. Take Ethan Martin (Phillies): Just watched him pitch two innings on TV today, and he threw a great fastball, slider, and curve. But I then take a look at his player comment, and it says that he's ditched the curve for a slider. (If this is not what was meant, it was heavily implied.) From what I saw today, he's still very much using it, and the Phillies Top 10 also lists it as a plus pitch. I know there's a lot of incomplete information with minor leaguers (and I'm not expecting to agree with all opinions), but this one for a fairly major prospect surprised me. I hope that the increased prospect coverage can cover some of these holes next year.

I'll echo this. Bought last year's at the promotional price. Enjoyed it for the most part, though hoping for some aesthetic improvements this year.

Are you sure about the throwing motion point? A pitcher who throws a two-seamer from a 3/4 arm slot will get more diagonal fade to the arm side as far as I'm aware. I'm thinking of Brandon Webb here. Over-the-top should produce relatively more lateral movement (with the drop being due to gravity instead of arm angle).

I think your biggest issue is that stepwise regression can find some pretty spurious correlations—I'd be a big proponent of using a validation data set here.

I'd love some diagnostics as well. Not a fan of stepwise regression personally. tbwhite's suggestion is a good place to start by showing risk/actual by quantiles for your validation data.

Maybe there's some disconnect between the percentile forecasts and the overall one listed at top: The latter shows 1.8 WARP, while the weighted mean is 0.1 WARP. I'm guessing it may have something to do with FRAA. After all, the main projection shows a .235 TAv in 588 PAs, while the weighted mean is .240 in 550 PAs—opposite direction suggested by the WARP.

Great to have this on the depth charts and player cards, so thank you! But first odd thing I've caught: Darwin Barney's projected -3 to -2 FRAA at 2B in the percentile forecasts. His only negative FRAA was in his first 44 games in low A ball. Is there an error here, or counterintuitive mechanic to the fielding projections?

Oustanding back-and-forth argument in this one. I thought the rebuttals from the player side really won this, particularly the CF argument and Jose Bautista career comp. That said, the club poked a big hole in the Ethier comp and really dented that the player's case. Unsurprising to me that the votes are so close in this one—it was the first time I split mine.

I think the player's case was stretched too thin. I'd have limited the argument to a much smaller handful of (favorable) comparables. For instance, the Win-Loss comparables should have never been presented if the average salary came out on the team's side. I'd try W-L% instead, if at all. The comp to Jose Contreras seemed pretty useful but was diluted by the rest.

Consider the numbers he threw out there—he's giving his personal break-even point. That's perfectly fair since everyone obviously has some trade they'd make of precision for timing. (I'd actually be interested to know what the range of them may be.)

This was outstandingly thorough. Avoided making any firm conclusions while illustrating a pretty robust picture. I feel like I learned three articles' worth here.

Really, it just needed "dusty" to not be capitalized to work better. Clever headline though.

YOU ARE 110 PERCENT WRONG ABOUT WHATEVER IT IS YOU ARE TRYING TO SAY.

Does this factor in position? A shortstop-to-shortstop comparison would be better.

It's interesting to note that WGN and Comacast SportsNet in Chicago have different producers and production teams, even though the broadcasters are the same. This definitely affects how the game "looks." So which station was carrying the White Sox game that day?

The pro scouting department entails coverage of all major and minor league players, including one's own team. They're crucial to the player acquisition process (trades, wavier claims, etc.) for every player who is not an amateur.

They did have to clear waivers, but since the waiver period is something like 48 hours, players like Loney and Punto had probably already cleared. Would you really be willing to put in a claim and risk having to take Crawford or Beckett? Plus, I just don't see how making the threat would get you a prospect -- only one team is awarded the waiver claim, so if you let a player by, you're out of the discussion. If you put in a claim and was awarded it, then congrats, you ruined the deal already anyway.

Love all of these. Would find it really helpful to participate in a "quiz" piece, where we try to take the knowledge we've learned and pick out certain aspects of a pitcher's delivery. I'd like to see if I'm on the right track when not knowing the answer first.

Fair point that some of them are quite simple, but I think an effort is being made to get more comments in here because people have demanded them. A lot seems to be from MLB or advance scouts who are much more outcome-oriented anyway -- yes, we can come to similar conclusions as well. But, I get value out of the comment about Ivan Nova that suggests he's hanging his breaking ball too much. That's something hard to know without paying a lot of attention -- hard to evaluate in PITCHf/x, too -- and it can be investigated further. For me, if I learn one thing from an article, it's a success. That's my one (at least in that section).

I know you're just being flippant, but let's not criticize the scouting industry's value (or even a single, anonymous scout) based on a handful of two-sentence comments. Better to learn from them than assume that you could do their jobs.

I do understand what you're saying, and I appreciate your perspective because it's really founded in having equal expectations for everyone. That's valid and admirable. I just don't think it works in the general case because I do believe cultural roots play a large part in shaping people -- not forever and not universally, just in shades of gray. We're talking about a young player here, and I think Castro has shown a lot of aptitude as a 20-22 year old. He doesn't look like he's done adjusting even if he's not where you expect him to be yet. Perhaps our disagreement comes mostly from evaluating from different perspectives? Maybe you're looking from a manager's view of holding players to high expectations, while SaberTJ and I are taking a long-term view of what we expect him to become through an average expectation of development. Those are not the same thing. I don't want to belabor the country thing too much because I think you just view/value things differently than I do, but let's compare the types of players produced out of Venezuela and the Dominican Republic. You'll rarely see catchers come out of the Dominican -- the all-time major league list is very small -- whereas they are more abundant and successful out of Venezuela. One potential reason: There's a better education system in Venezuela, and many of these kids play actual baseball games as youth. In the Dominican, they have no such structure until they are signed by a team. The catcher thing is just a symptom, but the education issue is going to be something that impacts players no matter if you think every player deserves equal expectations.

In my opinion, it's relevant insofar as how one may project his future maturity. There's a gap with where he is now and where he needs to be, but the expectations of a player with Castro's background are different than those of many U.S.-born players, especially as they are still young compared to their peers. Personally, I think you're making the mistake of lumping every foreign player together as part of Castro's peer group, including a 26-year-old Cuban. Time will tell, but isn't this a case where you have to give weight to the (pro/con) judgement of the people who know the individual much, much, much better than anyone on the outside?

Perhaps this would be viewed differently if Crane agreed to not sell the team sooner than 10 years or something like that. The main concern has seemed to be whether owners were mortgaging the team's future when they were only going to be around another couple years.

I actually noticed this with the AL Central. On July 18, Minnesota is shown to have a 34 percent chance, while Cleveland had flat-lined. I had to actually go back and check if I somehow missed Minnesota being above Cleveland in the standings that late in the season, but it looks like they're just reversed.

Right, that's where I realized there may be disagreement on the rebuild timeframe. I'd start it with this past offseason considering the substantial turnover throughout the front office. You may be talking from a purely financial standpoint, but I think "rebuilding" is more of a baseball-side term. The Cubs' payroll has dipped, though they have said they are merely investing the dollars elsewhere within the baseball budget -- Dominican academy, scouts, technology, amateur signings, etc. The Astros may be doing some of the same things. Regardless, as a relative proportion of revenues, the Cubs' payroll decline right now is at least in the same direction as that of the Astros'. The recent records are similar. And ultimately, I'm still not convinced this is a bad thing, just as in Chicago. It needs to be done. And done well.

Maury, I can't tell if you're letting your issues with Jim Crane get in the way here. It certainly seems at odds with some of the praise being thrown around for the Astros' commitment to rebuilding with Jeff Luhnow. How does this differ from what's happening (belatedly) in Chicago? And it's surprising to me for a sports-business reporter to be implying that something is amiss when payroll won't match 2013 revenue. There's of course a value judgement there that everyone can make for him/herself, but ultimately 1) we expect that all owners are somewhat profit maximizers and 2) what's the point of an artificially high payroll versus rebuilding thoroughly? Perhaps we just disagree when the rebuild started -- I'd say you start evaluating that from now, no matter if a few pieces were moved last year.

You always have to take anonymous comments with a grain of salt, but you have to trust John, who has a lot of access, experience, and sources, to filter the comments with his best judgement. He likely feels that they give shades of the truth that wouldn't be revealed if not for granting anonymity. Any anonymous quote could have the issue you cite -- that includes the scout/executive comments seen in several columns across BP. It's an industry where few people will say anything of interest with their names next to it. Give it the weight you wish, but I'd rather make that determination myself.

Underappreciated bonus of having the list published here, even if it contains different content than the ebook: Kevin's been answering questions. Lots of value in that.

My first thought when people jump to conclusions of who "won" a trade is that the person doesn't really grasp the concept of market exchanges. If both parties are better off -- as they clearly both believe -- then they both win. Start from that point and go from there.

Because when you're negotiating multiple contracts simultaneously, you're not really concerned about getting it all to a round number. It's easy to look at it now and ask why the Blue Jays didn't up their existing offer to a player by a buck or 11 or 341 once they knew they had the money, but none of those amounts would be noticeable/significant enough for an agent or the team to worry about it in active negotiations. It's just so ... trivial.

Very useful to me -- it's more actionable.

Fantastic work on the visualization here, trumping most other tools I've seen. Easy to explore, quick, responsive; plus, we know it's built on great data. Love it. Can't praise it enough.

I think it still has an incredible societal perspective too. I found the descriptions of U.S. society during the Vietnam War resonated with me more than three decades later when I was reading it during the start of the wars in Afghanistan and Iraq.

They clearly stated they wanted to hear from readers via email. Be fair.

I'm somewhat OK with separate leaderboards in theory because of the DH and it's impact on some stats. But the cons outweigh the pros once you consider midseason trades.

Phenomenal piece.

Looks like there's a bug in CDD -- sort by AGL, which I assume is Average Games Lost. Burnett, Crawford and Howard all have totals (and thus return dates) that are way off according to their comments.

Glad to once again hear a commitment to an overhauled site format. Can't come soon enough. In theory I appreciate all the new writers -- what I really struggle with is keeping up with all the contributions, particularly from those unfamiliar to me.

Very exciting news!

Hope you don't mind me asking ... is there a new ETA for the iOS app? First-minute purchase for me.

I should also add that the difference in the save totals the last two years can be traced to the five extra blown saves he had in 2011.

You're a bit overconfident. He had 38 two years ago, 34 last year. You may be right that 38 seems like it's on the high end, but it's still well within the realm of possibility.

Agreed. It was visible in the Upside projections earlier this year as well, though I think that has been fixed.

Does the Cubs' injury situation factor in Andrew Cashner?

Great stuff. I'd suggest ranking players who have a negative WARP as the worst for $/WARP. Right now, if I rank left fielders for which players are being paid the most $ per WARP, Alfonso Soriano doesn't appear among the worst. I thought he may be missing at first.

Thanks for the quick summary. I would imagine this was the case just based off Marcel not being built to project breakouts or collapses. I guess I am most interested in how the SD of PECOTA compares to previous versions as well as projection systems outside of Marcel. Your analysis has made me question whether i care about using the projection system with the lowest RSME -- I may be willing to sacrifice a certain amount of error if the SD is more representative of reality, taking into consideration the difference between observed variance and true talent variance. Other tests (like heads-up) can be used, of course, to meet those goals.

Seriously. Where Mike says in these comments that he has learned a lot from Colin in analysis, I think Colin would benefit a lot from Mike in terms of communication. His tone is diminishing some of his own points here and drawing some negative attention to BP as it still has some work to do fixing its own metrics. By the way, re-reading Colin's point about SIERA's smaller standard deviation makes me wonder if the heavily regressed PECOTA projections this year are going to benefit in terms of RMSE relative to previous projections that had a higher SD. So perhaps the same standardization will benefit the comparison between projection systems after this year.

I too have been perplexed by the loss of WXRL. I found it useful for some relief-pitching analysis because of the lineup adjustments.

I also value open disagreements, and many people will find that they are valuable learning experiences. Where I struggle is with the strength of your criticisms -- saying something is flat wrong -- weighed against the struggles to come up with sound, constructive alternatives. The former is much easier than the latter, and more balance from you would gain you, from my perspective, more patience in rolling out such ambitious projects.

I'd like to request more consistent reports on NL Central, like for the other divisions, please.

Ah, thanks. I'd suggest adding that, in fairly noticeable type, on the depth charts page. If it's there, I missed it reading over again.

Thanks, Sky. But be aware that the current depth charts are showing just 146- to 148-game schedules for each team. Don't let the weather postponements so far get you that far down.

Chi-ca-go! *clap-clap-clap*

At a glance, the rate stats do seem very similar, but the WARP are way off. Take Chad Billingsley: Same IP, H, BB, K, HR, but the depth chart (and player card "beta value" graph) lists 2.9 WARP. The weighted mean lists 4.3.

Completely agree that a site redesign will help by categorizing the content. I both love all the new content and feel a bit overwhelmed by how much there is, so I can sympathize with both views in this thread.

It's just that you seem to have a lot of anger toward dorks at Apple, Apple itself, fanboys, Will and people who disagree with you.

Wow, you come off as though everyone -- fanboys and all -- is out to get you. I think we appreciate your perspective as a developer who has had bad experiences, but there are many of us who also like how the app store runs for end users. And we're willing to pay for that experience. Some of the safeguards and approvals might be inconsistent, illogical or overkill, but the successes do carry significant weight.