Bill James Projection System Bullish on Royals offense
by kcdc1 on Nov 14, 2011 5:53 PM EST
Email
PrintYou might remember this much-maligned post that tried to take a very early stab at how the Royals might do next season. At the time, I estimated that the Royals offense would cumulatively be about 46 runs above average, and people basically thought I was some combination of crazy, stupid, and blindly optimistic.
Well, today the Bill James projections were posted on Fangraphs, and if you add up the wRAA of the 9 projected starters (Perez, Butler, Hosmer, Gio, Escobar, Moose, Frenchy, Cain, Gordon), you get a whopping 96.3 runs above average. Check out the projections for free on the players' Fangraphs pages. Items like Hosmer's projected .293/.362/.494 are pretty fun.
I'm not a paying subscriber, so I can't download the spreadsheet to see how this ranks in the AL, but last year, 96.3 wRAA would have been just 5 runs behind the Tigers offense, good for 5th in the Majors. It would also represent a 50 run improvement over the 2011 Royals offensive output, despite significant projected regression from Alex Gordon and Jeff Francoeur.
It should be noted that the Bill James projections have a reputation for being more optimistic than other leading projections systems, but I don't have the data as to how warranted that reputation really is.
Once the roster settles and more of the projections are published, it will be easier to get a handle on where the Royals stand for next season. But this first data point lends support to my expectation that the projections will look rosier than many in the Royals Review community are currently expecting.
81 comments
|
4 recs |
Do you like this story?
Comments
Well, that's darned...
…exciting. I’d be happy to split the difference between a 46 and 93 run improvement.
This is the great thing about having young players performing at high levels—-the numbers actually can go up.
"Sir,--It has been wittily remarked that there are three kinds of falsehood: the first is a 'fib,' the second is a downright lie, and the third and most aggravated is statistics." *The National Observer* (June 13, 1891): p. 93-94.
AGord - 2011, BJ proj, Career
BB% / K% / ISO / BABIP
9.7 % 20.1% .200 .358
11.1% 21.4% .191 .331
9.8 % 21.5 % .172 .314
2011 must be weighted very heavily. Except BB rate, for some reason. BJ projection has higher than career ISO and BABIP. wOBA is .359 vs career .341, which is not totally out of whack. At a glance, I’d say the whole thing looks a bit optimistic, but not crazy or anything – though the BABIP rise seems particularly sketchy.
Kila's slash for Apr 20 to May 4, 2011, right before he was sent down: .276 / .344 / .448
by SagehenMacGyver47 on Nov 14, 2011 8:08 PM EST reply actions
FYI, Francoeur projection looks like it has more regression. Pretty close to his career averages,
which is not good enough to be an average RF, IIRC. .321 wOBA from BJ.
Kila's slash for Apr 20 to May 4, 2011, right before he was sent down: .276 / .344 / .448
by SagehenMacGyver47 on Nov 14, 2011 8:11 PM EST reply actions
Isn't Bill James routinely optimistic on everyone though?
Your 2010 Royals Review Fantasy Football Keeper League Champion
Since 2005: Royals win% = .4100, Chiefs win% = .4095
And I should have finished reading before posting.
Your 2010 Royals Review Fantasy Football Keeper League Champion
Since 2005: Royals win% = .4100, Chiefs win% = .4095
by averagegatsby on Nov 14, 2011 8:19 PM EST up reply actions
Yes very optimistic, he doesn't regress any of the values.
Also his yearly weighting are seem a little off.
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Nov 14, 2011 8:50 PM EST up reply actions
Are you sure? I can't imagine a modern projection system missing something as obvious as regression.....
Either way, there’s no such thing as optimistic. If there’s a problem, it’s going to hurt either hitting or pitching.
I’m sure there are studies out there that rate the accuracy of projection systems.
The "Bill James" projection system is widely maligned
First, just to get it out of the way, Bill James really doesn’t have anything to do with this system. From what I’ve read, he just gets paid so they can put his name on it. Second, this projection system’s methodology is widely maligned, and has been for years. It doesn’t use even a basic regression, as Jeff said. BJ projections for position players are almost always wildly positive. It’s a joke. It’s worthless. And that doesn’t have anything to do with what it happens to say about Royals players this year. It’s been a joke since its inception. No one takes it seriously.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 9:21 AM EST up reply actions
im really surprised that bill james lets that happen....
i mean…they cant be paying him THAT much to use his name and hurt his reputation
Fire Everyone
by billybeingbilly on Nov 15, 2011 9:26 AM EST up reply actions
A quick google search suggests that this comment is generally innacurate
Re: Bill Jame’s involvement:
After reviewing the comments on your original article, I wanted to mention one more thing in case you do a follow-up piece: Bill is most certainly still involved in the projections. He and John Dewan worked on the original coding shortly after BIS began publishing the Handbook, so most of the work is just the mechanics of running the code. Any changes to the code are instituted by Bill and John, and Bill, John, Steve Moyer, and others review the projections at least twice each and make manual playing time adjustments before we run the final version for the Handbook.
http://www.hardballtimes.com/main/fantasy/article/the-thought-process-behind-the-bill-james-system/
Re: “It doesn’t even use a basic regression.”
If you’re talking about playing time allocation, then you’re sort of correct. The projections assume a full, healthy season which I think is the right decision for projecting an individual player, but hurts projections for groups of players because some will get hurt. Either way, this doesn’t affect the rate stats. See:
I looked at the wOBA projections on FanGraphs for 154 players with at least 500 PAs in 2009 (I think everyone with 500 PAs except Colby Rasmus, Chris Coghlin, and Elvis Andrus, for whom I didn’t have BIS IDs). Taking the straight average of the wOBAs for the 154 players so that each is weighted the same amount and projected playing time is not considered, I get the following averages for CHONE, Marcel, and James:
.350 CHONE
.349 Marcel
.353 James
http://www.insidethebook.com/ee/index.php/site/comments/are_bill_james_forecasts_optimistic/
If you continue reading that link, you’ll see that the Bill James system is more optimistic than Marcel and CHONE on pitchers. Tango suggests that the playing time allocations cause this, and he could be right if the player pool includes very good pitchers that are assumed to stay healthy and AAA depth pitchers that are assumed to be kept in AAA. It’s possible that the playing time thresholds Kincaid used (500 AB’s vs 100 IP) mostly prevented back-ups and scrubs from getting into the offensive pool, but allowed them into the pitching pool, and this would explain the discrepancy.
Anyway, your blasting of the system is mostly unfounded. It may not be as accurate as other leading projection systems, but it’s probably pretty close, and a lot of the discrepancy can be accounted for by understanding the choice they’ve made in dealing with projecting playing time.
by kcdc1 on Nov 15, 2011 10:15 AM EST up reply actions 5 recs
It may not be as accurate as other leading projection systems,
It’s horribly inaccurate. It’s awful. It’s amazingly high every year. There’s no reason to take it seriously. Really. But hey, it’s positive about the Royals, so it must be pretty close.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 10:19 AM EST up reply actions
Source
Did you even read my comment?
I looked at the wOBA projections on FanGraphs for 154 players with at least 500 PAs in 2009 (I think everyone with 500 PAs except Colby Rasmus, Chris Coghlin, and Elvis Andrus, for whom I didn’t have BIS IDs). Taking the straight average of the wOBAs for the 154 players so that each is weighted the same amount and projected playing time is not considered, I get the following averages for CHONE, Marcel, and James:
.350 CHONE
.349 Marcel
.353 James
The rate stats appear fine.
And I thought we’d put this “positivity thesis” behind us. Do you want me to pick out things that I think you’re wrong on and link those inaccuracies to psychological motivations?
That part about wOBA's should be block quoted as it was in the previous comment
I didn’t do any of that work.
If these projections were all low for the Royals instead of all high, would you even be talking about them?
I mean, really. There are good reasons why BJ’s projections have this reputation. There are good reasons why (IIRC), Tango doesn’t even include the BJ projections in his semi-annual evaluation of forecast systems.
But hey, maybe all of these guys really are going to be very good and the Royals will win 90+ games. Feel free to hang your hat on these projections if you like. It’s fun. I’ll be interested in see what a good system like ZiPS says.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 10:33 AM EST up reply actions
Ok, so you're not interested in evidence or data.
My apologies. I shouldn’t have wasted my time.
If you choose to continue this conversation, please source your statements.
I shouldn’t have wasted my time.
Or mine. I think what you should try to do is honestly and objectively evaluate this team, instead of this.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 10:53 AM EST up reply actions
It’s just pointing out what I see as bias. It’s not like I’m calling you ugly, stupid or talking about your mother.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 11:08 AM EST up reply actions
Then argue against what is said rather than the person who says it. Every position should stand on its own merits.
For what it’s worth, for purposes of 2012, I think you’re unrealistically negative on the Royals chances, but I don’t respond to all of your inaccurate statements with a blanket, “Old Negative Scott, at it again!”
FTR, do you think a prediction of say 76-77 wins is unrealistically negative? And do I only make fanposts/fanshots which are negative about the team? If all of my fanposts/fanshots were negative about the Royals, wouldn’t that lead you to consider that maybe my analysis was biased (either concsiously or subconsciously)? I’m happy to recognize how good the good Royals players are. Hell, I’ve even written fanposts giving props to Moore for his contract extensions for young players and how great he does with arbitration settlements. It just seems like your analysis is a process of trying to put flesh on a skeleton of wishcasting.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 11:50 AM EST up reply actions
Glad to see your predictions inching up. Last I checked, you were in the 75-76 win range; maybe the Sanchez trade made the difference.
And yes, I do think 76-77 wins is unrealistically negative. When the ZiPS projections are published, I’ll sum batting runs above average and pitching runs above average for a projected run differential, and I’ll use that run differential to project a W-L record. I’ll bet that it will come in above 77 wins.
And as for what I write posts/links for
I don’t write about everything I consider or observe. I open threads when I have something to say that I think is interesting and is not being said elsewhere. Since the negative positions tend to get plenty of coverage at RR, I don’t write about them.
Taking my “Francoeur Is Better Than You Think” post as an example, at the time I wrote that, I felt that the community thought Francouer was awful. If I had thought Francoeur was awful, I would have remained silent on the issue because that position already had plenty of voices. But since I thought Francoeur was kind of okay (average-ish), I opened a thread making that case.
Also, I'm looking forward to the release of more reputable projection systems as well
The James projections look optimistic to me as well, but I’m not inclined to write them off completely unless someone has done a study and found that they’re wildly inaccurate. What I’ve read suggests that they probably aren’t as good as CHONE, but they’re not awful.
Who knows, maybe these projections are in the right ballpark. I doubt it. It doesn’t look like it to me, or that BJ’s projections are ever in the right ballpark. But we’ll see what systems like ZiPS and PECOTA say. I wish Chone was still published.
One problem with accurately projecting some Royals players is that all of the Royals 2011 rookies are going to have projections which will still significantly be based on MLE’s. That doesn’t make those projections worthless, but less reliable. Moustakas will be projected to have a decent 2012 because of his minor league numbers. But of course how a player makes the transition to the majors is all over the place. It’s hard to accurately project MLB performance from MiLB data. But I’m certainly willing to accept ZiPS and PECOTA projections as the best information we have about how good these players and this team should be in 2012.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 11:01 AM EST up reply actions
This is the correct response
Any projection system projecting the Royals offense is going to be doing so projecting Johnny Giavotella, Lorenzo Cain, Mike Moustakas, Eric Hosmer, and Salvador Perez.
None of those guys even have a full season in the majors. If MLE equivalents were reliable Kila and Clint Robinson would be Hall of Famers.
For the Tigers
The James projections are regularly almost exactly 5% higher in OPS than the ZiPS projections. The eyeball test suggests that ZiPS is the sane party at the table. It’s almost like the James projections missed the boat on the declining scoring environment.
You say it has a reputation for being too optimistic,
he’s got a citation of a specific instance where that was not really true (rate stats in 2009). But you don’t seem to address that. Perhaps the rate stats are less optimistic, while the counting stats are more optimistic.
Kila's slash for Apr 20 to May 4, 2011, right before he was sent down: .276 / .344 / .448
by SagehenMacGyver47 on Nov 15, 2011 12:33 PM EST up reply actions
The system works in such a way that the counting stats are going to be too optimistic. The James projections do try to get the rate stats right tho, and it’s not clear to me whether they do a good job with that. A cursory glance at projected rate stats for 2012 leads me to suspect that they’re systematically high, but I wouldn’t throw them out entirely without a proper study.
Would you assume that they are “pretty close” without a proper study?
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 12:45 PM EST up reply actions
I was going off Kincaid’s finding that the 2010 James wOBA projections were within 1% of the 2010 CHONE wOBA projections. At that point, I thought that they were probably pretty close.
Since then, I’ve compared the James and ZiPS OPS projections for the Tigers starters, and found that the James projections were about 5% higher. New information, new conclusion.
I’ll note that a sample of 9 projections is not conclusive, and that my current position on the James projections is not at all fixed. And I’ll also note that you haven’t bothered to incorporate evidence or data at all.
by kcdc1 on Nov 15, 2011 12:58 PM EST up reply actions 4 recs
What about this comment from Tango?
Kincaid: great stuff!
4.22 Marcel
4.13 James
Now, that was ENTIRELY predictable, no?
There’s no question that the Bill James forecasts are optimistic for both hitting and pitching, and it has nothing to do with any logical rationalization by Ben or anyone.
If you were to apply any kind of reasonable playing time for every player, you will find that the Marcel runs scored v runs allowed will pretty much match to each other (differential of 0), while the Bill James forecasts will end up with a differential of something like 1000 runs.
It doesn’t really matter if you treat each forecasting system as its own universe, since you can re-calibrate as you need it. But, you can’t then try to compare individual players from each system.
Glad I came, just wish I hadn't stayed so long.
Rock Chalk Talk
by Warden11 on Nov 15, 2011 11:51 AM EST up reply actions 1 recs
This comment has to do with the playing time allocations. This is what I was talking about when I said:
If you continue reading that link, you’ll see that the Bill James system is more optimistic than Marcel and CHONE on pitchers. Tango suggests that the playing time allocations cause this, and he could be right if the player pool includes very good pitchers that are assumed to stay healthy and AAA depth pitchers that are assumed to be kept in AAA. It’s possible that the playing time thresholds Kincaid used (500 AB’s vs 100 IP) mostly prevented back-ups and scrubs from getting into the offensive pool, but allowed them into the pitching pool, and this would explain the discrepancy.
The “differential of something like 1000 runs” comes from assuming perfect health for hitters and ordinary health for pitchers for the hitting projection, and assuming perfect health for the pitchers and ordinary health for hitters for the pitching projection. Tango is right that this poses problems for looking at the whole population, but it’s arguably a good way to project individual players, as a fantasy player might be looking for.
Are we talking about this for fantasy purposes?
but it’s arguably a good way to project individual players, as a fantasy player might be looking for.
Glad I came, just wish I hadn't stayed so long.
Rock Chalk Talk
No
I’m just explaining why the James system projects playing time in the way that it does. For projecting playing time, you know some players will get injured and miss games, but you don’t know who it will be so you can (1) say everyone will be healthy and allot them full seasons, or (2) apply a standard discount so that everyone gets something like 85% of a season. Option (1) is arguably better if you’re looking at an individual because he’ll likely stay healthy, but option (2) is better if you’re looking at a group of players like a team because some players in that group will likely get hurt.
I think the James system wants to cater to fantasy players and to project how players will do if they stay healthy, so they go with option (1).
Naturally, this assumption of perfect health will make team projections unrealistically optimistic.
Naturally, this assumption of perfect health will make team projections unrealistically optimistic.
That makes me question this comment from your post and wonder why it’s any use to even think about at this moment.
I’m not a paying subscriber, so I can’t download the spreadsheet to see how this ranks in the AL, but last year, 96.3 wRAA would have been just 5 runs behind the Tigers offense, good for 5th in the Majors. It would also represent a 50 run improvement over the 2011 Royals offensive output
Why would I care about this when you just stated that they’re unrealistically optimistic?
Glad I came, just wish I hadn't stayed so long.
Rock Chalk Talk
. The projections assume a full, healthy season which I think is the right decision for projecting an individual player, but hurts projections for groups of players because some will get hurt. Either way, this doesn’t affect the rate stats.
I can’t speak to the rest of the projections, but I think this the appropriate take. I don’t like that ZIPS tries to predict injuries. I’m not sure injuries can be all that predictive, except for maybe in a general sense (Mark Prior is an injury threat). I don’t think you can predict actual playing time based on health. I’d rather see a projection system that sets up a baseline and lets the user take that info and project injuries/playing time on their own.
Relive Royals History at royalsretro.blogspot.com
by RoyalsRetro on Nov 15, 2011 11:01 AM EST up reply actions
I guess that is problematic in team sims though
Relive Royals History at royalsretro.blogspot.com
by RoyalsRetro on Nov 15, 2011 11:02 AM EST up reply actions
Only to a point it would seem
I will readily admit I am not following the full conversation. But it seems to me the problem only comes if you compare projections to other systems or actual performance data from previous years. If you wanted to get a snapshot about actual team talent and projected performance, maybe relative to league or division, you could still do a team-by-team comparison because the inflation from healthy projections would theoretically be stable for all teams. In fact, it might be much better for measuring talent level than a system that chooses to factor in very unpredictable play time factors. Using that data, you might be able to say the Royals could or could not compete (assuming a healthy team).
I agree
Kila's slash for Apr 20 to May 4, 2011, right before he was sent down: .276 / .344 / .448
by SagehenMacGyver47 on Nov 15, 2011 12:33 PM EST up reply actions
Or this quote from the original post at THT,
The most important concept I’d like to stress is that of relativity. The kinds of articles I just mentioned operate under the assumption that the James projection for a player should be looked at relative to another system’s projection for him or relative to last year. This is incorrect, though. What we should be doing is examining the James projection for a player relative to all of the other players the James system projects.
As noted in the comments of that post, I don’t understand why we’d want to use numbers from this projection system when the numbers don’t match the league’s run environment.
Glad I came, just wish I hadn't stayed so long.
Rock Chalk Talk
This is a fair point
See above where I wrote:
The James projections are regularly almost exactly 5% higher in OPS than the ZiPS projections. The eyeball test suggests that ZiPS is the sane party at the table. It’s almost like the James projections missed the boat on the declining scoring environment.
I think it’s interesting that for projecting 2010, the James projections were within 3 points of wOBA of the CHONE projections. I doubt that will be the case for this season, and I wonder if it may be that the James projections haven’t accounted well for the decline in scoring environment.
Excessively bullish, I'd say
The projections that leaped out at me were
Alcides .267/.309/.360, wOBA .297
Hosmer .311/.362/.494, wOBA .369
Moustakas .278/.323/.447, wOBA .336
Cain .284/.337/.416, wOBA .334
I’ll take that. Where do I sign up?
Alcides is projected as about a league-average shortstop at the bat (669 OPS). I can actually buy that: he’s not a slap-hitter like Getz, he drives the ball (21 2B, 8 3B, 4 HR last year) at least sometimes. If he fields anywhere near as well as he did last year, he’s a 2.5 WAR player.
"All the boys think she's a guy
She's got crazy Frenchy eyes."
According to the BJ projections, apparently every team is going to win 85+ games next year
Before you tell me that’s unrealistic, prove to me that it won’t happen!!!
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 10:16 AM EST up reply actions
Do you have any evidence for that statement?
Or are you just throwing it out there?
I’ll do some work for you.
The Tiger’s projections (given their best offensive starting 9 according to the Bill James system), sums to 143.8 wRAA.
If you discount the Royals projection by the percentage that the Tigers’ projection overshoots last season’s total, it bumps the Royals projection down to 68 wRAA.
Do you have any evidence for that statement? Or are you just throwing it out there?
Just having a little fun with the worst projection system.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 10:46 AM EST up reply actions
It's "probably pretty close"
Just like Alcides Escobar’s jump from a .282 wOBA to to .297. That seems reasonable.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 10:36 AM EST up reply actions 1 recs
It looks slightly high to me
But almost all of that improvement comes from his BABIP raising from .285 to .299. So the question for that projection is whether a .299 BABIP sound unreasonable for Escobar. I’d say it’d take some improvement, but he is on the improving side of the aging curve.
Is that "laughably" optimistic?
What would the MLEs of a guy who for a minor league career is .305/.375/.437 and hit .338/.390/.491 at AAA the previous (age 24) season?
Nick Swisher and Johnny Giavatella deserve treats in the Kaufmann parking lot.
For comparison, Hosmer's line this year was .293/.334/.465 with a wOBA of .342.
Ackley’s 2011 number’s for 2011 were: .273/.348/.417, wOBA of.340
To suggest that Gio will be as good as Ackley and almost as good as Hosmer were last year is laughable to me.
by Loose Seal on Nov 15, 2011 3:39 PM EST up reply actions 1 recs
We're talking about projections thought?
Right?
Nick Swisher and Johnny Giavatella deserve treats in the Kaufmann parking lot.
My point is that Bill James is projecting Gio to put up ROY-like numbers
(though I dunno if Gio would stil be eligible…then again, he’d have strong competition from Cain and Salvador Perez according to the projections…lets just stop dreaming ).
To paraphrase NYRoyal, that projection is not outside the realm of possibility, but it is far from the most likely outcome for Gio’s 2012 production.
by Loose Seal on Nov 15, 2011 5:01 PM EST up reply actions 1 recs
We're talking about projections though.
right?
Nick Swisher and Johnny Giavatella deserve treats in the Kaufmann parking lot.
Oliver, offense, and standings
I don’t know if they are accessible any more, but THT’s Oliver had six-year “true talent” projections on its player pages that were updated right through the end of the season. Obviously, you want to take long-term projections for any individual player with a grain of salt, but by the end of the season, the 2012 numbers are pretty much what they will be when Oliver 2012 comes out.. I managed to get the basic numbers for the Royals and other AL Central teams onto a spreadsheet for a basic standings projection for 2012, I don’t know if I will do a “big” one this spring or not,but this was interesting.
Remember, this are not my projections, I’m just doing stuff with them. Complain to Brian Cartwright. I’m not going to comment much on individual projections. I suspect these will be pretty much in line with what we will see from ZiPS.
First, the Royals projected starting offense (by wOBA):
Perez .297
Hosmer .349
Giavotella .316
Escobar .268
Moustakas .321
Gordon .354
Francoeur .320
Butler .362
The simple average is .322. The Royals team wOBA was .325 in 2011.
To make things more interesting, I did a crude “projection” of the AL Central using used each team’s currenty projected starting lineup (as best I could put them together given that things will change), starting five pitchers, and top five relievers. I did not deal with defense or baserunning. I simply generated runs allowed based on an average of the five starting pitchers and the five relievers, (weighting starting pitchers at 940 innings, relievers at 505). Then I converted wOBA to “absolute” runs created based on the 2011 run environment (and 6250 PA per team). That allowed me to get a run differential and “neutral” winning percentage for each team. Then I applied that to 162 games.
Obviously, this leaves a lot of things out: adjusting for stuff like playing time, benches, and better offensive teams getting more PAs (and worse getting fewer). I am not going to post projected “records” here, since that would be even more confusing — I would need to project every team in the league then run a log5/Odds ratio thing, and I wanted to do a relatively quick one. This assumes that each team would play the same schedule.
I did have to fill some holes beyond current rosters — I gave the Royals Francis as their fifth starter, and let the Twins have Kubel as their DH (they needed mercy).
Again, this is just something fun to mess with, at least for me. Before I get to the projected pseudo-standings, here are he projected 2012 AL Central offenses by wOBA according to this method:
DET .336
CLE .335
KCA .322
CHA .322
MIN .326 (Kubel!)
As far as the projected standings go, I’ll list the team with the most projected “neutral” wins (I mean assuming all these teams play a schedule full of true talent .500 teams, which obviously isn’t the case, but gives a rough idea of the projected relative talent levels of teach team). Remember — I’m not presenting these as “my” projected standings, this is just for the curious and for discussion. Team name, then “games back from leader”
1. DET 0
2. CLE -5
3. CHA -11
4. KCA -14
5. MIN -19
Making watching baseball as fun as doing your taxes.
My Twitter feed.
Before getting tweaked, read up on regression.
by Matt Klaassen on Nov 15, 2011 12:20 PM EST reply actions 1 recs
The simple average is .322. The Royals team wOBA was .325 in 2011.
Thank god the Royals good offense is going to carry it’s comparatively poor pitching.
You may know me as NYRoyal.
by Scott McKinney on Nov 15, 2011 12:26 PM EST up reply actions
Interesting--thanks for posting
Most of those numbers look reasonable, tho I wonder why Escobar’s projected at a .268 wOBA when his last two years were .270 and .282. I’d also expect more from Hosmer, and slightly more from Moose and Perez.
Why did you leave off Cain in making the predictions?
You only have 8 Royals starters.
Go Royals!
I noticed that too, but figured that it didn't have a projection posted for Cain at the time Matt pulled the data.
Cain only got into something like 3 games at the very end of the season, so it’d be understandable.
Or maybe Matt thinks Gordon and Frenchy are so good that the Royals don’t need a centerfielder.
Actually, the numbers add up to a .323 wOBA for the Royals, so I think he forgot to add him but he is in
the results because the has them at .322 overall.
Go Royals!
Oops, he's included in the projections, just skipped that line when I typed the comment
.314 wOBA
Making watching baseball as fun as doing your taxes.
My Twitter feed.
Before getting tweaked, read up on regression.
by Matt Klaassen on Nov 15, 2011 2:38 PM EST up reply actions
Seems optimistic for Cleveland...
Generally I have no patience for individual projections because the error bands are so wide, but I do think that aggregated projections are helpful. And this gibes with my own thoughts on the mean outcome for 2012. But the Cleveland love I don’t necessarily get. I think that Detroit’s probably 10 games better than any team in the division next year.
I will say that the nature of the Royals roster makes projecting next season’s offense particularly problematic because of the lack of major league experience for five projected starters. And it’s precisely because of that lack of experience that makes me want to be much more conservative for the team as a whole.
The Royals will probably be a pretty bad team next year, which is fine with me. One question for devilfingers or whomever: are players with limited MLB experience less likely to come close to their mean projection? If so, doesn’t that significantly change the team projections because some players will be unplayably bad? And so won’t play? And will be replaced by someone ostensibly better?
by billexgordler on Nov 15, 2011 11:19 PM EST up reply actions
are players with limited MLB experience less likely to come close to their mean projection?
I would say definitely yes.
If so, doesn’t that significantly change the team projections because some players will be unplayably bad? And so won’t play? And will be replaced by someone ostensibly better?
That’s a valid point, but how do we figure that in? For instance, let’s say that a young player’s projection is to have a .310 wOBA. But instead he actually hits .280 for the first month or two and is replaced. And the replacement hits .290. I think that’s a reasonable scenario. How would we figure that into projections? But certainly projecting a team with a lot of young players is very difficult to do.
You may know me as NYRoyal.
by Scott McKinney on Nov 16, 2011 9:56 AM EST up reply actions
Exactly.
I think erring on the side of conservatism is probably right. My conclusion will be to treat mean projections for the 2012 Royals as highly speculative and not significantly more or less likely than optimistic or pessimistic projections which I imagine can realistically range from 65-85 wins.
by billexgordler on Nov 16, 2011 12:03 PM EST via mobile up reply actions
I would say that the 50th percentile projection for the team is significantly more likely than the 75th or 25th percentile projection. While the magnitude of that likelihood is likely smaller for a team with a lot of young players with little MLB data that it is for a team full of vets, I think it is still true.
You may know me as NYRoyal.
by Scott McKinney on Nov 16, 2011 12:20 PM EST up reply actions
Projections for young players have more variance, but that doesn't mean that they tend to come in lower than projected
Let’s say Player X is projected at a .310 wOBA. Assuming the projection is good, he’s as likely hit better than that as he is to hit worse. Your point about replacements actually tends to favor teams with young players because it effectively puts a floor on the performance the team gets. If Player X is really a .350 wOBA hitter, the team enjoys that production, but if the projection is equally wrong in the other direction and Player X is a .270 wOBA hitter, the team replaces him with a .290 wOBA hitter and gets the .290 wOBA production. Since the production the team gets has a floor, this actually skews the team’s expected production slightly above the mean of the player’s production distribution.
That complicated explanation aside, there’s a lot of uncertainty with the Royals projection, and if you want to take the under, there’s a good chance you’ll be right.
Without any evidence, I think they will in the case of the Bill James Projections
There are too many guys who crush AAA and AA statistically but aren’t actually very good like Kila.
Sure, my point only applies if the projection is good
Before you start tweaking expectations up and down from a projection, you need to know if the projection is worth using as a baseline.
My projections...
suggest that the Royals are going to be really fun to watch this year. Heck, they were fun last year. For a change.
BJ predictions
Do the BJ predictions take into account whether she has a headache or the time of the month?
by ks.cowboy on Nov 15, 2011 1:56 PM EST reply actions 1 recs
Error in a measurement device is fine as long as it is consistent.
All you have to do is recalibrate to fix the error.
by BlueEyes_Austin on Nov 16, 2011 9:53 AM EST reply actions
A 5% error is pretty strange tho
I suspect the BJ projections weren’t properly tuned to the declining scoring environment which might mean that you can just recalibrate everything down 5%…..but they might also just be bad projections.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
In Association With
Big League Stew - MLB - Yahoo! Sports
- College player makes impossible over the fence catch to save tournament game (Video)
- Security!: Camden Yards guard collides with Royals’ Mike Moustakas
- The Juice: Nelson Cruz matches career high eight RBIs, Pujols stays hot in Seattle
- Benches clear at Fenway as Red Sox-Rays rivalry heats up
- Keeping good company: Albert Pujols becomes fourth-youngest player to hit 450 home runs
