CHONE/BtBS/d_f Quasi-Projection for the 2009 Royals BETA
by Matt Klaassen on Jan 10, 2009 12:53 AM EST
Email
Print
Yup, it's the moment you've all been waiting for -- me to give away the dirty little secret that was going to net me a RR t-shirt next fall.
Or not.
Originally, this was just going to be a FanShot, but I just kept going, and going, and -- you get the picture.
Sky Kalkman of Beyond the Box Score fame came out with a spreadsheet (still in BETA) in which we could enter information from a projection system or whatever for a team -- it then generates the WAR and gives a win prediction. So I did it for our beloved Royals. I didn't want to at first, but then I got excited to see what Willie Bloomquist would add the the pennant aspirations of the boys in blue. How did I do it? Well, I started with Chone's projections. The used the "official" 50th percentiles for offense and pitching rate stats and stuff. That what I use in "Royals 2009 projection," where you should start -- the one that "predicts" 79.3 wins.
[NB: I don't know if this is a problem, but while Sky used a .335 lgwOBA, I messed around and believe I figured out that Fangraphs using .328 for CHONE's lgwOBA, so I changed the sheet accordingly.]
Then I had to tweak PAs, IPs, and leverage ratings for the relievers. I tried to strike a balance between meeting the minimums for the sheet to "work," the cream rising to the top (or, in the case of, e.g., the pr0fe550r, Hillman somehow eventually realizing that he sucks and shouldn't be setting up), and so on. Some projections you won't agree with. Some I don't either. I know the playing time stuff probably doesn't quite add up right yet. Most of it is on my end, some of it are kinks Sky is in the process of working out in the spreadsheet.
For defense, I started with CHONE's projections. Then I went to bUZR at Fangraphs to balance it out a bit for projections that were too extreme one way or the other. In some cases, where there was a lot of disagreement and/or less than three years of data, I looked at the Fans Scouting Report, although that had less priority. So, for example, despite Gordon's bad year defensively this year, almost all the metrics had him above average last year, and the FSR had him great last year, and average this year. Given the paucity of data, I simply have him as average. In general, when in doubt, I move guys a bit closer to "0" (average). With Butler and Jacobs, I realize (and hope, in Butler's case), that they will see time at 1st. I simply assume that either it will be negligible, or that (particularly in Jacobs case, as I've shown elsewhere) that they are so bad at 1B they're better off DHing anyway. Again, I'm working within the bounds of the spreadsheet balanced with my guesses as to what Trey/Dayton will do as well what I think should be done.
For baserunning, I looked at Baseball Prospectus' EQBRR stat, minus the stolen base portion (since wOBA includes that already), and again, used it pretty conservatively, given the paucity of years available and the difficulty in projecting something like that.
79 wins probably doesn't set your heart abaze, so I did another "less official" projection found in "Royals Bumped" where I used Alex Gordon, Billy Butler, and Luke Hochevar's 60th percentile projections, as well as Zack Greinke's 70th (keep in mind I upped all the IPs and PAs for these guys in all the different versions). That took the Royals to 82 (81.8, to be exact) wins. I do think that the 70th is close for Zack and 60th for Gordon, in particular. I think it's also reasonable for Butler and Hochevar as well. Great job on drafting those guys, DMGM!
For even more fun, I put together an "If I Ran the Zoo" sheet, which reflects what I would have done in the offseason. No Jacobs, Olivo, Coco, or even Slick Willie. I either try to move Joe Guillen after a hot streak for whoever will take most of his contract for a bag of balls or put him in a platoon with Teahen. Ryan Langerhans (former Brave!) is claimed off of waivers when he passed through and installed in the OF. Brayan Pena splits time/backs up Buck. J. R. House sees a bit of time at the end of the season at catcher, 1B, and DH (hence the bad defensive rating for him). And other stuff. I use the same numbers for everyone as I did on the "bump" tab. Keep in mind that Leo and RamRam are still around, and that this is all a lot cheaper, so imagine what would happen if Orlando Cabrera was plugged in for 3/27 with all the money I saved!
And, yes, I'm taking suggestions for alterations on the main sheet -- particularly regarding playing time. The projections I'm unlikely to alter much, although that doesn't mean we can't argue about them. Again, I'm not necessarily putting this forward as what I think players will do -- just packaging the CHONE projections in a particular form (although CHONE obviously can't be responsible for my silly contributions, of course).
It's just for chuckles and discussion, so just chill. Remember, I'm just disseminating projections in a certain form -- don't kill the messenger.
[Update: 1.31 AM EST -- oops, I forgot on the "Zoo" sheet to get rid of Farnsworth and HoRam and put back Nunez and Ramirez... I just kept Waechter for the heck of it. I'm even more of a genius now.]
[Updated Sun, January 12, 3:12 PM EST: I justed added another tab, 'Standing Pat,' which basically uses last year's roster, as if no moves were made. I did have to make some changes, but I kept it as close to the playing time, etc. of currently projected team. I did also leave Waechter in to soak up some innings, but note that this doesn't really help this projection, since even with minimal leverage assigned to him, he's still a sub-replacement pitcher according to his the spreadsheet. Note that I include what I consider to be completely gratuitous numbers of PAs for Ross Gload and Tony Pena, Jr. I believe that a simple comparison of the respective projections needs no commentary from me.]
[Update, Sun, January 11, 10:55 P.M. EST: Here's a chart I made using the binomial distribution function in Excel that should give a sense of the win percentages based on the 79 wins prediction. The percentages are that the chances Royals win will at least that many games assuming 79.3 win talent. At least that's what they tell me.]
[Update January 14, 2:20 EST-ish: I redid all the win probability curves to make more sense visually (thanks to ZeppelinDZ for the solution), and put them all on one sheet. I also ended up adding a faint curve for "Standing Pat" without data labels, but I think it speaks for itself. Enjoy.]
Click on the image to enlarge it (line).
144 comments
|
4 recs |
Do you like this story?
Comments
Excellent work
79-82 wins sounds pretty realistic. The really hard part is making all of the playing time work out realistically. Of course we don’t know how Hillman will use these guys, so all we can do is make our best guesses. And your estimates are very realistic. The only thing I’d tweak is adding in some more scrubs. Because of some DL trips and poor performances by some players, every team is going to have some scrubs come in and play. These are usually going to be guys called up from AAA, but they may also be a mid-season waiver wire claim or FA signing. You have accounted for this in part by projecting playing time for 15 pitchers and 17 position players. But there will be others. A couple injuries here or there and you’ll have playing time for Musser, Lowery, Dawkins, Matranga and god knows who else.
So, since your total PA’s and IP reflect what should be expected from a MLB team over a 162 game season, I think you need to subtract some of that PT from the current players and add in a category for Replacement Players in both the hitter and pitcher categories. How much? That’s difficult. Last year, for the pitchers, the guys in the category I’m thinking of were Fulchino, HoRam, Wells, Lowery, Rosa, Nomo, Newman and Musser (and those are the guys who got fewer ip than the top 16 pitchers for the Royals). Those eight pitchers totaled 68 ip.
For position players, I think your 17 players cover more of this scrub territory. But there were still 50-60 scrub and Sept. call-up PA’s from guys like Jason Smith and Ka’aihue. And then there’s maybe 30 PA’s for pitchers. Of course, to do an even more serious estimate of the PT a team’s scrubs should get, one could look at even more seasons for the Royals and other teams to see how much PT this group of players usually gets.
Now, if you want to be anal enough to include these things, you could either make room for these additions by decreasing the PA’s and IP across the board on a percentage basis. Or you could manually do it by decreasing the worse players’ PT at a greater rate than the better players, as they are more likely to be losing PT to the scrubs in question.
Did the above make sense? It’s late.
The immoderate moderator
it does make sense -- I still need to figure it out
In a way, I don’t need to “add” replacement players, since as replacement players they don’t add any marginal wins. I just need to decrease the numbers of PAs for the regulars, since they will be “replaced” anyway
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 10, 2009 11:54 AM EST up reply actions
no wait,you're right
the outs won’t meet the mins, although it will still work.
crap
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 10, 2009 11:59 AM EST up reply actions
Great work.
Inflate that shit though. Find a way to tease me with 95 and talk me into it, then I’ll be convinced.
Great work d_f
now I don’t have to watch the games. Whew ! I didn’t know how I was gonna fit those in with my kids baseball this summer. You’ve relieved alot of worries where do I send the check?
It takes 46 muscles to frown but only 4 to flip 'em the bird.
I know, I’m going to cancel my cable subscription, too
Send the check to
devil_fingers
665 1/2, PoopVille Rd.
the Sue, ON G6G D0D
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 10, 2009 12:00 PM EST up reply actions
thanks, although you just reminded me about the real offseason: suicide now
Seriously, I know he wanted to go to a winner, but geez, surely 2/20 could have got Pat Burrell (I know it’s too much, but you see my point), have him share 1B/DH with Butler, and give the Royals and chance at contending in a weak AL Central with a “breakout” or two
Oh well, at least Farnsworth will kick someone’s ass after he give up 3 jacks in one inning to the White Sox
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 10, 2009 11:57 AM EST up reply actions
Yes, I understand that projections don't always translate into reality...
But, damn, this shit is saddening.
"If I Ran the Zoo"
Who would take your place in the Zoo projections at midseason when JoGui put you in a coma?
It takes 46 muscles to frown but only 4 to flip 'em the bird.
You and NYRoyal will co-GM
with “Joel” as executive vice-president
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 2:43 AM EST up reply actions
I ran the win probabilities for you DF
This is with a 51 win %
Wins %
115 0.012844914
110 0.070766467
105 0.612564028
100 2.009307973
95 8.41392181
90 17.59129201
85 39.60648754
80 57.16492048
75 80.1443306
70 90.17773467
65 97.51074744
60 99.20483497
55 99.90013802
2% shot at 100, yay!
thanks, RJ
Maybe I’ll post the chart
However, I’m sort of mad at you for stealing my thunder when I impressed everyone here with something else I “borrowed” from BtBS or some other place.
The illusion of my intelligience takes another shot.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 2:44 AM EST up reply actions
HIs defense and the positional adjustment help a lot
The immoderate moderator
by Scott McKinney on Jan 11, 2009 2:26 AM EST up reply actions
Wouldn't be surprised
if 4 of the projected 5 starters outperform their projections.
Updated
Now includes a Standing Pat tab. See the explanation added at the end of the above fanpost.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
Yikes
Standing pat = 80 wins
Current team = 79.3 wins
So it appears that Moore’s moves haven’t even minimally improved the team? Ugly.
The immoderate moderator
by Scott McKinney on Jan 11, 2009 3:59 PM EST up reply actions
a lot of that can be explained
by the “Standing Pat” projection not factoring in Gload’s positional adjustment and limiting him to only 150 AB’s. Gload would likely cut a bigger slice out of Kila’s PA’s.
I have to critique the "standing pat" PT estimates a bit
I think the key issue with the accuracy of the “standing pat” projections is the PT and reliability of projections for Kila/Shealy/Gload. First, I think if the Royals stood pat, Gload would have gotten more playing time and Kila would have ended up getting significantly less. I also think that the projection for Kila (based on little MLB data and perhaps giving too much weight to one uncharacteristic minor league season) aren’t very reliable. With more PT for Gload, less for Kila and going by the 40th percentile projection for Kila (which I think is fair), the “standing pat” projection moves down closer to 79 wins.
But the overall point stands. Any net improvement Moore has made to this team is likely minimal.
The immoderate moderator
Ooopppss
thanks, kddc, for catching the lack of positional adjustment for Ross Gload… also forgot it for TPJ, though, so the real loss is about .1 wins.
As for playing time, I left it as close to the other as possible, and was generous to Hillman/Moore in assuing they’d recognize horrible performance. I’m not even sure TPJ and Gload would have been on the team at all.
As for using Kila and/or Brayan Pena’s 40th percentile (and BP gets 50 ABs in both versions) — I used the 50th percentile on all of them, and in any case, I use the 50th percentile on the projections for Moore’s version, too. The point is that I didn’t do any particular “favors” for the “Standing Pat” sheet that I didn’t do elsewhere/ The 2009 Royals could have been worse if I’d assuing — and there’s some justification — that Moore and Hillman would be stupid enough to have Farnsworth keep setting up even when his “true talent” shone through in line with the CHONE projections, but I assume that they would [eventually] figure it out.
Anyway, once I made the positional adj. corrections (reflected), now in the spreadsheet, just for giggles I eliminated Kila entirely and gave Gload the bulk of the PAs at first (thus reducing Shealy’s PT), and also also had TPJ and Aviles share the SS job. Guess what? it takes the wins all the way down to 78. Even Hillman and Moore aren’t that stupid (knock on wood), but it even that team is abot a win worse than the one I constructed.
Heck, I wouldn’t have picked up Olivo’s option and given Buck the job again, and used Pena as the backup. Even with BP’s 40th percentile, that comes out as around as good or better than with Olivo, without checking it out.
In any case, I think “Standing Pat” (now with proper positional adjustments, but still better than the 2009 projections!) isn’t given any advantages I didn’t give the other projecitons. Any differences are relatively trivial. Meh… not worth arguing about
If you guys want to know how awesome the team would be with Jeff Francoeur instead of Mark Teahen, let me know.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
"thanks, KCDC"
sorry
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 5:40 PM EST up reply actions
What do you get if you assume Jacobs and Shealy platoon?
They likely will to a certain extent. Jacobs’s PA’s would come down a bit, but his OPS would jump 50 points.
that's sort of what I did
although not very scientifically. If I read up on it soon, I’ll make alteration. While Jacobs rate of performance would increase, his ABs might decrease, as well, so the net gain isn’t as much as one would thing ( and we’ll assuming that Shealys weird splits will, as for almost all righties, normalize give a larger sample size).
I doubt there would be even a win’s difference, though.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 11:14 PM EST up reply actions
I don't know how to find splits for wOBA
but Jacobs should OPS around .850 against RHP. I really don’t know, but I’d have thought that would translate to higher than a .340 wOBA.
One more thing, d_f
They don’t play the game on paper or in a computer. so I found this entire exercise and insult to Abner Doubleday, Henry Chadwick, Harry Wright and Cap Anson.
The immoderate moderator
and where is the grit factor?
surely that must account for 5-6 wins?
The concept of progress acts as a protective mechanism to shield us from the terrors of the future. - Collected sayings of Muad'Dib
Actually it is only the grit that gets you the wins
Everything else is just about padding personal stats
The immoderate moderator
by Scott McKinney on Jan 11, 2009 7:24 PM EST up reply actions
uh oh
you probably shouldn’t tell those guys I spent the last two hours trying to do a cool win probabilities chart, but am too stupid to figure it out without some nice people at BtBS holding my hand
devil_fingers: awesome sabermagician
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 7:33 PM EST up reply actions
Maybe the next revolution in sabermetrics will be the use of magic, voodoo, the occult and various dark arts
That could make for one hell of a projection system.
The immoderate moderator
by Scott McKinney on Jan 11, 2009 8:13 PM EST up reply actions
if that's the case, along with pimpbot 5000 as manager, I think we know who the next GM should be
Yeah, that stuff I put out in the press about not trading Teahen?
ILLUUUUUUUUSION MICHAEL!
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 8:27 PM EST up reply actions
that's how I draft my sucky fantasy baseball teams....
The concept of progress acts as a protective mechanism to shield us from the terrors of the future. - Collected sayings of Muad'Dib
yeah... .it's weird
what do you need to do?
Although it won’t give you the “percentiles,” you can downlaod the entire projections for hitters and pitchers as a csv file for use in Excel, Access, or MySQL, etc.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 9:36 PM EST up reply actions
Yet another update
added a chart showing win probabilities… you can browse to a larger version by clicking here
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
I mean
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 11, 2009 11:07 PM EST up reply actions
updated, just once more
flattery will get you everywhere, Warden11… here’s the probabilities graph for the “bumped” projection.
Some of the the “random” win numbers weren’t so random but where win #s taken from very good/bad seasons in Royals history.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
Excellent work, d_f
I just wish it was more than assembling random stats to “prove” your point.
The immoderate moderator
An almost 1 in 4 chance of having 86 wins
Moore has gotten us close to being in contention. What more could Royals fans possibly ask of him?
I love, love, love the new sarcasm font.
The immoderate moderator
Impressive but..
Lets put this analysis in perspective. I doubt that had you ran this formula of numbers for the Tamp Bay Rays before the start of last year you would have concluded they would win the AL East. Not saying it’s not a reasonable barometer, but such statistical scholarship shouldn’t be used to justify all the gloom and doom I have been reading on this blog lately.
Honestly, given that bullpen production should be about the same and we added a good CF and a 30HR guy, plus Bloomquist being godd for speed off the bench and giving us insurance on Callaspo, I really think all Royals fans should be excited about next year. Hope is supposed to be eternal in the preseason is it not?
Lets put this analysis in perspective. I doubt that had you ran this formula of numbers for the Tamp Bay Rays before the start of last year you would have concluded they would win the AL East.
Not true. Similar stat-based models ran before last season had the Rays with a high win total, certainly well into contention. Baseball Prospectus predicted that kind of success for the Rays using their PECOTA projections.
Not saying it’s not a reasonable barometer, but such statistical scholarship shouldn’t be used to justify all the gloom and doom I have been reading on this blog lately.
Do stats and predictions tell us a lot about how good players are and should be in the near future? If so, then we can fairly evaluate the moves Moore has made and much or how little his moves have improved the team.
Honestly, given that bullpen production should be about the same
Really, why? I’ve read a few “things aren’t so bad” posts in which this kind of statement has been made without any support or even explanation. Please tell me why a bullpen with key elements like RamRam and Nunez gone and below average pitchers like Farnsworth, HoRam and Waechter added, and with another key element (Mahay) turning 38 should be “about the same”? It shouldn’t be about the same. It should be more than a little worse. The only question is how much worse.
we added a good CF
By any measure, we added a league average CF.
and a 30HR guy
I wish that one counting stat were all that was important.
Hope is supposed to be eternal in the preseason is it not?
There’s a difference between hope and reality. Reality is what is. Hope is wishing things will be different from reality. Hope is great. I hope the Royals will be good. I hope the past stats of these guys means nothing and they’ll all suddenly be much better. I hope the Royals win 90 games. But at the same time, I recognize reality. One can recognize and describe reality and at the same time hope for something better. That’s what being a Royals fan is for me.
The immoderate moderator
by Scott McKinney on Jan 12, 2009 5:55 AM EST up reply actions
our bullpen production will be roughly the same?
Ramirez who was almost as good as Soria turned into Farnsworth….Nunez who was really good for most of the season turned into doug waechter….those are the guys pitching most of our high leverage innings….oh…and they cost almost $5 million extra….AWESOME
TPJ...you're dead to me
by billybeingbilly on Jan 12, 2009 11:31 PM EST up reply actions
now if Hillman
were to use Soria in a less conventional manner, it wouldn’t be a problem.
I don’t expect that to happen, but I’m always optimistic.
ha.
Replies to NYRoyal
First I will give credit that the formula did predict the Rays to do well, honestly I did not know that, good point. However I will still respectfully disagree with about everthing else, see as follows:
NY Royal said “By any measure, we added a league average CF.”
My Reply: Maybe, but offensively and defensively he is better than Teahan so that means the outfield as a whole has been upgaded. Guillen would be in the outfield with or without CoCo and as bad as Guillen is, when they tried Teahan in LF his defensive was even worse. Teahan may have defensive superiority in right over Gullen but it is no where close to making up the difference in offensive production between them. So logic concludes that this should be viewed as a nice upgrade, giving us better production and defense.
NY Royal said the “bullpen will be worse”
My Reply: In my estimation you over estimate Nunez as he did get rocked more than a few times in his career as a Royal, but I will concede the loss of RamRam hurt. But Farnsworth has a legit arm and Wachter has put up some decent numbers, plus it looks like Brandon Duckworth will be a solid contributor for the full year. Mahay may be a bit long in the tooth but Bale was perfect in a relief role last year, and presuming he doesn’t hit any more walls we should get a good full year from him. Hence, all that considered I gurantee you the production of last year’s pen to this years pen will be at least as good, if not better, and through trades we got a legit 30HR hitter and an upgraded outfield as a result.
NY Royal effectively said “If only HRs were the only stat to be concerned with with Jacobs”
Mt reply: If he hits 30 HR I will take that over Gloads putrid for a first baseman offensive numbers even with a few more errors. Remember he is stll learning the position. I do hope his attitide towards taking walks improves, on that I will agree with you.
NY Royals said “There is hope and there is reality”
You should have hope because each year Dayton has made us better and almost instantaniously took us from a laughing stock to respectable. With all the rukus you have made over Daytons trades, I gurantee you that if we do not contend this year it will be due to the starting rotation not being as effective as we need. My biggest concerns are that Bannister won’t rebound, Davies might regress, and HoRam and Hochaver give us more sub-par numbers. But really, what could Dayton have done to improve the starting rotation in this years free agent market?
Anyway bottom line for me is I have a lot of respect for Dayton’s getting us from a team that was laughing stock that people made fun of and said should fold to bringing the Royals back to feeling like they matter. I doubt Dayton ever reads these things but if he does I want him to know that at least one fan here really believes in what he is doing and thanks you for the effort. I think Dayton is doing a great job!
Maybe, but offensively and defensively he is better than Teahan
Defensively, yes. Offensively, certainly not. But yes with him in CF and DeJesus in LF the OF has been upgraded a bit. I said at the time this was a mildly positive move. Not that the Royals should have gone for a new CFer, but there you go. But you need to realize that the numbers show this to be a very small upgrade for the OF overall (unless Crisp is about to have some kind of unexpected breakout).
In my estimation you over estimate Nunez as he did get rocked more than a few times in his career as a Royal, but I will concede the loss of RamRam hurt
When Nunez was rushed to the majors at age 21 (2005), he got rocked. He wasn’t ready. But in 2007 and 2008, he was very good. Good ERA, good peripherals. Not great, but good.
But Farnsworth has a legit arm
Lots of pitchers have “legit arms.” Leo Nunez has a legit arm. Nunez’s stuff was and is every bit as good as Farnsworth’s. The difference is that Nunez actually has good results to go along with it. What’s the use of Farnsworth’s “legit arm” if it doesn’t translate into good results on the field? Farnsworth has been shitty for the last two years. He has big control problems, gives up way too many walks and gives up flyballs, and thus HR’s, like crazy. That’s what Farnsy’s arm does for the teams he’s been on.
Wachter has put up some decent numbers
Not many. He might make for a decent reliever. Maybe. And signing him for $640K is fine. Just don’t expect him to slot in and replace the production of genuinely good pitchers who have departed.
plus it looks like Brandon Duckworth will be a solid contributor for the full year.
Wow, why? His unimpressive track record, unimpressive stuff and unimpressive control leave me….unimpressed. He hasn’t been a solid contributor to a major league team for quite a while. That is, unless when you say “solid contributor” you mean “below average.”
. Mahay may be a bit long in the tooth but Bale was perfect in a relief role last year, and presuming he doesn’t hit any more walls we should get a good full year from him.
I like Bale well enough, but… He lost significant time due to injury in both 2007 and 2008 (that includes an arm injury in 2008 before he hit that wall – he was actually on a rehab assignment for an existing injury when he did that). So why should we expect a full year from him in 2009?
Hence, all that considered I gurantee you the production of last year’s pen to this years pen will be at least as good, if not better,
You’re kidding, right? Compare these two bullpens:
Soria
RamRam
Mahay
Nunez
Peralta
Gobble
Bale (half season)
HoRam (half season)
Soria
Farnsworth
Mahay
Waechter
Peralta
Gobble
Bale (partial season)
Duckworth (partial season)
You can guarantee me that the second group is as good as or better than the first group? Why? Because Farnsworth’s arm is “legit” despite his actual performance on the field? Because Waechter has had “some decent numbers”? Because Duckworth is going to put up a solid season? Because Bale is going to have a good full season? You’ve got nothing to base this on but wishcasting. Look at what has been subtracted and what has been added:
Farnsworth
HoRam
Waechter
RamRam
Nunez
Please tell me you understand that losing the latter two and adding the former three is a net loss in talent. There is absolutely no reason to believe that the 2009 bullpen will be as good as the 2008 bullpen. None.
If he hits 30 HR I will take that over Gloads putrid for a first baseman offensive numbers even with a few more errors.
The fact that Jacobs is likely better than Gload isn’t the issue (for the millionth time this has been said on this site). If that’s the only issue, then the Royals could have just slid Shealy into that spot. I mean, he’s a lock to do better in 2009 than Gload did in 2008. The point is that Moore gave up Nunez and $3.5M for Jacobs. He could have gotten that kind of production for less money on the FA market (Branyan). Or he could have pulled off a better trade (Willingham). Or he could have used the numerous existing options, which includes players the Royals really need to evaluate for the future (Shealy, Ka’aihue). The mere fact that Jacobs is going to be better than Gload was is meaningless. Several dozen options were better than Gload. Of those options, Moore chose poorly.
Remember he is stll learning the position.
He’s been playing first base for four years. And, for what it’s worth, his defensive stats over that period have gotten worse every year, not better. His problem isn’t experience; it is physical gifts. He doesn’t have the agility/speed/quickness to give him any kind of range. He fields his position as if both of his feet were buried in the dirt.
With all the rukus you have made over Daytons trades, I gurantee you that if we do not contend this year it will be due to the starting rotation not being as effective as we need.
Another guarantee. Good lord. The rotation is literally the last problem on this team. The rotation is better than the offense, the defense and the bullpen. Hell, it is the team’s major strength. This team is going to have one hell of a difficult time scoring runs. Moore did nothing to improve the team’s OBP. In fact, I think his moves decreased likely team OBP (a decrease from what it likely would have been if he stood pat this offseason). How is this team going to score if people aren’t getting on base? Jacobs 25 HR’s? Please.
But really, what could Dayton have done to improve the starting rotation in this years free agent market?
No, there were a half dozen other things Moore could and should have addressed this offseason…and didn’t.
Anyway bottom line for me is I have a lot of respect for Dayton’s getting us from a team that was laughing stock that people made fun of and said should fold to bringing the Royals back to feeling like they matter
Wow, you give him a lot of respect for getting the Royals up to the lofty heights of 75 wins. That’s fine. And I think the Royals will get up to 80ish wins this year. And I’m sure you’ll heap more praise on him for getting the Royals into mediocrity. But a mediocre GM can get his team to mediocrity. And mediocrity is only ok if it is just a brief stop on your way to contention. And for a small market team, it takes a genuinely good GM to get you into contention. Also, while the Royals have improved their record under Moore’s tenure, he doesn’t deserve credit for all of that. He doesn’t get credit for good play from DeJesus, Gordon, Greinke or anyone else he didn’t acquire. So what has Moore done to get more wins for the major league team? Meche and Soria. Perhaps re-signing Grudz (although I don’t think that was smart for the Royals long-term). Some good bullpen moves. That’s about it. Let’s not pretend that Moore’s moves account for all of the improvement in the Royals record.
I think Dayton is doing a great job!
Yikes.
The immoderate moderator
by Scott McKinney on Jan 12, 2009 9:31 AM EST up reply actions
Excellent work
I’m not sure if you are in the “Soria for Starter” camp, but I’d be curious to see how your “zoo” would come out if you replaced Davies with Soria and say Waechter with Davies and slid everyone else up?? Looks like a couple of wins, eh?
We always did feel the same, We just saw it from a different point of view, Tangled up in blue.
-Bob Dylan
Thanks
really, it’s Sky, CHONE, and my laptop doing all the work
I am/was in the Soria to starter camp (with the usual qualifications about “if they think he can stay healthy” and doing it safely and stuff), and I do think it would add some wins in general. However, I’m not comfortable enough with figuring out what his numbers would be like vis-a-vis CHONE’s projections for him. I remember TucsonRoyal asking Tango about it, if I’m not mistaken, and got a good general answer. I might look it up later if I have time (I’m feeling guilty already about not do a more detailed “what if Jacobs was platooned” post) and see what i can come up with.
Of course… let’s see if I can do this without being a Debbie Downer… with the current bullpen, rather than having Ram-Ram and Nunez being thrust into higher leverage roles, you’d have Farnsworth, HoRam, and Waechter… given the “multiplicative” effect of leverage, that’s much, much worse than having RamRam or Nunez in there (I’m actually an holding out hope for Rosa [if he ends up in the ’pen] and Tejada to be the new RamRod and Nunez by the break…), so it would be curious to see how it would balance out.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 12, 2009 11:15 AM EST up reply actions
Yeah, the problem is estimating Soria's stats as a starter
…which of course wouldn’t be as good as they are as a reliever. Adding a run per 9 innings to his ERA or RA would be fair.
The immoderate moderator
by Scott McKinney on Jan 12, 2009 11:21 AM EST up reply actions
Reply to NYRoyal
First let me say I am enjoying this discussion. I do want to respond to a few things you said:
NYRoyal effectively said: “Teahan is better offiensively than CoCo.”
My Reply: While I think you are very learned, this is the first of two occasions where you “jump the shark.” Teahan in 572 ABs had 15HRs 59RBIs and a .255 avg. CoCo, in only 361 ABs had 7HRs and 41 RBIs and batted ,283. CoCo out stole Teahan 20 to 4 and despite 211 more ABs, Teahan had only 11 more walks. When you look at the number of ABs differential, CoCo outperforms Teahan in nearly every meaningful category. I love Mark Teahan too my friend, but the numbers don’t lie and that postion is indefensible.
NYRoyal said “The fact that Jacobs will be better than Gload is meaningless”
My reply: 30Hrs, especially coupled with another 20plus from Guilen would be anything but meaningless.
NY Royal said: " Duckworth is unimpressive"
My reply: I guess you wern’t watching in September when as the 5th starter he did really well, but didn’t get many wins due to almost zero run support. I guess you also forget the decent year he had for us in 2007.
NY Royal said “The starting rotation is the least of the Royals worries”
My reply: This my friend is where you really jump the shark. We have no credible left handed starter, our #3 guy Bannister’s ERA was 5,76 last year, our #5 Hochaver finished at 5.51, and our #4 Davies would have been hovering in the high 4.00s but for a truly brilliant month of September. I am pretty optimistic about Davies continuing to perform well, but with no Left Hander and 2 starters in the mid to upper 5.00s in ERA, the Royals rotation is not nearly where it needs to be. Try not to let Jaws nip those skies as you fly over flaunting eloquently about the solidity of the Royals starting rotation.
BTW for me this is all in fun and I admire your passion for the Royals, it mirrors my own, even if we are on opposite ends of the spectrum with our analysis of Dayton Moore.
Teahan in 572 ABs… CoCo, in only 361 ABs…
You should never evaluate any player(s) by looking at only one year of stats. Compare their last three years each. Compare their projections. Teahen wins. Also, evaluating players by using counting stats like HR’s and RBI is a big mistake.
I love Mark Teahan too my friend, but the numbers don’t lie and that postion is indefensible.
Indefensible? You’re quoting me RBI’s and yet my position is indefensible. Teahen has been better overall over the last three years. That’s why every projection system projects Teahen to hit better than Crisp in 2009. Are those projections indefensible too? Now, that is just the hitting numbers. If you take into account defense and position adjustment, Crisp gets the edge by a little.
My reply: 30Hrs, especially coupled with another 20plus from Guilen would be anything but meaningless.
You’ve really got to get off of the counting stats. Rate stats are your friend.
My reply: I guess you wern’t watching in September when as the 5th starter he did really well, but didn’t get many wins due to almost zero run support. I guess you also forget the decent year he had for us in 2007.
First, his 2007 and 2008 numbers add up to a whopping 84.2 ip. Second, they weren’t impressive innings. What was it that impressed you about those 86 innings? His 4.60 ERA? His 41 K’s and 42 BB’s? He gets some groundballs, so he’s not useless. He’s also not very useful.
This my friend is where you really jump the shark. We have no credible left handed starter
Where do you get this stuff? Who cares if the Royals don’t have a LH starter? What makes you thinking having a LH starter is important?
our #3 guy Bannister’s ERA was 5,76 last year,
And you’re look at only one season of stats again.
our #5 Hochaver finished at 5.51
One season of stats from a rookie who is a very good prospect and who has a statistical profile similar to Chien-Ming Wang and other extreme sinkerballers. Peripherals count. GB% counts.
Try not to let Jaws nip those skies as you fly over flaunting eloquently about the solidity of the Royals starting rotation.
Dan, I’ve got a tip for you. You should really only condescend and ridicule others when your statistical analysis is at least decent. Your reliance on one year of stats and things like RBI, ERA and small sample sizes is weak. Using that to condescend to someone using better and more meaningful stats makes you look foolish.
The immoderate moderator
by Scott McKinney on Jan 12, 2009 8:18 PM EST up reply actions
My thrid reply to NYRoyal
First I was not trying to be condescending, toungue in cheek funny perhaps.
Secondly in praising Teahan and your precious “three year projections” you woefully overlook that Teahan has went way way down over his first season over that time frame and that would really sque the numbers. Teahan has woefully undeachieved since his first season and that’s the reason the Royals couldn’t get anything for him in the trade market. He’s an average outfielder with below avg. batting stats for a RF. Nuff said.
Counting stats vs. Rate states, baseball games are won and lost based on ERA, hits, HR, RBIs, Walks, Stolen Bases and Errors. They call them count states for one reason, because they count!
Speaking of which, I cannot believe you think the Royals rotation is great with two guys last year in the high 5.00s? They have promise yes, but then again I didn’t think you put much value in “hope” and were all stuck in your perecption of reality. Thanks to Dayton my friend, you will find this season is truly a season of hope!
Luke was unlucky
Really unlucky. I can’t remember off the top of my head but his FIP was something like a run below his ERA. He wasn’t great, but it was a solid rookie season, and there is nothing to suggest he won’t continue to improve.
The Allard Baird of incisive internet discourse.
relevant post on pitcher evaluation coming soon
Hochevar was no great shakes, but the Royals crappy D did him no favors when you look at his FIP and tRA… he’s at least a #4 this past year (when he played), and possible #3, depending on if his RA normalizes.
Somehwat related post going up later tonight on driveline (I hope).
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 12, 2009 8:56 PM EST up reply actions
I think baseball games are won by IBB's, HBP's, sac bunts and caught stealing
They count those stats too.
Speaking of which, I cannot believe you think the Royals rotation is great with two guys last year in the high 5.00s? They have promise yes, but then again I didn’t think you put much value in "hope" and were all stuck in your perecption of reality
Did I say that I think the Royals rotation is great? No, I said that it is better than the offense, defense and bullpen. And I think the stats back me up. Greinke and Meche are excellent pitchers. Greinke is an ace and Meche is a very, very good #2. Hochevar profiles as at least a pretty good #3 SP. I don’t know that he’ll get there in 2009, but he should get there. Perhaps even better. Again, Hochevar’s statistical profile compares favorably to some good extreme sinkerballers/groundball pitchers. Bannister and Davies are both iffy. An average MLB rotation has iffy pitchers at #4 and #5. Both could turn out to be decent…or not. If Bannister can pitch as an average #4 SP and eat 180+ innings, he’s a definite asset. That’s doable. Davies is a question mark, but I’m eager to see what he can do in 2009. And I also think Rosa would make a good MLB starter based on his minor league performance.
The immoderate moderator
by Scott McKinney on Jan 12, 2009 9:01 PM EST up reply actions
I get all the milb guys confused
are there any minor league starters who look like they have a shot of making it to Omaha this year and perhaps seeing time pre-September in 2010, or, Xenu willing, 2009?
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 12, 2009 9:29 PM EST up reply actions
Cortes and Rosa are the only guys that could probably make that move.
If Cortes can get his off speed stuff under control he could get a look Sept of this year. While Rosa has a shot to start in the bullpen this year and if he can develop his offspeed stuff he might get a switch to the rotation. The only other guy I could see making the move to the big leagues as a solid starter type would be Blake Wood. I think he starts in AA this year but he definitely has the size, stuff and the pitches to be a big league starter. He could move fast if he starts to command his stuff more consistently. The other guy that I like to make some spot starts would be Dusty Hughes. He’s a lefty and ended the season in Omaha last year and while his stuff is average those two factors give him a leg up if KC struggles with the #5-6 starter this season.
It takes 46 muscles to frown but only 4 to flip 'em the bird.
how old is Cortes?
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 12, 2009 11:07 PM EST up reply actions
and thanks, I was hoping a number of you MiLB guys would answer
I get most all of my MiLB info from ya’ll
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 12, 2009 11:08 PM EST up reply actions
Cortes will be 22 in March, Rosa 25 in Sept, Wood 24 in August. All three have the same problem consistency with their offspeed stuff. All three have plus fastballs topping around 96 mph. If I had to bet I would say Rosa stays in the bullpen while Cortes and Wood get a better chance to stay as starters just based on Rosa’s injury history. Rosa would be flat sick out of the pen probably touching 97-98(max speed last year according to pitch f/x 96.5) with a flat nasty slider (if he ever trust it).
It takes 46 muscles to frown but only 4 to flip 'em the bird.
While Cortes does have issues with his third and fourth pitches, his biggest problem is control. And that is what differentiates Rosa from him. Rosa has very good stuff and very good control. Of course he’s also older and has injury/durability issues. So Cortes is the better prospect, but Rosa has a good SP skill set. But he may never be a 150 ip per year guy.
The immoderate moderator
by Scott McKinney on Jan 12, 2009 11:28 PM EST up reply actions
so....is your likely/hopeful scenario for Rosa...
that he’ll turn into Sorias replacement, freeing him up for the rotation? That seems to make the most sense to me. Of course, Rosa will probably be lights out this year and be traded for Greg Zaun next offseason
TPJ...you're dead to me
by billybeingbilly on Jan 12, 2009 11:42 PM EST up reply actions
Rosa's scenarios
What I’d do: Make him the Royals swingman/6th starter in 2009, making sure his innings don’t go over around 110 last year (due to his injury and ip last year). Then, if healthy for all of 2009, I give him a shot at the 2010 rotation. In short, I’d continue to try to develop him as a starter. I don’t know if he has the durability to be a full season MLB starter, but I wouldn’t give up on it yet.
I’d also move Soria to the rotation, but I don’t think Rosa needs to necessarily be the new closer. Quite frankly, I’m not worried about who the closer is in 2009. I’d rather build the best rotation for 2010 and beyond. I’d let the bullpen figure itself out. If that means having no good closer for 2009, so be it. Then next offseason Moore can rebuild the pen and hopefully he’ll do it more like 2007 and 2008 than what he did for 2009.
If Rosa doesn’t pan out as a starter or can’t stay healthy enough to handle that kind of workload, then I’d use him as a high leverage reliever as either a closer or 8th inning setup man.
What’s most likely: I’m not sure, but I think he’ll start the season Omaha, probably in the rotation. He’d probably be one of the first pitchers called up when necessary, and then probably used in the bullpen. My feeling is that Moore views the SP depth chart as:
Greinke
Meche
Hochevar
Bannister
Davies
HoRam
Duckworth
Rosa
The immoderate moderator
by Scott McKinney on Jan 12, 2009 11:51 PM EST up reply actions
Have you considered?
I like your thinking but what about Tejeda. I know it was only one start last year but there were some things to build on in that start. His average fastball that night was 95.14 mph with a top end speed of 98.4 (decent movement) that has the makings of some impressive stuff if McClure can get him to trust his offspeed stuff more he could be a decent 5th-6th starter. Playing in the K with a faster outfield this year could give him the confidence to succeed.
I would use Rosa in a similar way as Tejeda with a couple more spot starts towards the end of the year.
It takes 46 muscles to frown but only 4 to flip 'em the bird.
I really think that Tejeda’s stuff, pitches and control make him best suited for relief. The control is the big thing. 5.01 BB/9 for his career and 4.73 BB/9 last year. But extremely impressive stuff. I think he’s the kind of guy who might be a decent #5, but could be a very good setup man, or maybe even a closer. I think you get max value from him in the bullpen.
Now, I want McClure to keep working with him and if he can figure out something which will improve Tejeda’s control, that could change things. I just haven’t seen any control improvement from him yet.
The immoderate moderator
by Scott McKinney on Jan 13, 2009 12:23 AM EST up reply actions
I would probably go
Greinke
Meche
Hoch
Davies
Bannister
Tejeda
HoRam
Duckworth
Rosa end of the year
It takes 46 muscles to frown but only 4 to flip 'em the bird.
As ya'll know, I'm a stats rook, still reading BB btwn the Numbers
But, was curious what any of you think about the Rs going to a 4 man rotation. That was pretty normal back when I was a 70s kid. It seems like even then they used to go to a 5 man rotation later in the year.
Look, I know that Hillman/DM would NEVER be the ones to do something so “unconventional” but my question really is about whether we have the staff to do it and whether it would be a good thing for us to do.
If we did, we would have more resources for the pen where some people feel we’ve regressed this offseason or for an additional position man.
meat
Rany did a lot of interesting work on that
And of course how well that would work and how risky it is is all speculative. If this were done, Hillman would have to monitor pitch counts more closely and put a hard ceiling on every pitcher for every start. I think this kind of thing would work better if pitchers had been developed throughout their professional careers to pitch like that. Instead, they’ve all pitched in 5-man rotations throughout their careers. Their bodies are used to those kinds of innings for a season and that kind of rest. I think the change would be risky. If I were a GM/manager, I’d try a lot of different, unconventional things but I don’t know if I’d take this risk.
For 2009, I’d definitely use a 5-man rotation because the Royals need to take a look at a number of starters. In addition to the obvious Greinke-Meche-Hochevar, I’d like to see 30 starts each for Davies and Bannister (at least 20). The Royals need to figure out if those guys should be a part of the Royals rotation beyond 2009.
Beyond that, a rotation of Greinke, Meche, Hochevar and Soria would be amazing. But probably not a good idea.
The immoderate moderator
by Scott McKinney on Jan 13, 2009 12:38 AM EST up reply actions
KC isn't heading in that direction
but it sounds like Nolan Ryan and the Texas Rangers could be. He is getting rid of pitch counts from what I heard. This type of move would definitely have to take place in the minors.
It takes 46 muscles to frown but only 4 to flip 'em the bird.
Nolan Ryan to his minor league pitching coaches
“I want to see a 1:1 K:BB ration or you’ll gone~!”
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 13, 2009 1:13 AM EST up reply actions
When I hear about his theories that pitchers can throw a lot more pitches in a game and more innings in a sason
I can’t help but think that he’s thinking that because he was an iron man who threw 5300 MLB innings through age 46, then anyone can. That seems pretty dangerous to me. We’ll see what happens. If the Rangers are truly moving forward with this kind of thing organization-wide, I wonder how they’ll react to the first or second torn labrum or TJ surgery in the organization.
The immoderate moderator
by Scott McKinney on Jan 13, 2009 1:19 AM EST up reply actions
I'm fine with it too
If it ruins some of their prospects, I’m not going to lose much sleep. I’m not saying this will definitely fail for them, but there’s a lot of risk. I’d much rather the Rangers incur that risk than for the Royals to risk Cortes, Duffy, Rosa, Wood, Montgomery, Melville, etc.
I think throwing out pitch counts at any level is a bad idea. I think doing it for 18-22 year old pitchers is reckless at best, and organizational suicide at worst.
The immoderate moderator
by Scott McKinney on Jan 13, 2009 1:27 AM EST up reply actions
Is that you, Dan Glass??
We always did feel the same, We just saw it from a different point of view, Tangled up in blue.
-Bob Dylan
by Royal Kingdom on Jan 13, 2009 10:09 AM EST up reply actions
COMPLETELY unrelated to this thread...
but I really really HATE Boston…b/w Gammons’ ridiculous justification for Rice in the Hall and KG and Paul Pierce being really really annoying in a post game press conference….its just Boston obnoxiousness overload…
TPJ...you're dead to me
by billybeingbilly on Jan 12, 2009 11:28 PM EST reply actions
my Boston obnoxiousness meter
broke a few years ago. Now they’re really like the teacher in Charlie Brown.
heh
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 13, 2009 11:53 AM EST up reply actions
I missed Gammons' defending Rice's induction
How bad was it?
Stathead, Zack Greinke fan, and Rock Band 2 singer extraordinaire.
I need to read it, too
For the millionth time, I think “big time” writers need to just take a pass on talking about “their” teams. I like Gammons, but his annoying on the Sox.
I know this might get me banned, but I like Neyer, too, when he isn’t talking about the Royals.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 13, 2009 2:49 PM EST up reply actions
Neyer is very down to earth,
and his analysis isn’t bad when he gets off the perpetually negative track. Good luck having both things happen when he writes on the Royals. I know this off-season has been a disappointment, but—as you’ll see in my article I’m working on—this doesn’t mean we’ll “never win.”
Gammons has a lifetime pass from me….but he’s very silly about the Sox these days.
Stathead, Zack Greinke fan, and Rock Band 2 singer extraordinaire.
I don't think it means we'll never win, either
but it does mean that winning soon is much less likely
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 13, 2009 10:12 PM EST up reply actions
Just catching up
but very nice work on this.
thanks
you do realize much of my whole pretending to be a wanna-be sabermetrics guy is due in large part to you turning me on to insidethebook, statcorner, and stuff.
I hope you have liability insurance.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 13, 2009 2:07 PM EST up reply actions
We added an umbrella policy last time we bought a car
so go to town with the spreadsheets.
Cool.
Why is the chart “bumpy”? It should be a continuous curve with an inflection point.
Beyond the Boxscore // Calling BJ Upton lazy is lazy.
Let me clarify. It's already continuous. I meant "simpler". Fewer changes of slope.
Beyond the Boxscore // Calling BJ Upton lazy is lazy.
Good point
It almost flattens out around 78.5 wins and then dives back down again at about the same slope as before.
The immoderate moderator
by Scott McKinney on Jan 13, 2009 5:18 PM EST up reply actions
that a good question, with a very good answer
…
….
anyone got that good answer? Perhaps I’m doing the wrong chart type in Excel?
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 13, 2009 10:13 PM EST up reply actions
I think the numbers have to be off
I have no idea why they would be though.
The immoderate moderator
by Scott McKinney on Jan 13, 2009 11:53 PM EST up reply actions
I think I'm just doing the chart the wrong way
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 14, 2009 12:04 AM EST up reply actions
Yes, I think the numbers are off.
Beyond the Boxscore // Calling BJ Upton lazy is lazy.
by Sky Kalkman on Jan 14, 2009 12:00 PM EST up reply actions
but why would that throw off the curve?
DO you mean the projection is off? well, maybe, but that wouldn’t explain why the curve isn’t “shaped” correctly… they’re just a series of numbers plotted onto a graph with a “smoothed” line connecting them…
do you mean that I did the binom dist incorrectly? Maybe — I just did the same thing all the other guys did with 1-BINOM
maybe I’m misunderstanding what you mean by the “numbers are off”
I think we need a scouting opinion
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 14, 2009 12:49 PM EST up reply actions
I meant the calculation is off.
Given the 76 win projection (or whatever), the resulting probabilities aren’t right. They should produce a smooth curve. Maybe not all of them are wrong, but at least one is.
Beyond the Boxscore // Calling BJ Upton lazy is lazy.
see my post below
the curve is off because I didn’t have a regular interval between numbers. ZeppelinDZ (resident real stats-guy) helped me out with it. My new ones are fine.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 6:05 PM EST up reply actions
OK
Zeppelin DZ generously looked over it,a nd the problem is the intervals between the numbers on the X-Axis throwing off the “Look”…. so the numbers of the probability oaren’t off, but the visual effect is distorted by how I thought I need to do it…
so I think I can fix it as some point, but right now, I’m back at square one
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 14, 2009 1:16 PM EST up reply actions
Homer
That chart says we can’t lose more than 122 times. I beg to differ!!!!!
/joke
The Allard Baird of incisive internet discourse.
OK
Beyond the Box Score is now doing a set of projections, and I volunteered to be the Royals “conduit,” hosting it here in Fanposts… Sorry if that steps on anyone’s toes. I just figured since I had already done it, knew how to work the spreadsheet, pored over defensive and baserunning stats, etc… it would be Okay
IN any case, I’ve redone the projections with the latest version of the spreadsheet…. While I’m still a bit uncomfortable with the lgwOBA which makes the Royals offense a bit crappier than one might think (more than in the projections above), that something I’ll figure out with the BrBS guys.
Anyway, this is the kind of thing that’s more fun if we get input from the Fan Base (more than just Mel). I especially need help figuring out who the heck is actually going to play this season. But I need people to try to be objective — I’m a homer, too, but we have to try to get a midpoint, and not just assume that Greinke and Meche can be put down for 200 IP each with a Sharpie, or that all projection systems are biased if they don’t have Kila hitting 30 jacks, or don’t see Aviles pulling another 1999 Jeter on us, you know?
Is that cool?
The thing is, this post will be “gone” from the recs section pretty soon. I don’t think we need to keep the discussion going for months, but is the next 10 days enough? I don’t know. Can we “sticker” this one to stay up somewhere for a while, if that’s cool with RR. et. al., or once I get things worked out, should I create a new post? Or will we be sick of this in another 10 days…
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
Maybe do a new fanpost for it?
This post is old enough that people might not even be reading it anymore. And are you looking for a “community projection” for all players? Or just a little tweaking of CHONE or something like that.
The immoderate moderator
by Scott McKinney on Jan 14, 2009 8:05 PM EST up reply actions
A little bit of both
First, we need to come to an informed consensus about who is going to be on the team. For example, I’m not sure this “Grienke” kid is going to make it out of camp…
but seriously, we’re obviously not going to get everything right by ST (when we’ll “set” it), but we can tweak things as we go along.
I would say the first thing that I need the most help with, then, is who people think will be on the team, and what roles they will play. We could even make that fun by having polls, if that isn’t too much.
The projection part is trickier… on one hand, i wish we could get a “community projection,” but then incoporating that with a “system” that assumes certain league averages and stuff, although if Royals Nation, of exampe, wants to send me his complete set of “Dempsey Projections”™ I’d be willing to work up a separate spreadsheet for those when I have the time…
One way would be to average several systems, and if I were doing it all myself, that’s what I would do, but if we want a bunch of input, that actually makes it more complicated.
CHONE is a great system, although choosing between CHONE, ZiPS, and PECOTA is a bit like choosing between Jesus, Buddha, and Tim Tebow. But there’s a couple reasons I think CHONE might for doing something like this with a group of people:
1) It’s all there right now, we don’t have to wait,
2) it’s free, so everyone can look at it, and
3) It’s already set up for wOBA, which makes things tons easier for estimating WAR
4) my idea with how to balance our natural (and enjoyable) homerism with a realistic projection
Now, obviously, people can and should look into other systems to see where CHONE needs to be corrected. But one cool thing about CHONE is that, as far as I know, it’s the only “free” public system that has publishes percentiles for each player. That way, we can “bump” players up and down while remaining within the constraints of the system.
And here’s an idea about the way we could do it. Obviously, it’s not a very good projection to say that half the team is going to hit their 60th, and the other half is “only” going to hit their 50th — that makes for a lot of 88+ win teams. But my idea (and I’m open to suggestions) would be that we could even have “votes” (it wouldn’t have to be based direcltly on CHONE) about whether a certain player in question is going to hit 60th, or 70th, or whatever.
But here’s the thing — I say that after one or two players get bumped, for every other player that gets “bumped,” to 60th or 70th (or whatever), another player on the team that we’ve projected for similar playing time gets bumped down a respective couple of notches.
I dunno, to me it sounds kind of fun and could get people involved, as long as we try to remember we’re at least pretending to be objective and doing our best. As you know, projections need to be closer to “midpoint” to be more accurate.
Does that sound doable and interesting? I’ve done most of the hard work already, in terms of organizing it and stuff. If people want to tweak the baserunning and defense numbers, that’s OK, but I think I’ve got some pretty good baselines there. I’ll probably report my new, slightly more conservative CHONE-based projection on editgrid soon, since google docs doesn’t like BINOMDIST.
What say you (all)?
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 14, 2009 8:24 PM EST up reply actions
well
you also have to get Dick Kaegel on an RR podcast…
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 14, 2009 10:51 PM EST up reply actions
Sounds good to me
I’m sure there will be some difficulties and you’ll have to figure out who gets bumped down because if put to a vote just about every player is going to get bumped up.
The immoderate moderator
by Scott McKinney on Jan 15, 2009 10:40 AM EST up reply actions
yeah
but we can solve that by having “equivalencies” — if we bump up Greinke’s FIP over 200 (just an example), the “equivalent” can’t be “okay, I guess we’ll take Gobble downa few notches,” it would have to be Meche or Hochevar or something
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 11:05 AM EST up reply actions
My thoughts on the IP
I just thought I’d blurt them out because they were on the tip of my brain
These all seem fair to me:
Meche 200
Greinke 200
Bannister 150
Davies 150
(although of course either Bannister or Davies could easily go 180+. But those estimates are fair and reasonable.
I think Hochevar is going to make a full season of starts, barring injury (and I don’t think we can or should predict injury for him), so I think he should be more like 175 (conservative estimate). While I would love for Rosa to be the 6th starter and get 90 IP for the Royals, I don’t think that will happen. He might not even start the season in KC. If so, I think it will be in the bullpen and he’ll get no more than 50 ip for the Royals. I think the guy who ends up with the 6th most starts on the team will be HoRam, but of course this is very speculative. If I had to estimate now, I’d push his IP estimate to 90.
The bullpen is harder to judge. I think Mahay will get more than 55, but perhaps that projection makes sense as he might be traded by the deadline. I think 50 is on the high side for both Waechter and Yabuta, as they are both decent shots to not make the team on opening day. I would say 40 would be more reasonable for both of them. I really think Soria is going to pitch more like 80+ innings. I think the significantly lowered quality of the bullpen is going to lead to Hillman using Soria more.
The immoderate moderator
by Scott McKinney on Jan 15, 2009 11:23 AM EST up reply actions
yeah, we'll get to that
I do think both Greinke and Meche are likelyto pitch 200 next year, but even one stint to the DL means a pitcher probably won’t, so I think we can be a bit more conservative, at least at first.
I think filling up 6-7 slots, and starting the #1 and #2 guys at 180, 3-4 at 140, #5 at 100, then 6-7 at much lower things, then moving 1-4 or 5 each up by intervals of 5 or 10 until the target is reached. but that’s something we can vote on. I’ve got a baseline KC projection up on BtBS EditGrid page, when a good time comes up, I’ll do a post linking to it and opening things up for suggestions.
I’m just ready for the revolt when I have Jacobs in the DH slot, rather than at 1B, where’s he’s totally average on defense.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 11:34 AM EST up reply actions
Do you think Hillman will use Jacobs primarily as a DH?
And, if so, who gets most of the time at 1B? Again, we’re talking “will” not “should,” right?
The immoderate moderator
by Scott McKinney on Jan 15, 2009 11:41 AM EST up reply actions
will vs. should
that’s something I’m trying to figure out myself, maybe something to talk about in the thread amongst ourselves, and also Sky.
In my initial sheet, I balance them out, and frankly, it shurts the Royals taking HIllman/Moore’s preferences/badsignings into account. It costs the Royals at least half-a win starting Guillen over Teahen. I balanced out the bullpen’s leverage ratings, but it hurts the Royals’ projection to have Farnsworth with even a leverage of “1” rather than giving that to Bale, Mahay, Tejada, or even Joel Peralta (!) (I end up givng a number of guys 1.1 — it just has to average to 1).
I think we also have to decide whether Gload, TPJ, and German will be on the team at all next year, and/or how we should handle that, but those are all things I think will make for fun discussions and certainly can’t be projected by a computer.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 11:45 AM EST up reply actions
I think these playing time estimates should be about "will"
If we’re predicting/projecting how this team will perform in 2009, we should use the most realistic playing time estimates we can manage. If that means Jacobs gets 650 PA’s with 3/4ths of his PT at first base, then so be it. Reality is a bitch sometimes. The same goes for Guillen, HoRam and the rest.
About Farnsworth. While I think he’ll start the season as the #1 RH setup man, if he pitches as he’s projected to pitch, then I think his leverage will drop. Hillman doesn’t stick with the same bullpen roles throughout the season. But if you don’t want to try to figure out everyone’s average leverage over the full season, then giving a bunch of guys 1.1 should work.
The immoderate moderator
by Scott McKinney on Jan 15, 2009 11:54 AM EST up reply actions
well, that's what I figured
Farns will start with the 1.3 “standard setup” leverage, then drop…
I really don’t think Trey/DMGM are THAT blind. They did everything they could not to give Aviles a chance, but TPJ forced their hand, and Aviles shocked the world (and me) with a 4.4 WAR over 109 games. And DMGM gets credit for improving the Royals 3 years in a row… funny how that works out.
So, on the bright side, mightn’t there be a possibility that Treyton realizes the Farns thing isn’t working out, and for Tejada and Rosa to become the new Leo and RamRam? I think it’s fairly realistic talent-wise, at least. But what do I know.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 12:01 PM EST up reply actions
Yeah, that's definitely possible
Even if Rosa starts the year in Omaha, I think he’d be one of the first pitchers called up. And Hillman is wiling to go with the hot hand, as it were, as far as who gets the highest leverage middle relief innings. Of course, one of the joys of Farnsworth is that if he’s moved down to less important innings, he’ll bitch about it very publicly demanding props, respek, a trade or all of the above.
Even after trading away RamRam and Nunez, this still could have been a genuinely good bullpen. Tejeda and Rosa could/would/should have stepped up. Moore could have made a few cheap acquisitions, of which one of them would have panned out. And now we have to slog through the Farnsworth/HoRam/Waecther crap until the cream will (hopefully) be allowed to rise to the top.
The immoderate moderator
by Scott McKinney on Jan 15, 2009 12:07 PM EST up reply actions
Except
you said it yourself, the Aviles move only happened because they were forced.
So, not only does Farns have to suck and REALLY suck, they have to have someone that they think can be “adequate”. I guess my point is that following the TPJ model, just sucking isn’t bad enough. One must REALLY suck.
meat
oh, yeah
and I think putting Jacobs at DH does a good job of accounting for his “value” even if he does play mostly 1B. Well, it actually might be a bit generous…
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 11:46 AM EST up reply actions
updated and consolidated the probability graph onto one
one last time, Sky and NYRoyal pointed out that it didn’t look “right,” and ZeppelinDZ pointed out how to make it look better. Thanks, all.
I put “Standing Pat” on there, too, but just faintly…
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
DF -- Split adjustments - I can't find where we talked about splits so I will post it here
There are definitely splits and can be seen by using the yearly complete splits. I went ahead and figured out the the average split in OBP and SLG for all four instances:
Situation OBP adjustment SLG adjustment
vs RHP as RH 0.320 -0.015 0.410 -0.014
vs RHP as LH 0.349 0.013 0.438 0.014
vs LHP as RH 0.347 0.012 0.440 0.017
vs LHP as LH 0.320 -0.016 0.386 -0.037
by Jeff Zimmerman on Jan 15, 2009 1:10 PM EST up reply actions
thanks
I had wondered about this, not sure how to incorporate this into a “community” projection given other factors, but good stuff nonetheless
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 1:25 PM EST up reply actions
I am working on a manager scorecard
and needed the values for lineup generation. Using Marcels and a lineup generator here is what the maximize Royal’s lineup should be:
vs RHP 4.952 runs/game
Dejesus Gordon Crisp Jacobs Teahen Aviles Butler Buck Callaspo
vs LHP 5.734 runs/game
Butler Aviles Crisp Dejesus Shealy Guillen Buck Gordon Callaspo
by Jeff Zimmerman on Jan 15, 2009 1:53 PM EST up reply actions
Be sure an email to Treyton
well, perhaps a carrier pigeon would be the best approach
anyone want a non-cum comparison of initial AL Central projections for 2009?
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 1:55 PM EST up reply actions
Question on Lineups --
— when would a coach start taking into account current season stats, vice previous/projected stats. ~100 AB’s or do they just start to get averaged into the projected stats?
when would, or should
I remember reading a thread about this at insidethebook a while back… check the archives there. MGL, perhaps typcially, said “never,” while Tango and others said in August current season performance makes sense
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 2:35 PM EST up reply actions
Thanks, found it
Alot of theories, little backed up except 130 ABs with Marcels. I might just add in YTD stats into Marcels/Chone/Bill James and go with those numbers.
I should not have bitten off this project, but I am slowly moving forward.
by Jeff Zimmerman on Jan 15, 2009 3:23 PM EST up reply actions
(line)
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 15, 2009 3:27 PM EST up reply actions
THE EDITGRID IS READY FOR NERDIFICATION
I left all the defensive stuff I put in, as well as the baserunning numbers — it didn’t half-ass t them, either. As above, I looked at multiple sources where available to get decent numbers for everything.
The CHONE projections are set at th 50th percentile, but we can argue over basic modifications. That will actually be a minor part of it, as we are going for “midpoint accuracy” on this particular projection, not optimistic or pessimistic.
The main thing is to figure out playing time, particularly for the bench guys. And also who is really going to be on the roster — the bullpen and the “#6” and “#7” rotaiton spots are the biggest difficulties. I flattened out the playing time projections in general from my initial guesses so that it wouldn’t bias people. I dunno, maybe I’ll put them back, what would you think, “flatten them” first and then go from there, or do you want me to contour them a bit first. For example, right now, I have starters 1-5 all with 160 innings, that won’t happen, of course, but the more conservative we start, the less outlandish our final community projection will be. Should I just not worry about meeting the minimm’s at this point, and give a basic “countour” anyway?
In other words, are we ready for a “community projection” fanpost?
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
I wouldn't flatten them
Start with a more realistic baseline from which we knuckleheads can adjust them. Other than that, I guess we’re ready.
The immoderate moderator
by Scott McKinney on Jan 15, 2009 11:13 PM EST up reply actions
Sorry for not diving in too much earlier, but no Duck numbers?
Right now I am pretty sure he would be the #6 starter and possibly Rosa’s numbers with be in the pen instead.
by Jeff Zimmerman on Jan 16, 2009 11:20 AM EST up reply actions
I'm working on it right now, this is the stuff we can argue about
although converting pen/start numbers will be “fun” for me. Yes I know add/subtract….
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 16, 2009 11:37 AM EST up reply actions
So are we in the argue mode or is it still just the inital setup stage?
by Jeff Zimmerman on Jan 16, 2009 11:40 AM EST up reply actions
I'm setting up a new fanpost
The above represents an initial stab with my own ideas. The new post will more of a genuine “community” projection. I agree that Rosa will probably be used in the ‘pen, for now I just want to get the thing published so we can begin. As far as PT, within reasonable projections (Meche and Greinke aren’t both going to be projected for 220 IP next year with 3.2 FIP, Gordon isn’t going project for 680 PAs with a .375 wOBA, etc.), it’s going to be up to all of us to give a reasonable projection, not an optimistic prediction (which I hope people will publish on their own).
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 16, 2009 11:46 AM EST up reply actions
One more thing ...
..do you have the at bats per position over the season handy.
by Jeff Zimmerman on Jan 16, 2009 11:53 AM EST reply actions
me?
no. I’m going with an average of just uner 700 (695) per position per team, as per Sky. Trying to work it out with wOBA/OBP as a community will be a nightmare, so I’m going to suggest we just try to hit the 6250 target.
Bringing you more-or-less replacement level analysis and commentary since sometime in 2008.
by Matt Klaassen on Jan 16, 2009 12:04 PM EST up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Royals Review
Related FanShots on Royals Review
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
