 |
US Presswire photo Several projections show a drop-off in production from Ian Desmond this year. |
Two of the more noteworthy folks who produce annual projections are Bill James (the grandfather of sabermetrics) and Dan Szymborski (who created the ZiPS projections that are now featured on Fangraphs.com). Both utilize a player's performance over several past seasons, plus historical trends, to project how his numbers will look the following season.
They're by no means perfect -- last winter, James projected Ian Desmond to hit .268 with a .394 slugging percentage; he wound up hitting .292 and slugging .511 -- but they're a fun baseline to begin imagining how certain players will face in the upcoming season.
So let's take a look at what James and Szymborski say about
members of the Nationals' lineup and bench for 2013. We'll look at the pitching staff in a future blog post...
BRYCE HARPER PA HR RBI AVG OBP SLG
2012 Actual 597 22 59 .270 .340 .477
2013 James 659 24 65 .272 .347 .476
2013 ZiPS 641 26 70 .274 .346 .486
COMMENT: Both projections seem awfully conservative for a player just about everybody in baseball expects to make a huge leap at age 20.
RYAN ZIMMERMAN PA HR RBI AVG OBP SLG
2012 Actual 641 25 95 .282 .346 .478
2013 James 659 25 93 .287 .359 .486
2013 ZiPS 607 23 87 .280 .348 .471
COMMENT: At this point in his career, we know what Zimmerman is going to produce. The only question is how many games he plays.
IAN DESMOND PA HR RBI AVG OBP SLG
2012 Actual 547 25 73 .292 .335 .511
2013 James 595 18 68 .279 .326 .445
2013 ZiPS 611 18 70 .269 .312 .432
COMMENT: Obviously, there's some skepticism about Desmond's ability to pick up where he left off last season, though neither projects him to revert all the way back to his 2011 numbers.
ADAM LaROCHE PA HR RBI AVG OBP SLG
2012 Actual 647 33 100 .271 .343 .510
2013 James 592 26 87 .256 .334 .471
2013 ZiPS 478 19 72 .249 .322 .445
COMMENT: It's probably correct to assume some regression from LaRoche in his age-33 season, but does anyone honestly believe he'll only hit .249 with 19 homers as ZiPS suggests?
MICHAEL MORSE PA HR RBI AVG OBP SLG
2012 Actual 430 18 62 .291 .321 .470
2013 James 515 23 80 .295 .342 .501
2013 ZiPS 481 21 73 .272 .320 .467
COMMENT: If he can stay healthy, there's every reason to believe Morse will put up numbers that match or exceed James' projection.
JAYSON WERTH PA HR RBI AVG OBP SLG
2012 Actual 344 5 31 .300 .387 .440
2013 James 636 20 74 .267 .367 .440
2013 ZiPS 501 15 53 .253 .345 .422
COMMENT: James projects Werth to produce pretty close to his career norms; ZiPS says he'll more closely resemble the disappointing 2011 version.
DANNY ESPINOSA PA HR RBI AVG OBP SLG
2012 Actual 658 17 56 .247 .315 .402
2013 James 602 21 66 .253 .327 .434
2013 ZiPS 653 20 66 .235 .308 .403
COMMENT: The Nationals hope Espinosa can make the kind of strides in his third big-league season that Desmond did in his. James and ZiPS see more of the same in 2013.
DENARD SPAN PA HR RBI AVG OBP SLG
2012 Actual 668 4 41 .283 .342 .395
2013 James 593 5 50 .281 .350 .379
2013 ZiPS 568 4 42 .273 .330 .380
COMMENT: Neither projection deviates much from Span's 2012 numbers, but James thinks he'll perform slightly better than ZiPS does.
KURT SUZUKI PA HR RBI AVG OBP SLG
2012 Actual 442 6 43 .235 .276 .328
2013 James 424 8 47 .247 .306 .360
2013 ZiPS 509 12 63 .263 .308 .400
COMMENT: Suzuki finished the season strong. ZiPS seems to think that was the sign of an upward trend; James seems to think it was a mere blip.
WILSON RAMOS PA HR RBI AVG OBP SLG
2012 Actual 96 3 10 .265 .354 .398
2013 James 446 13 50 .269 .324 .426
2013 ZiPS 311 9 34 .260 .314 .411
COMMENT: Both projection methods believe Ramos can make a full recovery from a torn ACL and produce like he did as a rookie. The question is how much playing time he'll get.
ROGER BERNADINA PA HR RBI AVG OBP SLG
2012 Actual 261 5 25 .291 .372 .405
2013 James 348 8 32 .272 .343 .408
2013 ZiPS 369 9 31 .248 .317 .385
COMMENT: Bernadina figures to be the first guy off the bench if a starting outfielder gets hurt. These projections suggest he's still best off when he doesn't play on a daily basis.
TYLER MOORE PA HR RBI AVG OBP SLG
2012 Actual 171 10 29 .263 .327 .513
2013 James 250 13 47 .259 .315 .491
2013 ZiPS 479 20 71 .233 .280 .433
COMMENT: Moore would put up some big power numbers if he played regularly, but his batting average and OBP would probably suffer for it.
STEVE LOMBARDOZZI PA HR RBI AVG OBP SLG
2012 Actual 416 3 27 .273 .317 .354
2013 James 348 4 28 .293 .339 .399
2013 ZiPS 584 5 42 .267 .316 .372
COMMENT: As is the case with Moore, the projections suggest Lombardozzi wouldn't be as productive as an everyday player at this early stage of his career.
150 comments:
I believe these numbers are based on just individual stats from previous years. I would like to see them adjusted for Nats current lineup. Basically two lineups, one with Morse and one with ALR as our 1B.
I also expect desi to regress a bit. I expect avg and hr to take a hit. his HR/FB% was significantly higher than league average and so was his BABIP. But if his LD and GB percentage keep trending in opposite directions, i can see more HRs over the visitors bullpen for him.
I have no doubt in RZ's abilities. Like mark said, it is matter of health only. if he can be healthy all season long, I expect MVP numbers from him.
for Werth, I don't like ZiPs numbers. I think he is better than that.
by the way, fans can also make predictions for individual players over at FG.
It's probably correct to assume some regression from LaRoche in his age-33 season, but does anyone honestly believe he'll only hit .249 with 19 homers as ZiPS suggests?
Why should we care? Whatever he does, he'll be doing it with some other team.
Ridiculous! To start with, Harper's estimates are just plumb dumb, as the kid will be much more secure with himself, not to mention the table should be set for him to drive in more runs.This is a classic case of someone drawing conclusions about a player that they saw little, if anything,with their own eyes.I really expect to see a much more grown up and mature, Bryce Harper. If ALR returns I think Harp has a good chance to lead the club in RBI's. And where is Bernadina going to get 348-369 AB's from? Desmond? Agreed, he may have a drop off,but I doubt it will be appreciable. I think in Ian's case that it's just as likely that he came into his own last year and will establish that his 2012 numbers will be closer to the norm for him.You can keep the grain of salt, these projections aren't worth one.
I think you have to respect the experience that goes into these projections. Without a recap of last year's numbers my guess is that most of the Nats exceeded their projections. I'm sure that is part of what drove some posters to demand that they go all out to win and to a degree it was a valid position.
I certainly don't see anything spectacular that would indicate a championship team. Far from it. But, I didn't see anything spectacular last season and look where they finished. Harper is our one major player. Hopefully the pitching will hold up to carry the team.
Both James and his ZiPS friend are more knowledgeable about these projections than I am.
However, these projections are probably reflecting very static metrics and don't seem to adjust well to individual player growth patterns, e.g., Desi 2011 vs. Desi 2012.
The projections that I seem to intuitively agree with most are those for ALR. Where ever they both play, I think that The Beast will out perform ALR.
So are these projections purely based on statistical measures, or do they include, as terms like "awfully conservative" and "skeptical" imply, a fudge factor for other input? Harper, for instance: Mark and Boswell, among others, expect him to do much better than these numbers, based on the history of other 20-something phenoms. Do Bill James and/or ZiPS take that sort of analysis into account to come up with these?
*20-year-old, not 20-something.
.This is a classic case of someone drawing conclusions about a player that they saw little, if anything,with their own eyes.
Which makes me wonder, are local writers, who see their own team's players regularly, better predictors of the next year's production for a given player than James or other national writers, who are trying to show and refine (as I understand it) what we can say based on the numbers alone?
Realize that when they come up with these projections, its not them (Bill James, or Syzbro...) they are inputting numbers from previous years into a computer along with some other variables, no need to get angry at the people in fact Dan Syzbro... has said he would take Harper over Trout and compare the stats no need to overreact, there are a lot of factors that can't be quantified that we all know and see.
Two thoughts, neither flattering:
(1) These projections are not derived from "experience," just the amount of some computer's memory. GIGO.
(2) They are about as reliable as Tarot cards.
You all seem to be missing the point of these projections. They essentially represent the peak of bell curve projections for each player based on computer models for similar players. There are lots of individual variables that might alter those projections if that was the intent- for example, I don't think the models can account for the fact that Ramos tore his ACL, they only know how many games he missed and can compare that to other players recovering from other players who missed time.
They are always gonna miss wildly on a bunch of players. The guys who do them will admit as much, and there's no reason to change that, since tinkering to make them "soft" projections will do more harm than good. Despite what Theophilus says, these types of projections are still more accurate and useful on average than any other system.
They're also far more useful if you do look at them in the aggregate instead of focusing on one player or one statistic.
I wonder if they make any small adjustments after ST. I doubt it given sample size but watching them can indicate whether a particular player has made any particular adjustments to their swings or approach which could lead to better stats. so yeah, overall we can not put much weight into these predictions.
For those interested, you can find a quick description of how these projections are calculated at:
http://www.fangraphs.com/library/index.php/principles/projections/
Obviously, take 'em for what you think they're worth. They are in fact better than Tarot cards, but are better at some kinds of things than others--for example, they're better at rates than at absolute #s--see for example the calculation of PA, esp. for non-starters (here, the likes of Lombo and Bernie).
Final thought--I find it a little amusing to hear commenters on an internet blog expressing all this animus towards the use of ... computers.
NatsJack said @9:12
Foretelling the supposed future of individual baseball players is for those fantasy league people.
I prefer watching the games and let them tell the actual story.
_____________________________________________
I'm not sure that I ever agreed more about a post on this board.
(1) These projections are not derived from "experience," just the amount of some computer's memory. GIGO.
Computer programs don't write themselves. Some person or persons wrote them, and used their "experience" to tell the computer how to interpret and massage the input data in order to get the output data. So "garbage in, garbage out" does not apply here. What's going into these projection programs is not garbage, it's actual statistics from the players' previous years. If what's coming out is garbage, it's because the projection program is a garbage maker. And there's some "expert" somewhere who told that program how to make garbage.
This, BTW, is the same problem that the so-called "advanced stats" have. They are only as good as the formulas that create them, and those formulas are only as good as the "expert" who developed them. People seem to think advanced stats like WAR are gospel truth, but do they even know how these numbers are calculated? Can the formula really be trusted? The closer an advanced stat is to the real stats, the simpler the formula, the more it can be trusted. Everyone knows what WHIP is. You take walks + hits and divide by innings pitched. That's a formula you can trust. But WAR, pRAA, xFIP, all that other alphabet soup, who knows what the hell that is? Why are people so willing to put their trust in it?
NatsJack in Florida said...
I take all these projections with massive amounts of salt. Foretelling the supposed future of induvidual baseball players is for those fantasy league people.
I prefer watching the games and let them tell the actual story.
January 03, 2013 9:12 AM
Its a cruel game, "What have you done for me lately". Every player starts at the same point 0.000
Very few players change what they are. Ian Desmond is one of the few and he has to show he can replicate what he did in 2012. Then you have Jayson Werth and Michael Morse who found themselves in their latter 20's.
"You are what you are until you prove otherwise."
If Espi has 20 or 21 knocks then he will strike out close to 190 times
I think the reason some people look to advanced stats is frustration with using the older stats for projections. For example, if you look at ERA, first of all, the "earned" in ERA is very arbitrary. Second, some of ERA depends on things beyond the pitcher's control, such as how good the defense is, and nature of his home park.
Hence, FIP is invented, attempting to look at the pitcher independent of his fielders. How would this pitcher do for my team, which might have different fielders than his present team? This is a legit question a GM (or fantasy player) has. But FIP has its problems. HRs can be dependent on park factors and small sample sizes. Some pitchers don't care how many players they walk as long as those player's DON'T SCORE! Ooops, we're back to "runs allowed."
I'm a fan of stats like WPA and Leverage Index. These stats run game situations through hundreds of thousands of iterations to evaluate how much a guy who strikes out in the bottom of the 7th with two men on base has cost his team. YET--neither of these two stats figures in the actual man hitting or pitching, and so, although we can recognize a high-leverage situation when we see one, we can't, in the end, predict what a given player will do. You might get closer to a prediction if you factored in the current hitter and pitcher, weather, crowd noise, importance of the game, etc. But those factors are what makes it FUN to watch the games.
My own feeling is that there is no one perfect stat or method. However, the ability of smart people to invent stats is also part of the fun of the game. Predictions were that the Oakland A's would lose 100 games in 2012. We all know they didn't, and it wasn't just a matter of "luck." The question is, is there anyway to factor in what the A's did to improve pre-season predictions? I don't know, but I bet somewhere someone is trying. :)
If they were looking for a definition of "academic exercise" to put in the dictionary, this would be it. As far as I know, the computers don't play the game, unless it's chess.
And in the end, all stats can be sabotaged by simple but unpredictable things like a flu bug in the clubhouse or a guy needed contact lenses...
I find it a little amusing to hear commenters on an Internet blog expressing all this animus towards the use of ... computers.
Well, that, and the fact that "the amount of computer memory" for algebra like this hasn't been a limiting factor since before most posters here, even us old farts, were born. This isn't any harder than, say, calculating a mortgage--easier, in fact. As several posters above point out, coming up with the observations that get expressed symbolically is the actual work.
The numbers on the backs of the cards have been a part of baseball for a very long time now. As Woody pointed out above, the problem isn't the numbers, or the predictions, it's believing in things that you don't understand. And when you believe in things that you don't understand, then you suffer.
I think Abraham Lincoln said that.
Laddie, no, they don't. But computers--and videos--are used to evaluated how a player performs compared to his peers and they are used to (attempt to) predict how he will perform in the future. Computers--and satellites--are why we don't laugh at weather forecasters like we did when I was a child.
Like.
Joe Seamhead said...
NatsJack said @9:12
Foretelling the supposed future of individual baseball players is for those fantasy league people.
I prefer watching the games and let them tell the actual story.
_____________________________________________
I'm not sure that I ever agreed more about a post on this board.
January 03, 2013 9:37 AM
Um, I still do laugh at them. :-)
NatsLady said...
Computers--and satellites--are why we don't laugh at weather forecasters like we did when I was a child.
January 03, 2013 10:01 AM
Seems several people are getting rather indignant about predictions about "our" boys being anything less than stellar next year.
I'll take these for what they are, historically-informed, statistically more-likely (but not certain) outcomes.
If these predictions pan out, the 2013 Nats will be like the 2012 Nats: a team with dominant starting pitching, with just enough offense (no 30-HR sluggers whether or not LaRoche comes back) and relief pitching to put the Nats at 90-something wins.
I'm somewhat surprised these projections are so pessimistic, particularly for Harper, Desmond, and Espinosa. They seem to make no allowance for improvement with age. I'll be very happy if Werth's power returns as James predicts.
In any event, it's going to be our pitching that will tell the tale this year. Even if the hitters produce just as projected we'll be fine if the pitching is as good as last year.
Make sure to check out the continually updated NIDO Spreadsheet. Lots of fun additions to the Glossary (h/t Gorse(?), among others).
Uh oh. I think I owe Snivius a drink!
I figured out the interface and wrote a few of them myself, 222. ;-)
NatsLady said...I'm a fan of stats like WPA and Leverage Index. These stats run game situations through hundreds of thousands of iterations to evaluate how much a guy who strikes out in the bottom of the 7th with two men on base has cost his team. YET--neither of these two stats figures in the actual man hitting or pitching, and so, although we can recognize a high-leverage situation when we see one, we can't, in the end, predict what a given player will do. You might get closer to a prediction if you factored in the current hitter and pitcher, weather, crowd noise, importance of the game, etc. But those factors are what makes it FUN to watch the games.
If a player is 1 for 7 with men on 2nd and 3rd with 2 outs and comes up again in that same situation, what do you think the outcome will be?
The odds generally are always in the favor of the pitcher as Rule #1 but statistics say each new occurrence are mutually exclusive of what previously happened so even though the odds are against that player getting a hit it's still possible and that's why they play the game and that's why people second-guess all the time based on negative outcomes.
The one factor I always consider is trending. Yes, those small sample sizes. How does Batter A do against Pitcher B. What has Batter A done lately. What does Batter A or Pitcher B do in those situations. Can Batter A look for a certain pitch that Pitcher B throws in those exact situations.
Trending or as some people call it "hot streaks" and "slumps" are really small sample sizes. Put 17 small sample sizes of 9 game stretches together and you have a full season.
Not sure I understand. Zuck wrote regarding Lombo: "As is the case with Moore, the projections suggest Lombardozzi wouldn't be as productive as an everyday player at this early stage of his career."
What in the projections showed how he'd do on an everyday basis versus part time (much less having anything to do with it being early in his career)?
Ghost Of Steve M. said...
Buster Olney has been compiling his best teams. Best outfield, best starting rotation, etc.
Nats are his best outfield in the NL and #2 in the Majors behind the Angels (Trout, Josh Hamilton).
January 02, 2013 12:26 PM
Sec222, I am calling Mark Z out that he owes me a drink.
He published this 45 minutes after I wrote my post
http://www.natsinsider.com/2013/01/where-nats-rank-by-unit.html ESPN's Buster Olney has been issuing top-10 lists the last few days, ranking what he believes are the best rotations, infields and outfields in baseball. Olney has both the Nats' rotation and outfield as the second-best units in the majors
I think as the keeper of the NIDO stats you need to make a ruling on this. LOL
Baseball for me is first and foremost a game to enjoy watching, or listening to on a radio. And yes, it is a game of stats, it always has been, but yesterday's stats are real. Tomorrow's stats and predictions are conjecture that can be fun for fans to debate, but it gets to a point that it starts turning a wonderful game of pitch, throw, and hit into a geekish, otherworldly something else that isn't exactly baseball anymore to this old man.I've read Moneyball, and I look at advanced stats, agree that they have a place, but I'm more on the same page as F.P. Santangelo when it comes to a lot of the prognosticators. For example, after a UZR was put up here last year that stated that Adam Dunn was a far better 1st baseman then Michael Morse, I now just glaze over UZR. When many predicted last spring that the Phillies would win the World Series last year I thought that they were baseball stupid. Tick-tock, tick-tock had to have been obvious to anybody actually watching the Phillies at the end of 2011. Many of them aren't watching the same game as I am. In fact, often they never actually watch the players that they draw their conclusions with their numbers crunching. To those that don't agree, I'm not saying that I'm a better fan then you, or vice verse. If it brings you joy to play with the numbers, more power to you. Like Nats Jack said, I prefer watching the games and let them tell the actual story.
In conclusion: GYFNG!!!
As a few others have mentioned, the projection program that's applied to developing players shouldn't be the same as that applied to older players. As long as the historic stats for both types of players are fed through the same program, the projections will continue to underestimate the performance of young players, like Desi, Espi and Harper, who showed improvement from early in their most recent season to later in the same season. Desi's 2nd half of 2011 was an indication that he would be better in 2012. Believing the projections rather than expecting similar improvement from Espi based on his July, August and September 2012 and Harper based on his September and October would require ignoring what your eyes tell you.
… I always knew I was conservative. (Quite a confession for an admitted 'liberal'.) But I didn't know how much until I realised I was nowhere near as optimistic about Bryce's 2013 as Mark claims most fans are. Having had a great 2012 and being only a kid just isn't enough for me to see him doing any better in 2013, except that he'll have more AB's. In the end, if he does the same, I'll be more than satisfied; if he does as Mark projects, I'll be tickled pink.
… in regards to Ian D, I'll easily agree that he'll do much better than the projectors say.
Go Nats, and get packing!
I am a betting man and I'd be willing to wager that Bryce Harper exceeds the above projections of both sources in all five statistical categories listed. I'd make the bet based on what I witnessed with my own eyes last year.
"For example, after a UZR was put up here last year that stated that Adam Dunn was a far better 1st baseman then Michael Morse, I now just glaze over UZR."
I mostly agree, Joe, but if advanced stats are used, they should be used in the way their creators intended. For example, isn't it true that UZR isn't very useful to evaluate a player who has less than 3 seasons playing a position? Morse has played 1b for less than a full season and the only time Dunn played a full season at 1b was in 2010.
Natsfan1a - you can't owe yourself a drink.
For example, after a UZR was put up here last year that stated that Adam Dunn was a far better 1st baseman then Michael Morse, I now just glaze over UZR.
Really, Joe? A single bad example, which may or may not even have been properly related and applied, and you're going to throw the whole thing out?
As for the larger question about projections--Nate Silver has a good chapter about projection systems in his new book, the Signal and the Noise. He dismisses the "scouts vs. statheads" debate that followed Moneyball. The best projection systems combine the two but doesn't dismiss either one--which some folks here some awfully ready to do.
I quickly compared the 2012 actual stats vs Zips projections for 8 position players (I left out catcher, but included Lombo among the regular position players). The projections were within about 5% for Zimm, Morse, Espinosa, and Lombo. They were 20+% off for Harper, Werth, LaRoche, and Ian.
Take out Ian (he had a great year in 2011, but it's not a "breakout" year til he repeats it), and the cases it got most wrong were the very youngest and oldest players--Harper, Werth, and ALR. In the former case, there's just not much data; in the later cases, most projection systems tend to regress older players pretty sharply. Combine that with the fact that both Werth and ALR had very bad 2011s, and it's even easier to see how Zips arrived at the poor projections.
A final thought: BA is the most unstable stat--variable from year to year, and HRs can be too. The projection systems did much better at estimating things like OBP and OPS+. So if you're going to pay any attention to the projections at all, pay most attention to those stats.
That's why I said it was a trick question. :-P
Candide said...
Natsfan1a - you can't owe yourself a drink.
January 03, 2013 10:55 AM
Natsfan1a - you can't owe yourself a drink.
Why not? MASN owes the Orioles TV fees.
In w/ NatsJack, Seamhead and natsfan1a on this. Watch the game, believe what you see, not some crap written in Fortran.
If you want to live your baseball life in nos., see if you can find an intact version of Negamco Major League Baseball on E-Bay.
And then apply an "age-regression" factor if it suits your prejudices.
Back in the '50s, '60s, even into the '70s and '80s, many players flamed out in their early 30s. Careers of many players since then have been extended by a decade or more because of (A) year-around training; (B) better training equipment and protocols; (C) better medical procedures (diagnostic and surgical); (D) better rehabilitation; (E) bigger salaries and, therefore, more incentive to extend careers at a high level; (F) better conservation of players (because of the huge investment in bonuses, salaries, etc. "Back in the day," careers like Musial's (.337 at age 37), Williams (.328 at age 39), Yaz (until age 43) were novelties. These days I have no difficulty foreseeing a good (not necessarily great) player (e.g., LaRoche, Ibanez) with an exceptional skill (e.g., LaRoche's fluid swing) being very productive at 37 or 38. If you want to complain about Jeter's "reduced range" at 40, tell me you wouldn't want him somewhere in your lineup.
Improve your lives, people; buy a ticket.
A Strong Package for Gorse Hackage! said...Natsfan1a - you can't owe yourself a drink.
Why not? MASN owes the Orioles TV fees.
Yeah, but we aren't the Orioles, Section 222 isn't Peter Angelos, and the NIDO sheet ain't MASN.
And on this, there can be no dispute.
Seriously, you deleted it? It's all in good fun, right?
natsfan1a said...
That's why I said it was a trick question. :-P
Candide said...
Natsfan1a - you can't owe yourself a drink.
January 03, 2013 10:55 AM
January 03, 2013 11:06 AM
Eh, never mind. I have work to do.
Sec222, I am calling Mark Z out that he owes me a drink.
He published this 45 minutes after I wrote my post
Ghost, you said this yesterday. Does that mean you owe yourself a drink? No, that's against the rules, as enforced by Candide's ingenious programming of the NIDO Spreadsheet!
Anyway, as I said yesterday (and with this intro I immunize myself from owing myself a drink, which I can't do anyway), Mark provided much more information and analysis about the article, and he also gave us a new post, for which, on many occasions, we all should owe him a drink. Besides, my view is that as tavern owner, he doesn't have to buy anyone a drink, though people are welcome to offer to buy him drinks (and some already have). However, a CSN sponsored event before opening day may provide the perfect opportunity to settle drink debts. Hint, hint.
The one rule I suggested at the outset that I still think makes sense is that only drink owe-ers can note that a drink is owed. I may think, for example, that Feel owes me many beverages for a variety of reasons, but only he can acknowledge a debt by adding it to the NIDO Spreadsheet.
Also, I may be wrong about this, and again, the drink owe-er will make his or her own decision, but my impression is that drink debts are most often incurred when one commenter beats another to the punch by a matter of minutes with essentially the same comment. This usually happens when both are writing comments at the same time and one happens to be a quicker typist or more concise writer and hits submit first. No one should feel it is necessary to read every previous comment for fear of repeating something that has been said hours before by someone and incurring a virtual drink debt. And no one should be deterred from expanding on, or adding to, a previous comment.
Again, NIDO is now the property of the collective. So my reaction is just my opinion. Your mileage may vary, as 1a would say. :-)
hiramhover said...
For example, after a UZR was put up here last year that stated that Adam Dunn was a far better 1st baseman then Michael Morse, I now just glaze over UZR.
Really, Joe? A single bad example, which may or may not even have been properly related and applied, and you're going to throw the whole thing out?
--------------------------------------
Define " whole thing." If by whole thing you are asking my take on UZR, then yes I am throwing that whole stat thing out for my own purposes, but not just because of that one example. I find it a flawed, subjective "stat". Do I throw all advanced statistics out? Of course not, but I they also are not where I find my joy in baseball.
UZR. Like I said yesterday looking at the top Nats CFs in UZR with Ryan Church and Nyjer Morgan at the top of the list. Its flawed. It can't judge the jumps you get on the ball but supposedly gives points for getting to tough balls. Very subjective. It also doesn't factor in your arm. You could always take an extra base on Nyjer Morgan aka noodle arm.
natsfan1a said...Seriously, you deleted it? It's all in good fun, right?
I had fun deleting it...
I left your infinite number of drinks you claim to owe Gorse, 'cuz I'd like to see THAT debt paid off...
And bon-bons is good...
Theo at 11:20, I agree, but also add in there is little place in the game today for the John Kruk mentality that,"I'm not an athlete, I'm a ballplayer." Today's game is mostly passing those guys by now, though of course there are still exceptions.
I may think, for example, that Feel owes me many beverages for a variety of reasons, but only he can acknowledge a debt by adding it to the NIDO Spreadsheet.
I don't participate in the drink-owing nonsense.
I think this is a false discordance, we're talking at cross-purposes.
Very few people who keep score well do it exactly the same way, but they are all recording (or not) the same information--the game. Different people keep track of what they've observed, or somebody else observed and told them about, in different ways. But everybody is observing real games, keeping track somehow, whether it's written down or just in their heads, and drawing conclusions and forming expectations of future performance based on those conclusions.
Computers are no different from pencil and paper, or the TV you watch games on, except they're faster and bigger; they can't think for themselves, or make any deductions, any more than a pencil and scoresheet can. The game is always primary, and nobody is suggesting the actual games are unnecessary because we have video games.
Theophilus-
I buy tickets and watch the games AND I look at the numbers, because I love baseball and consume it as much and in as many ways as possible. Why should I limit my efforts to enjoy and understand the game?
Those of you complaining about stats and statheads in baseball might want to take a few minutes and read this excellent column:
http://www.sportsonearth.com/article/39977596/
I don't participate in the drink-owing nonsense.
Knew I'd get a rise out of you. :)
If you haven't seen the movie "Trouble With The Curve" with Clint Eastwood you may enjoy it. It was predictable but from so much of what I talk about here the movie strongly reinforces "scouting" over pure stats.
Hell, I used to play in a bar's football pool that one guy won more often then any other players and he never, and I mean never, watched football. He was a computer geek that helped write training manuals for submarines. I agree that the advanced stats information can be useful, and also entertaining. But when watching a batter, they don't tell me that he drops his hands, or throws his head out, or that an outfielder reflexively takes a step in at the crack of the bat,or if he has a natural "radar" to get a good jump, or whether a first baseman just"knows" when to come off the bag to stop an errant throw, or "knows" when to charge a bunt. These things are some of the nuances about the game that are more entertaining to me. I also love it when a play happens that four people just witnessed and you'll hear four different versions of what just happened!
Still, any body has an answer for my questions from around 9:00?
Joe
Fair enough, on your opinion of UZR. It seems to me that UZR--when properly understood and used, which is a big caveat--actually conforms, more often than not, with the "eye test" evaluation applied by experienced and diligent observers--which is also a big caveat.
When the 2 widely diverge, you've got a case worth looking at more closely. Sometimes, it's because the eye is seeing something the #s don't; sometimes, it's the reverse. Best to use both, correctly and in conjunction with one another.
I'm a stats geek but still prefer objective numerical stats like BA & OBP & ERA & Errors & Outfield Assists to subjective numbers like UZR.
As I have said I look at the stats and use my eyes to confirm.
Gorse-
No fudge factor in ZiPS and most of the other projections like PECOTA. I don't know how Bill James does it but I'd be shocked if he uses any kind of fudge factor.
Also, it's funny to see so much complaining about UZR in the comments to an article about offensive projections.
Ghost
I'm with you much of the way--tho I confess that I chuckled and thought of Cole Hamels when I got to the part about ERA and errors being "objective" stats.
I also want to reiterate that I am glad for all of the fans that love playing with advanced stats, and I am in no way meaning to complain about stats and statheads. I don't begin to try to hold a candle to many here in that regard, though I am known to look things up for comparison's sake.
hiramhover, good point. I guess there is a little bit of subjectivity with official scorers and costing Morse on that Cole Hamels hit/error call.
I wish there was a stat for batters that purposely hit behind runners to sacrifice them over even though its scored as an "out" via groundout.
The system isn't perfect by any means.
MLBTradeRumors has us looking at Vasquez on a minor league deal with an out date if he isn't he ML roster
NIDO spreadsheet modified to account for non-drink-owing-paying-curmudgeon.
Natsfan1a and Gorse, you both owe each other infinite drinks?
Dang. Who's driving YOU home?
Todd Boss does an in-depth article for 2013 based on Mark's WAR article a year ago.
http://www.nationalsarmrace.com/?p=3522
My biggest problem with James' projections for Bryce was the RBI total.It was the first thing I saw, and it set me off! Where is Bryce going to bat in the order? If he's 2nd like last year[highly unlikely] then 65 might not be too far off base, but if he moves down to 3, 4, or 5? Then I would fully expect his total to be well over 90.
Those of you complaining about stats and statheads in baseball might want to take a few minutes and read this excellent column:
http://www.sportsonearth.com/article/39977596/
The only thing excellent about it is that it perfectly highlights everything that's wrong with statheads in baseball - their smugness, their holier-than-thou attitude, and the chip they carry around on their shoulders. The author's main complaint seems to be that broadcasters and media don't emphasize the "advanced" stats when they report on baseball. He claims that these stats are used in every MLB front office, just like similar methods are used in the worlds of politics and finance. No argument there, but his conclusion that all baseball commentary should be based on these stats is ridiculous. There is a big difference between baseball and politics/finance. Baseball exists solely as entertainment, while politics/finance exist for other purposes, with any entertainment they offer being purely a side benefit. As entertainment, every single thing that feeds these SABR stats is available for viewing by the public, either via watching the games or by reading the box score. There is no need for broadcasters of the games to dwell on numbers generated by some complex formula as if the game could not possibly be understood without them. That would be like doing a PBP broadcast of trading on the stock market floor and talking about nothing but obscure financial data. There's a reason they don't do that - no one would watch it. Just because a bunch of nerds have invented some numbers that help them understand baseball does not mean that it's the only way to describe the game. It also doesn't mean that other ways of describing or understanding the game are not equally or even more valid. It's very easy to see why so many people who have played, coached or managed in MLB want nothing to do with the stats nerds.
There's this show called Clubhouse Confidential that bills itself as "the show for the thinking fan." There are lots of thinking fans like myself who don't agree with that, just like there are lots of Christians who don't agree with the Catholic church calling itself "the one true church."
Ghost, I liked the Todd Boss article.I also agree, that though the gist of this was if ALR left, but that if he stays then the team would rate even better.
Feelwood, I also agree with a couple of your points, especially about the point that nobody would watch it if the baseball broadcasts dwelled on SABR stats.
Joe Seamhead said...
My biggest problem with James' projections for Bryce was the RBI total.It was the first thing I saw, and it set me off! Where is Bryce going to bat in the order? If he's 2nd like last year[highly unlikely] then 65 might not be too far off base, but if he moves down to 3, 4, or 5? Then I would fully expect his total to be well over 90.
January 03, 2013 12:53 PM
Good point and who knows what Davey is thinking. The conventional wisdom is he won't stack the 2 pure lefties together in Denard & Bryce.
I'm thinking:
1. Span
2. Werth
3. Harper
4. Zim
5. Morse
6. Desmond
7. Espi
8. Ramozuki
Natsfan1a and Gorse, you both owe each other infinite drinks? Dang. Who's driving YOU home?
That's the beauty of infinite tabs--no one ever pays.
My biggest problem with James' projections for Bryce
That's part of the deal with bell curves--Harper isn't at the top of that, he's on the tail, an extreme case, and the general cases don't apply.
IOW, small sample size.
Joe Seamhead, in a perfect world 103 to 104 wins according to Todd and like he said injuries happen. 98 seems like another great season. Need to win the NL East again.
I think its interesting that he projected Gio Gonzalez down. Gio is a darn magician at times. As we know too many walks will kill you and it caught up with him in the playoffs but even during the regular season Gio looked like the fighter on the ropes who somehow kept getting up.
The problem is his high pitch counts and the reason behind it. When Gio is amp'd his curveball which is a "touch" pitch isn't working and he goes to the fastball on the corners and stops attacking the hitters.
If Gio can develop a slider, I believe he will be able to attack the zone more and get more swing & miss and keep his pitch count down and go deeper into games.
If he goes into 2013 relying on his success of 2012, I could also see him regressing some. He is one pitch away from being elite.
The only thing excellent about it is that it perfectly highlights everything that's wrong with statheads in baseball - their smugness, their holier-than-thou attitude, and the chip they carry around on their shoulders.
Irony of the day, perhaps, for Woody to complain about someone expressing frustration a little over-dramatically, but it did put me in mind of Kilgore's article on field goal kickers in the NFL, and Kai Forbath, who set a record this year for most kicks without a miss at the start of his career (17). Since I don't follow the NFL anymore, I did not know this:
"One prerequisite to becoming an NFL kicker is being one of the best kickers ever. Kicking clinics, coaching specialists and refined techniques have turned a gridiron afterthought into a cottage industry and transformed modern kicking. Twenty-five of the top 33 all-time leaders in field goal percentage, including 11 of the top 12, are active. Kickers have converted 83.9 percent of their field goal attempts this season, which would make the average NFL kicker in 2012 the 12th-most accurate of all time."
I think that's stunning. I imagine it represents a paradigm shift in how kicking is done, like when Fosbury started going over the highjump bar backwards. I wonder if using stats at all was intimidating to 19th century players?
I tried to look up their 2012 projections to see how close they were to actual numbers. Only then, we can put this complete disregard against favoritism debate to rest.
KURT SUZUKI PA HR RBI AVG OBP SLG
2012 Actual 442 6 43 .235 .276 .328
2013 James 424 8 47 .247 .306 .360
2013 ZiPS 509 12 63 .263 .308 .400
COMMENT: Suzuki finished the season strong. ZiPS seems to think that was the sign of an upward trend; James seems to think it was a mere blip.
WILSON RAMOS PA HR RBI AVG OBP SLG
2012 Actual 96 3 10 .265 .354 .398
2013 James 446 13 50 .269 .324 .426
2013 ZiPS 311 9 34 .260 .314 .411
COMMENT: Both projection methods believe Ramos can make a full recovery from a torn ACL and produce like he did as a rookie. The question is how much playing time he'll get.
I guess Bill James sees Ramos getting more of the reps as the season goes on while ZiPS has a 60/40 type of split on playing time.
I think the edge will go to whomever is hitting better and I expect Ramos to have a good year. I still think Ramos can be one of the best all-round catchers.
Zim is the best hitter o nthe club, he's going to bat 3rd.
Don, what's your batting order?
Anyone think there is even the slightest chance that Julian Vasquez would agree to a minor league deal and just be insurance for the Nats' rotation? He had a great year for the Marlins the in his last season before "retiring." Surely, someone's going to offer him a major league job.
Last time I checked baseball is a team sport and hitters are not batting in a vacuum. I think this line-up will get on base more then last year and hitters will get more looks at the pitcher and we will get deeper into the bullpen of the other team. If that happens everyone will hit better unless injured.
Runners on base open up holes and make it more likely that a pitcher will throw a fastball. I think ALR's career year last year had as much to do with him being on a good team as it was something he did. ALR is a a very good fastball hitter and if someone is on 2nd there will be no shift.
Last year when everyone got healthy the team started to score runs at a much great clip than earlier in the season. I have not checked but I would think that most of the batters had a better second half than first.
Or his brother, Javier...?
Section 222,
You are probably right that it would take a major league position to get Vasquez however, If the monies right and he was offered a future position of his choice other than manager with the organization he might take it?
Not sure if Vasquez wants a job in baseball or not. I'm sure he is set financially.
Don, would like to hear your proposed batting order.
Brutal straw-man murder in the comments today.
I'm failing to spot where anyone who has suggested that the projected numbers MIGHT have some merit has implied that neither scout nor fan needs to see ballgames anymore to inform player projections or to enjoy the game.
I cynically contend that if the sabermetricians had projected Bryce, Morse and/or Desmond as 30-100 guys next year, we wouldn't see half of these "projection numbers are crap" posts.
This spring, I will incorporate historically-based projections in my fantasy draft, and let my opponents draft on sentiment and past results alone.
Once the games roll out, I'll watch and celebrate those Nats who exceed projections, and grumble at those who fall short.
OK, here goes. If I had Davey Johnson's lineup pen and the season started tomorrow . . .
1. Span CF
2. Lombo 2B (Bernadina on days when he starts)
3. Zim 3B
4. Harper LF (move down to 6th, Morse to cleanup and Werth 5th for LHP)
5. Morse 1B
6. Werth RF
7. Desi SS
8. Suzuki C
Espi not starting is a tough call, but until he can make better contact Lombo's getting the nod from me. He can put the ball in play, drop a bunt, move a guy over and he does not K that much. The club needs guys on and over, they have enough guys who can mash it without Espi's 15-20 HRs for 190 Ks and .235 AVG (or something close to that over 650 PA).
Werth has had his best success as a big leaguer batting 5-6 and in this lineup he gets to swing away with RBI opps and his OBP ability gives Desi some RBI chances of his own -- plus this gives the club decent speed ahead of a bunting pitcher (Suzuki can run in a big way for a catcher).
I do think that it is unlikely that Espi actually loses the starting job, so if he does not, then I'd have him batting 7th bumping Desi to 6th, with Werth batting 2nd if for no other reason than no one else among the starters can really hit there.
Espi's slash line for the 2nd half: .264/.321/.432, with 10 HR.
Lombo's: .280/.306/.365 with 2 HR.
Sure, Espi strikes out a lot more than Lombo, but his pre-all star OPS was higher than Lombo's OPS in either half of the season, then he improved it by 70 points in the second half. I'm a big fan of Lombo, but I still think Espi's a much better hitter than his backup, just based on 2012.
Darn typos!
Well put Snivius. And welcome, by the way (or maybe you're a long time commenter and you just have a new nom de plume for the new year.) It seems to me that for purposes of this blog the projections are mostly interesting as a jumping off place for discussion. None of us here are going to stop rooting for or following the Nats because Bill James is less bullish on Harper than we are. Nor are we going to let his view dampen our hopes for a World Series bound team. Still, it's interesting to know what he thinks and discuss why it is or isn't valid. Over 90 comments for a January post with no news isn't bad.
One might say, "I prefer to watch the games and see what happens" in response to virtually any discussion of what might happen in the future, whether it's projected hitting stats or the best lineup against left handed pitching. But what fun would that be?
Not sure if Vasquez wants a job in baseball or not. I'm sure he is set financially.
Exactly. According to Baseball Reference, the guy's total career income is almost $100 million. I can kind of understand why he'd want to come back for another year while he can still pitch, but these days you don't often see ex-stars who made a boatload of money take non head-coaching jobs. Mark McGwire is an exception, but I can't think of a lot of others.
It seems like lots of folks want to see players stick around and be coaches. Remember when that was the perfect thing for Pudge to do? Now it's DeRosa who is the flavor of the month. But these guys have more interesting things to with their lives and their money. Being on the road with a baseball team isn't great for your family for one thing.
Snivius is no stranger, here... I recall seeing the handle, occasionally, since at least last season.... Always makes me think of Severus..... ;-)
Don.
Why is the 3rd spot reserved for the best hitter in your opinion?
In my opinion it's an advantage to have righty, lefty,righty etc.
Werth's greatest asset at this point in his career is his ability to get on base and spoil many good pitches; both should land him at or near the top of the batting order.
Span
Werth
Harper
Zim
Morse (ALR)
Desmond
Espi
Suzuki (Ramos)
That's my winter batting order prediction.
ManBearPig.. of that 2nd half on Espi, can you split out what was at SS vs 2B?
Convention and experience typically has the best hitter third to optimize ABs and RBI chances.
Feel Wood,
Smugness is not limited to people interested in advanced metrics; there are jerks on both sides of the argument.
What I find is that many veteran fans refuse to accept the fact that there are different and most probably better ways to measure performance than there was back in the day. Most new stats are hardly complicated; they just allow for a more analytical view of the game.
We used to think that a .300 hitter is great automatically; we now know to a fair degree of certainty that this is not necessarily the case. We used to think that RBI is the ultimate stat regarding run production; we now know that it's a very misleading stat. We used to ignore stadium effect on offense and pitching and now we don't.
We used to travel by horse and buggy and for the most part now we use a faster mode of transportation if we want to get somewhere in a hurry.
Finally, the idea that stats make scouting irrelevant is nonsense and is never claimed by people who study advanced metrics.
Desi was out from the last week in July through the middle of August, and Espi was hitting pretty well then (he hit well from July until the end of September, when he had his shoulder injury), so I'm sure it was better at SS than at 2b during the 2nd half. For the entire season, his slash line at SS was: .295/.340/.475 with 5 HR in 139 AB.
Gorse,
That's a theory often repeated but is not supported my many studies that disproved it.
I liked Don's lineup and he put some thought into various situations. I also think Davey will push Harper into the 4th spot and I think Bryce will lobby for it as well.
Span
Werth
Zimm
Harper
Morse /ALR
Desmond
Espinosa
Suzuki/Ramos
In my thoughts we have 7 positions out of 8 with the capability of 20+ homeruns. 5 out of 8 that can steal 20+ bases. Pretty damned excited when you add one of the top pitching staffs in the majors. Now,if I can just get HRod to take deep breaths and not dent the backstop I think we will be ok.
Go Nats!
One other thought, and I do not go back and look at much of the past stats, but, Zimm seems to really settle in well at the 3 spot and why would you move him if you don't have to. I would bet that Zimm has batted in the 3rd position more than any other.
I also think that Morse and Harper have more pure power than Zimm but that Zimm is our best hitter (hope that makes sense). So I want Zimm on base for the bashers behind him.
More on Espi - it looks like he did most of his 2nd half damage (to opponents, not to his own numbers) playing 2b rather than SS. From July 1 until Desmond went on the DL on July 22, Espi raised his BA from .225 to .250. During the rest of Desmond's time on the DL, Espi maintained his numbers at about the same level as they were on July 22. His line for July, his best month, was .301/.333/.553. Not quite as good for August (.255/.331/.420). September was about the same or maybe a little better than August, even with his shoulder problems (.255/.327/.436). It's interesting that even though his BA for August and Sept was almost 50 pts lower than it was for July, his OBP remained pretty steady at around .330 for the entire 3 month period.
Just came from NIDO and you guys are amazing! and definitely have way too much time on your hands. I still feel that if MicheleS wants to buy herself a drink she should be able to but I will defer here graciously. Thanks for all the fun!
Go Nats!! World Series or Bust!!
I can't see Vasquez coming here because of the offers he will get. He'd be great insurance for us but why come here and hope when some contender like Detroit will offer a better spot in their rotation (if Porcello is moved). Only way I could see if would be if he came here and said if he wasn't on the MLB roster when camp breaks then he goes free; gives him a chance to perhaps pitch on the best team in baseball while auditioning for other teams in case that spot doesn't develop.
JD -- on the 3 hole hitter being the best hitter on the club. I am not sure, but I think that is an old baseball philosophy. OBP-speed at 1, contact guy at 2 (can bunt, hit and run), best all around hitter 3, slugger 4th, power protection for clean-up guy 5th, etc. I think the idea is that you get your best hitter a chance to hit in the first inning of every game and maybe with someone in scoring position.
I like that lineup sjm. I'm taking your putting Suzuki/Ramos (not the other way around) into the 8th position as meaning you think Suzuki will get more starts. I would agree with that at this point, not so much because of Ramos' injury (which I think he'll be fully recovered from by spring training) but because I think Suzuki is more reliable defensively. Even before his injury, Ramos seemed to have some problems with receiving a throw cleanly and tagging a runner, so I would give the edge to Suzuki.
NatsLady,
I have long experience with computers, including programming them. I once arranged for a seminar on artificial intelligence, way before it became viable, as it is today, and headed up a team developing a computer model gaming contingencies for the use of the Strategic Petroleum Reserve. I know more than a little about computer tech.
With all due respect, anyone who has ever worked on computers and computer modeling can tell you that they all contain numerous variables which are assigned values according to one (or multiple) criteria, any one of which can drastically change the results of the model just by a change in the value of one of the variables, perhaps even an apparently minor such change.
I do not know of a single computer model that predicted that the Nats would win the NL East division last year. But I did, before the season started, and I would not, and did not, pay any heed to any computer modeling predicting otherwise.
To each his own.
NL-
What I said about computers does not apply to videos. I used videos to evaluate Anthony Rendon's swing, right here, just the other day, and looked at slo mo of Hank Aaron's swing, as well. Much different material when you are looking at the actual subject, instead of a programmer's model of it.
Any computer model is only as good as its programmer. A video of a hitter, or pitcher, is a rendering of the real thing, just as satellite imagery of a weather pattern will tell you what is actually occurring much more accurately than a computer model of the same thing.
ManBearPig.. Thanks valuable insight, maybe before he banged up his shoulder Espi actually figured it out (she says hopefully!!!)
Laddie: There are Lies, Damn Lies, and then there are statistics! As you stated, Everything can be manipulated to make a point. We see it here every day.
Also.. not saying that the SABR methods are bad. JD has an excellent point, use all the tools available to evaluate.
But I also believe in Trust but verify! The stats may say one thing, but it is always good to kick the tires in person. (ok.. enough cliches for today)
You're welcome, MicheleS! I've enjoyed reading and occasionally participating in the discussion here, and I'm glad I could make a contribution.
I wasn't sold on Espi's having turned the corner until I saw his monthly numbers from 2012. It gives me a lot of hope that in 2013 he'll build on the progress he made at the end of last season.
MicheleS, hummm Trust but vérifié. You may iwe me a drink!
JD said...
Feel Wood,
Smugness is not limited to people interested in advanced metrics; there are jerks on both sides of the argument.
What I find is that many veteran fans refuse to accept the fact that there are different and most probably better ways to measure performance than there was back in the day. Most new stats are hardly complicated; they just allow for a more analytical view of the game.
You are Exhibit A of the smugness of sabermetrics. Nerds aren't smarter just because they have nerdy formulas. Their so-called "advanced" stats aren't any better than the older stats. They're just different. They augment the older stats, they do not replace or negate them. The new ways aren't any better than the old ways. They are just two different ways of looking at the same thing.
Statistical modeling has come a long way Laddie Blah Blah, its why weather models on the new supers do so well at prediction where just a decade ago they too often fell on their faces.
A big part of that is the amount of memory available, meaning more variables can be used in "forever" calculations. Lots of parallel processing techniques are used to extend performance so that results can be out within hours. And if you've noticed the storm centers' often consult multiple models just as Mark has and then some do the additional calculations to find a middle ground between at least three models.
So, can we predict individual performance using statistical modeling adjusted and weighed using "eyeball" or "scouts" analyses? Yes, just as
weather modeling does. Can we predict the performance of an entire
teams, and leagues using clever statistical modeling. Yes, its possible but the appropriate algorithms and formulae must first be developed and
tested for accuracy. Something no one that I know of has paid to do as
yet.
I work and have worked in the parallel processing community for a number of years including for two of our FFRDC's. I'm not sure but I think I may have programmed in just about language and paradigm short a sparse few.
Have to admit though ... the time and money might be better spent on socio-economic prediction and analyses given the many woes of this world.
Manbear - that is exactly right and I agree with you. I like both our catchers and think Ramos will be fine, but I go with Suzuki right now
JD, don't take it personally. For some unknown reason, Feel (sometimes known as pRAA) has a chip on his shoulder against anyone who disagrees with him. I'm sure most readers of this blog saw your comment as well reasoned, friendly, and anything but smug. You make a lot of sense, NJ's opinion notwithstanding. Keep it up.
I like stats- I really do, but they only go so far. Some of these things seem like big, giant guesses. There are too many unpredictable variables. The "experts" said the Phil's would stay atop the NL East in 2012 and the Oakland A's would stay at the bottom of the AL West. Really? How did that work out?
Section 222 said...
JD, don't take it personally. For some unknown reason, Feel (sometimes known as pRAA) has a chip on his shoulder against anyone who disagrees with him.
No, I have a chip on my shoulder against people who are smug. Like, for instance, you. Smug people state an opinion and say people who don't agree with it aren't as smart as they are, or as good as they are. I don't do that - with two exceptions. I state my opinion and leave it at that. There are only two people here who I've called idiots, peric and JayB. I'm not proud of that, but I can't help myself. They are both idiots. And I don't think I'm the only one here who believes that.
JD, don't take it personally. For some unknown reason, Feel (sometimes known as pRAA) has a chip on his shoulder against anyone who disagrees with him.
Feel Wood, I feel your pain, but though some posters have been known to get under my skin,too,[and I've gotten under theirs], neither JayB or peric deserve to be called names, IMHO. The name calling is juvenile and only serves to lower the quality of this fine establishment that Mr. Zuckerman runs here. I happen to also identify with your feelings that some fans have an air of superiority because they spew out SABR stats, and that by being able to cut and paste stats from Fangraphs makes some feel "smarter" then those that don't do it. Jeez, just look at James' and ZiPs' stats above and though mostly pretty close, they are a mile apart on a few things. Hell, the next thing you know a Bill James type will come up with stats to analyze and measure how knowledgeable of a fan one is! Peace & harmony, everybody. I personally love having this site! Happy Nats Year to all of you! GYFNG!!!
Man, we *really* need some baseball.
Gorse.. I second that!!!!
"Man, we *really* need some baseball"
And I will post this again tomorrow..
Does anyone know a reasonable place to stay in New York? The Degenerate Nephew just asked for a weekend in NYC for Nats at Mets as his 18th birtday/Graduation present!!! HELP!!!! THE KID TOTALLY KNOWS HOW TO PLAY ME!!!!
MicheleS said...Does anyone know a reasonable place to stay in New York?
There are a lot of places to stay in New York City that are inexpensive and nice.
Unfortunately, the nice ones aren't inexpensive.
And the inexpensive ones aren't nice.
Try hotels.com, read the reviews.
Laddie, I read your post a couple of times, and I'm still not sure I understood your point. If you are saying you predicted the Nats would win the division and that makes you better than computer models, well, OK. I guess. If you are assuming I don't understand computers because I didn't spend twenty years programming them, being the absolute best in a specialized field, and as a consequence making a ton of money as a consultant, well, um, that's not a correct assumption.
I didn't predict how the Nats would do, because I don't make predictions; for me making predictions takes the fun out of the game, and that's also why I don't play fantasy sports. I'm struggling to read the books my mother sent me because I find fiction boring; facts, figures and history (and historical personalities and issues) are far more interesting to me than made-up stuff.
Seamhead, I can't speak for other posters. I don't cut and paste stats to make myself "feel" smart. If I put stats into an argument, it's because I believe that a particular stat may be relevant to the argument and add something to the discourse.
There are several reasonably priced hotels in Queens near the 7 train. We stay at the former Holiday Inn Express, which I believe is now called the the Queensboro Hotel.
Have to admit though ... the time and money might be better spent on socio-economic prediction and analyses given the many woes of this world.
peric, one of the best statements you've made! Not wanting to get into politics, I'll just applaud it and go back to wasting my time and resources on baseball... :)
MicheleS, the barter system works, if you are willing to take a risk. You find a family that wants to come to DC for the inauguration and put them up in return for that Mets weekend at their place. You might have to sleep on foutons, though.
Nats Lady,I have never felt that way about you, or most other posters that post stats.The operative word in my post was "some." I have always found your posts to be informative and well presented. And I agree with your last post.
"If you are assuming I don't understand computers because I didn't spend twenty years programming them, being the absolute best in a specialized field, and as a consequence making a ton of money as a consultant, well, um, that's not a correct assumption."
That goes for you, too. If you are well-versed in computer tech and computer modeling you should be well aware of the limitations of computer modeling. I could give you example after example, but since you are knowledgable in the field, that should not be necessary.
"If you are saying you predicted the Nats would win the division and that makes you better than computer models, well, OK. I guess."
Of course that's what I am saying. If computer models could accurately predict the outcomes of sporting events Las Vegas would go out of business, as would race tracks and private bookmakers. When you come up with a computer model that predicted that the Nats would win the division last year, let me know. Otherwise, on what basis do you question that assertion?
Just look at the diversion in the predictions of just the 2 models referenced in Mark's write-up. Which one do you buy? You could add a 3rd and a 4th, and there would still be a divergence of results because every model will be different from every other model because every computer modeler is different from every other.
You are free to place your confidence in them, as you wish.
Laddie, I am aware of the limitations of computer modeling, and I've said so on just about every post I've made concerning stats and the predictive value of stats. It's not only that a single, simple error by a programmer can destroy a model, so can bad data fed into the model. So what? Does that mean don't try?
As for "confidence," I don't know what you mean by that. For me, to place "confidence" in a model would mean that I would evacuate from an approaching hurricane (which I would), or that I would support investment in technology that can better predict tornadoes or earthquakes (which I also would).
I don't place "confidence" in models for sporting events. I find them amusing. Baseball games amuse me, and so do the related models that attempt to predict them.
It's fun for me to dig into the stats. Every year there is some nerd (usually young) who tries to invent a new one and make himself famous. Most of them are overly complex and don't add any insight. They remind me of songwriters who want that big hit or writers who think their semi-autobiographical first novel is interesting to someone besides their mother. Most of the constructs are complete drek, but every once in a while you find someone who actually has an interesting thought.
Nats Lady, I predicted that the Nats would win the NL East crown last Nov 1st and repeated it right before the season started.Lucky guess? Partly, but an educated,lucky guess based on I was convinced that the Phillies were in serious trouble,and though the Nats and the Braves were pretty evenly matched, the Nats had the advantage because of Davey Johnson. Unlike your policy, I enjoy making predictions, though I'm admittedly more often wrong then right.
NL
If you place no confidence in computer modeling, yourself, what is your point?
"Does that mean don't try?"
Try using your own ability to see, process and interpret information, instead. Works for me.
As you surely know, weather prognostications based on computer models do not predict the weather. They project a range of possibilities, instead, with a degree of confidence associated with each possibility. Those models are all collaborative efforts which involved millions of man-years of effort to produce, and they still are approximations, at best, and do not pretend to be anything else. The predictive weather models for the number of annual hurricanes has been notoriously inaccurate. even the old Farmer's Almanac has been better. Surely you know that.
The models in Mark's write-up make no such allowances for error nor does either predict a range of possibilities. If you do not place any confidence in them, either, then we share a sensible skepticism of such absurdly definitive predictions.
I do not make many predictions, either, but when I make them they are based on careful observation and many years of experience.
As you may have noticed, I have given up predicting what Rizzo will specifically do during the off-season, because I have no idea. However, I made a number of general predictions as to what he would not do during the off-seson, based on his past, observable strategy, his cost-benefit orientation to personnel matters, his long-term perspective, and his professed intentions, and I have been right about all of those predictions as well.
It ain't that hard.
Smug people state an opinion and say people who don't agree with it aren't as smart as they are, or as good as they are. I don't do that.
Actually, you do it all the time. You may not say it explicitly, but your tone and your way of stating your point makes it as plain as day that you feel superior to most of us peons who comment here. Fortunately, most people aren't deterred from stating their views even though they know there's a chance they will later find their words italicized, dissected, and ridiculed by you.
Honestly, I can't think of anyone here (well, actually there's one person, but he's been better lately) who posts stats to make himself feel smart. Many commenters here teach me something when they make an argument supported by data. Others have something special to offer because they've played and/or coached or have insights from close observation over many years. There's a rich variety of experience among us. I don't understand why anyone would reject any of it out of hand.
Try using your own ability to see, process and interpret information, instead. Works for me.
I do, WHEN I have that ability to see... But I don't always, because as I said in a previous post, I don't have those amazing eyes that can see every player in every game. And since I don't I rely on other people who input the data to "computer models" and out of these models come statistics. By no means do I think they are infallible. But they are useful.
I have been right about all of those predictions as well.
Must be nice. I haven't. When I've used my eyes and ability to process information, I've sometimes been mistaken, for example in my evaluation of C-M Wang, who I saw pitch a masterful game in Chicago. From that observation I thought he was talented and healthy. He was derailed by injury and then, I believe, lost confidence. Other people used his age and deconstructions of his previous stats to predict he would not be a success, but I used my eyes. Unless he reverses himself (he will be pitching for Taiwan in the WBC) I will have to conclude that they were right and I was wrong.
If you are saying you predicted the Nats would win the division and that makes you better than computer models, well, OK.
You can look it up Natslady. i predicted the Nats would win the division using computer models and statistical analyses. And I was right. My prediction was based on determining that the Nats bullpen would be the best in baseball along with a top 3-4 starters in the rotation. All else being equal to the previous year.
Look there are guys who know some stuff about computers and there are computists. Have computer will travel.
For some unknown reason, Feel (sometimes known as pRAA) has a chip on his shoulder against anyone who disagrees with him.
Which is why Mark deletes his comments from time-to-time. FeelWood is the ultimate troll on this blog. Bar none.
peric said...
"If you are saying you predicted the Nats would win the division and that makes you better than computer models, well, OK."
You can look it up Natslady. i predicted the Nats would win the division using computer models and statistical analyses. And I was right. My prediction was based on determining that the Nats bullpen would be the best in baseball along with a top 3-4 starters in the rotation. All else being equal to the previous year.
January 04, 2013 12:16 AM
______________________________________
I'm curious to see where this was written.
I did find people predicting the Nats would win the NL East in September 2012.
Ladies and Gentlemen,
I thank Mark for providing this forum and giving us so much food for thought.
I enjoy every poster but I don't enjoy every post. I can do without the personal attacks.
Arguing each other's conclusions and each other's processes are wonderful sources of entertainment for me.
Arguing the value of the individual who posted is not.
I don't post much because I'm always coming late to the argument, since I only read once or twice a day. For this reason I know I add very little value, and am basically a free rider.
I'll contribute more this year...a mixed blessing for all here.
For those who engage in personal attaks, unless you believe your attacks add to all our enjoyment, I recommend you reconsider before you send.
Happy New Year to all.
Let's Play Two!
Hear, hear!!! I am grateful that there are those iof us talking baseball this particular week. I have also been thinking how bad it is that there has been no hockey. Imagine how we would feel. The businesses in the Verzon Center area are being destroyed by this.
Corrected for content (yawn - yes, another cuppa would be great, thanks!).
SlowPitch, I would think that anyone who reads here provides value from the blog host's point of view, by virtue of the page view(s). They also serve who only sit and read. Further, imho, we're all equal, with no commenter being more valuable than another due to either content or frequency of posting. Loved your post, btw. :-)
Post a Comment