CAIRO Projections and the 2012 Royals
by Scott McKinney on Dec 15, 2011 3:23 PM EST
Email
PrintThe CAIRO projections have come out, and while this is not the best projection system, it tested well last year and is better than a simple weighted average and regression (Marcel), but isn't quite as good as the best projection systems (probably ZiPS and PECOTA).
The roll out of the projections also came with projected standings. The Royals were projected to have a 73-89 record. Even the guy who puts CAIRO together recognizes that the projected standings aren't worth a lot because rosters change and it is hard to estimate playing time. But I thought I'd give it a shot, and tweak the CAIRO data to come up with a more accurate projected win total for the 2012 Royals, based on CAIRO's invidual player projections.
The basic methodology involved making my own estimates as to playing time (PA and IP) and adjusting RAR and WAR accordingly. For position players, I added CAIRO's defensive projections to projected BRAR to come up with a total WAR (the defensive projection was also changed proportionately for my estimate of playing time). For pitchers, I recalculated WAR based on FIP, as CAIRO's pitching WAR was apparently based on (E)RA.
My playing time estimates were just my best guess. I tried to take into account what appears to be the player's likely usage, his injury history and his recent history with regard to playing time. I also gave every regular something less than a full season of playing time, to account for little injuries which cut into playing time and the DL stints which will pop up here and there.
My assumptions are just that: assumptions. They could be far off. Who knows. So take this for what it is. This is a win total based on certain playing time estimates, using a given system's projections. It's useful information, but far from the final word on how many games the Royals should win in 2012.
The following isn't going to be pretty, but I don't really have the time to tart this up.
Player, PA or IP, WAR
Billy Butler, 660, 2.40
Eric Hosmer, 660, 0.59
Jeff Francoeur, 635, 1.46
Alex Gordon, 610, 3.23
Lorenzo Cain, 580, 1.85
Alcides Escobar, 580, 1.62
Mike Moustakas, 580, 1.76
Salvador Perez, 560, 1.29
Johnny Giavotella, 510, 0.41
Chris Getz, 170, 0.29
Mitch Maier, 160, 0.41
Brayan Pena, 135, 0.52
Jarrod Dyson, 75, 0.36
Joaquin Arias, 50, 0.0
David Lough, 50, 0.03
Manuel Pina, 20, 0.06
Repl. Lev. Fill-ins, 75, 0.0
TOTAL, 6110, 16.28
Felipe Paulino, 180, 3.13
Luke Hochevar, 170, 2.05
Danny Duffy, 165, 1.55
Bruce Chen, 150, 1.55
Jonathan Sanchez, 150, 1.76
Aaron Crow, 65, 0.22
Greg Holland, 65, 0.66
Joakim Soria, 63, 1.20
Louis Coleman, 60, 0.18
Tim Collins, 60, 0.0
Blake Wood, 60, 0.03
Jonathan Broxton, 50, 0.75
Mike Montgomery, 50, 0.62
Kelvin Herrera, 50, -0.06
Sean O’Sullivan, 40, 0.16
Everett Teaford, 40, -0.49
Nathan Adcock, 10, 0.01
Vin Mazzaro, 10, 0.07
Jeremy Jeffress, 10, -0.09
TOTAL, 1448, 13.32
Replacement + Pos. Pl. WAR + Pitcher WAR = Projected Win Total
48.6 + 16.28 + 13.32 = 78.2 Wins
136 comments
|
2 recs |
Do you like this story?
Comments
The 0.59 is going to bring some heat.
Glad I came, just wish I hadn't stayed so long.
Rock Chalk Talk
CAIRO doesn't like him very much
Over 660 PA’s (my estimate), CAIRO projects him to hit .273/.324/.429 with -7.8 defensive runs.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 3:51 PM EST up reply actions
They don't measure the intangibles
BFIB
by tiquanunderwear on Dec 15, 2011 3:52 PM EST up reply actions
Clearly you haven't looked at the spreadsheet
Hosmer’s wGRIT projects at -4.9
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 4:02 PM EST up reply actions
He's done
Trade him for the grittiest player in the game
“Laser show”
BFIB
by tiquanunderwear on Dec 15, 2011 4:39 PM EST up reply actions
hey, I think you just came up with a good nickname for Lorenzo Cain
(besides Lorenzocaine)
"You're like that guy who wrote that thing about remembering stuff!!"
- Crow T. Robot
Already taken
by Pedroia
Many people would sooner die than think; in fact, they do so.-Bertrand Russell
by Dr. van Strijcker on Dec 15, 2011 5:02 PM EST up reply actions
Do they not regress defense at all?
Projecting a player with, what, 5 months of defensive data at 2 standard deviations below the mean is laughable.
Not sure his defensive projection methodology
It’s difficult as we don’t have minor league data to project from. FWIW, this defensive projection is a little better (per X innings) than how he did in 2011.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 4:07 PM EST up reply actions
Defensive projections are based on an average of zone rating, plus/minus, Totalzone and UZR. They are regressed, but I don’t think they are of any use when looking at a player like Hosmer since we have less than a full season of numbers on him.
And I think CAIRO’s way low on Hosmer overall, but it’s generally pessimistic on young players. Hell, I hate CAIRO’s projection for Jesus Montero too.
It’s basically just designed as Marcel with a few more inputs (MLEs and park mainly). I think ZiPS is closer on Hosmer, FWIW.
But CAIRO’s defensive projections have separate projections for DRS, UZR, ZR, etc. So I don’t think it’s projections some from a combination of defensive metrics.
You may know me as NYRoyal.
by Scott McKinney on Dec 16, 2011 12:37 PM EST up reply actions
Right, I show each one separately, but the last column is avg which is just a straight average of the four. Since defensive metrics are somewhat unreliable, I think the more decent systems we look at the closer we may get.
I fully agree
For this exercise, I just used UZR because I wanted to follow Fangraphs WAR model. But I really do like aggregate defensive stats, as long as each is PBP-based.
You may know me as NYRoyal.
by Scott McKinney on Dec 16, 2011 12:53 PM EST up reply actions
I was pretty surprised to see the disagreement – what do you have as Hosmer’s translation? Mine must be way higher than yours.
by D.Szymborski on Dec 16, 2011 4:56 PM EST up reply actions
No kidding.
It even seems laughably bad, but if that’s what it takes to get to 78 wins next season, I might take it.
Nick Swisher and Johnny Giavatella deserve treats in the Kaufmann parking lot.
Joaquin Arias just signed with SF
Your entire projection is thrown off
Relive Royals History at royalsretro.blogspot.com
by RoyalsRetro on Dec 15, 2011 3:47 PM EST reply actions 1 recs
Looks like someone else is going to have to manage his 0.0 WAR
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 3:52 PM EST up reply actions
Thanks for posting
I don’t like the ‘add up the WAR" method, but it’s quick and easy.
It’s pretty easy to see the team doing significantly better than these projections overall. There are several players cough Hosmer cough that I’d confidently take the over on.
It also doesn't include major injuries to any player, especially the position players
And while it doesn’t show great improvement from many young players, it also doesn’t include any collapses.
I don’t like the ‘add up the WAR" method, but it’s quick and easy.
It definitely has it’s limitations. But given more time to do it, what methodology would you prefer?
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 4:04 PM EST up reply actions
Limiting to quick-and-dirty methods, I conceptually prefer totaling RS and RA and calculating XWL from that. You could subtract defensive runs from the RA total.
I don’t know if that’s actually any more accurate, but it’s easier for my mind to connect the dots from rate stats to run differential and run differential to wins than it is to imagine a team of replacement level players winning 48.6 wins and then summing up how many wins above replacement for each player.
I might try that method with this data at some point tonight to see if it gives different results. Theoretically, it should be very close. Which XWL formula do you use? RS^2 / (RS^2 + RA^2)? Or do you use a more refined exponent like in pythagenpat?
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 4:20 PM EST up reply actions
I got 712.9 runs scored and 783.3 runs allowed. I would think that defensive runs should be subtracted from RA because they represent runs scored (the alternative would be adding those defensive runs to RS). Using those numbers and doing a simple pythag (exponent = 2), you get 74.0 wins. But using UZR doesn’t cover all runs saved by defense. I think if you capture all of the runs saved, you get a higher win total, but it wouldn’t exceed the 78 expected wins from the WAR method.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 4:43 PM EST up reply actions
Yeah, my limited experience with the WAR-summing method is that it tends to shoot a little high
I think it’s because the summing method assumes a linear relationship between runs and wins. Real-world data suggests that this relationship is not linear, and that’s why pythag formulas that attempt to define this relationship have exponents.
Assuming a linear relationship is a close enough approximation when you’re looking at the impact of a single player (a straight line can approximate a curve over 2 WAR), but when you start trying to sum up 30 WAR from a 40 man roster, the linear approximation doesn’t work as well.
There’s probably other considerations too, but my brain is only like 20% functional right now after working 24 of the last 32 hours.
The theory on the exponent is that it should depend on the scoring environment. The formula BP uses to get the exponent is found here:
http://www.baseballprospectus.com/glossary/index.php?mode=viewstat&stat=136
For that scoring environment, I get an exponent of 1.884.
Plugging in that exponent, you get 73.8 wins, which sounds about right if the players are as shitty as CAIRO thinks they are. If Hosmer is barely over replacement level and Bruce Chen is by far our best pitcher with a 4.2 ERA, we’re in for a long year.
I wonder how different the ZiPS and PECOTA projections will be
Undoubtedly there will be some players where the difference is significant (in both directions), but I wonder if the team totals will be significantly different. One problem is that PECOTA doesn’t project defense (IIRC) and ZiPS only does so in a general way (“above average,” etc.). So in order to turn those projections into a team projection, we’d have to include estimated defensive runs, which leaves a lot of room for us to inject our own subconscious bias.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 8:10 PM EST up reply actions
I think projecting the Royals as roughly average defensively is close enough
They’re certainly not exceptionally good or exceptionally bad, and I’m not interested in trying to project down to 5 or 10 runs.
CAIRO and ZiPS are in super-disagreement with Hosmer. ZiPS has Hosmer at 304/354/474, 123 OPS+
by D.Szymborski on Dec 15, 2011 11:14 PM EST up reply actions
Everyone here will enjoy that much more
And will feel much better about projection systems in general (except for the Royals that you don’t project to be so good).
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 11:21 PM EST up reply actions
We know, we know.
You keep beating that drum.
I don’t disagree with you, but your knee-jerk counterpoint is becoming as tiresome and immature as the unwashed masses.
"You're like that guy who wrote that thing about remembering stuff!!"
- Crow T. Robot
Lots of things become tiresome to one person or another. I guess we all just have to accept a lot around here.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 11:33 PM EST up reply actions
this sounded WAY harsher than I intended…..sorry.
"You're like that guy who wrote that thing about remembering stuff!!"
- Crow T. Robot
No problem.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 11:43 PM EST up reply actions
It looks more right to me
I don’t buy CAIRO’s projection that Hosmer will hit significantly worse next year
Been checking the ZiPS series every couple days for weeks hoping the Royals would show up soon
How far back on the list are they?
HOSMER SUCKS ASS
"You're like that guy who wrote that thing about remembering stuff!!"
- Crow T. Robot
basically
if he doesn’t clean up his defense and stop playing so close to the line. If his defense hurts the team that much and prevents them from an additonal 2-3 wins it could end up hurting in a weak AL central. i mean average AL central
Do these effectively hide my thunder?
his defensive alignment prolly has more to do with Yost and the other coaches
than it does with Hos-i-moto himself.
by DickHowser4ever on Dec 15, 2011 5:10 PM EST up reply actions
Hosmer!
Superior
Unique
Courteous
Knowledgeable
Supernatural
Awesome
Superpower
Superstar!!!!!!
by Fans-of Eric Hosmer on Dec 15, 2011 9:52 PM EST up reply actions 1 recs
“Not in ’Nam, of course.”
Obviously, you are not a golfer.
by Kyled85 on Dec 15, 2011 11:24 PM EST up reply actions 1 recs
How much do the Egyptians really know about baseball?
by Bronzillo on Dec 15, 2011 5:05 PM EST reply actions 8 recs
That's racist.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 5:07 PM EST up reply actions
I smell a pyramid scheme...
More Goofus than Gallant...
by Tracer Bullet 82 on Dec 15, 2011 5:16 PM EST up reply actions 5 recs
you forgot to include the 400 ABs promised to all RR posters in yesterdays topic
that would add at least 5 hits to the season total (i am adding up all hits from all posters)
by DickHowser4ever on Dec 15, 2011 5:09 PM EST reply actions
gosh
that’s almost a million plate appearances.
That would look like a batting average of .000
Okay, sorry, I know we don’t value batting average here.
That would look like an on-base percentage of .000
"You're like that guy who wrote that thing about remembering stuff!!"
- Crow T. Robot
hey look
.jpg)
The H in Jesus H. Christ stands for Hosmer.
by HippoLaBamba on Dec 16, 2011 12:59 AM EST up reply actions
Nice work, Scott.
Mid point of their projected range is 81, so your player look suggests that the Royals are more likely to perform in the bottom half of their range.
by BlueEyes_Austin on Dec 15, 2011 5:43 PM EST reply actions
I'd be happy with 78 wins
anything over that would be pretty damn good for the Royals, what the hell happened to will mcdonald?
by Kansas City Keith on Dec 15, 2011 9:29 PM EST reply actions
He's A Zombie
I used to be an A's fan until they left town and got good.
by philofthenorth on Dec 15, 2011 11:57 PM EST up reply actions
Honest question:
What makes CAIRO or ZiPS or PECOTA actually mean anything? How is it any better than pulling random (yet logical numbers) out of a hat?
I’m a little wary of these projection systems. They seem to be extremely wrong quite often—ZiPS 2011 projections for Gordon (244/.342/.407) and Kila (.244/.355/.412) were way off, for example, not to mention Frenchy and Cabrera’s projections (couldn’t find the numbers for them). Why are they a good indicator of future performance? Or are they?
They are objective systems to project future performance based on past performance, appropriately regressed and applying an aging curve. They definitely aren’t perfect, but the best systems have a pretty good track record. Sometimes they are going to be far off. Much will happen that isn’t predictable. But they are the best estimators of future performance out there.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 9:43 PM EST up reply actions
"Sometimes they are going to be far off"
You got that right! I don’t see why have a data system in place..that are projected to be far off or not perfect….if they are using a data system to evaluate players…it should be far more perfect..or better..then being projected far off. One reads the data and it makes the player viewed not as good….on the other hand…ones read the sports news or watches…hotstove..mlb…sports media..intentiional talk and so on…..and when they speak about the players…it surely contradicts everything you read on the data systems used to evaluate the players. So, I would think if something is not actually accurate..it should be changed to something more accurate.
by Fans-of Eric Hosmer on Dec 15, 2011 10:17 PM EST up reply actions
I don’t see why have a data system in place..that are projected to be far off or not perfect…
Not exactly sure what you’re saying here. If you’re saying we shouldn’t use projections that we know will be far off, then I guess you’re right. But there isn’t good reason to believe that these projections, overall, will be far off. It is essentially certain with every projection system that for some players they will be far off. Sometimes breakouts, collapses and aberrational seasons will happen. It wouldn’t be reasonable to expect a projection system to predict all of those occurrences. If you’re saying that a projection system which isn’t perfect is useless, then I disagree. Perfection is a good goal, but it isn’t a reasonable requirement.
if they are using a data system to evaluate players…it should be far more perfect..or better..then being projected far off
Again, not sure what this mean, but the system and its projections overall is not “projected to be far off.”
So, I would think if something is not actually accurate..it should be changed to something more accurate.
Um, I agree. What does this have to do with CAIRO projections, or projection systems in general? Are they insufficiently accurate for you? Are you familiar with their track record and how accurate they have been? Should they be thrown out or ignored if sometimes far off for some players? If you’re arguing that they should be improved, then you’ll be happy to know that systems like Oliver, CAIRO, ZiPS and PECOTA are tweaked every year in an attempt to improve them. Do you have any suggestions for how they could be improved?
Is this really all about the fact that you don’t like Hosmer’s projection? It’s really hard to say how good Hosmer will be next year. He could have a huge breakout and be an All-Star with MVP votes. Or he could have a sophomore slump and have a significantly worse year than his rookie season. Both are reasonably possible. And will his defensive positioning improve? If the Royals don’t position him differently, I think his defense (in terms of actual outcomes) will continue to be poor. So there’s real downside potential. While this isn’t the prediction for Hosmer that I would give (and I think some other systems will have different numbers), I don’t think it is an unreasonable projection, and it should not be dismissed out of hand because us Royals fans know for sure that he’s the bestest.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 10:32 PM EST up reply actions
Uh....Scott....
Let this one go.
Let it go.
"You're like that guy who wrote that thing about remembering stuff!!"
- Crow T. Robot
Oh no..Scott..your not letting me go---this is how I learn
If you’re saying that a projection system which isn’t perfect is useless, then I disagree. Perfection is a good goal, but it isn’t a reasonable requirement.
Scott, I’m trying to say… that a a projection system which isn’t perfect is useless if it provides in accurate information. If the figures are not accurate or the figures are far off…at times..(as it was posted above on another post)…than they should come up with a projection system that would be closer to accurate and not use a system that is not always accurate or throws the numbers far off.
What does this have to do with CAIRO projections, or projection systems in general?
Lost you on this part..What were you referring to?
Correct me please….my understanding on the CAIRO projection or projection system..that it is a data sheet.
Are you familiar with their track record and how accurate they have been? Should they be thrown out or ignored if sometimes far off for some players?
I am familiar with their track records.(.when I go to review them) Not as good as most of you are. So I visit Royals Review along with other sites to read, review and learn..what I don’t know. As far as ignored or thrown out…my my..of course not. Oh and for the record “I’m not arguing”
If you’re arguing that they should be improved, then you’ll be happy to know that systems like Oliver, CAIRO, ZiPS and PECOTA are tweaked every year in an attempt to improve them. Do you have any suggestions for how they could be improved?
Do they actally have people at the games that tally or do they review by videos?
Is this really all about the fact that you don’t like Hosmer’s projection?
No..not at all……—That I like Hoz! Yes, I do! But not just HOZ though! There is Butler, Gordon, Escobar, Perez, Giavotella, Francouer, Ghetz (Cabrera),who we no longer have “Moose” Moustakas ..and my list goes on! Actually, I care and like the whole team.
This team is the younest team right now in MLB..correct me if wrong please (thks) Hoz will do better this year then last year…Moose will do better this year to…same for Perez..and Eco.
Thanks for beng patient with me!
by Fans-of Eric Hosmer on Dec 16, 2011 2:27 AM EST up reply actions
than they should come up with a projection system that would be closer to accurate and not use a system that is not always accurate or throws the numbers far off
No projection system is always going to provide accurate projections.
Do they actally have people at the games that tally or do they review by videos?
They are using publicly available stats and making projections based on those stats.
You may know me as NYRoyal.
by Scott McKinney on Dec 16, 2011 12:41 PM EST up reply actions
Good post!
Thanks Scott, for explaining..kinda of getting the picture now!
by Fans-of Eric Hosmer on Dec 17, 2011 3:52 PM EST up reply actions
Still
Projections can’t account for injuries, spikes or valleys in playing time, off the field issues, or jumps in production that occur all the time. I think it’s good for some players (I’m interested in Butler’s projections) but a system that was way, way off on four Royals off the top of my head last year makes me think twice about taking them seriously, or as seriously as some.
Of course they're going to be wrong a lot
They just predict what “tends to” happen. Human behavior is only predictable some of the time. It would be impossible, and frankly, quite sad if we could predict with near certainty what will happen in baseball.
But it gives us a good estimate, which is useful data, although it should not be conclusive by any means.
Relive Royals History at royalsretro.blogspot.com
by RoyalsRetro on Dec 16, 2011 11:00 AM EST up reply actions
They’re pretty accurate in the aggregate. Individual projections can be pretty far off, just like specific players have individual seasons that are way off of their career norms.
Let's just trust the process.
by trusttheprocess on Dec 16, 2011 11:37 AM EST up reply actions
that's the way I look at them...
If you take all the projections for a team of 25, the projections are decent, but I don’t put a lot into what they think of a player, individually.
If women only slept with nice guys...guys would only be nice. And they don't. And we're not.
by setupunchtag on Dec 17, 2011 10:37 AM EST up reply actions
Wow on Paulino projection
And he’s not a lock for the rotation?
by YouDon'tPhaseMeGobble on Dec 15, 2011 10:51 PM EST via mobile reply actions
Exactly
He is on a very short list of pitchers who should be a lock for the rotation.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 10:55 PM EST up reply actions
Didn't CAIRO project his ERA to be 4.92?
What am I missing here?
Yes, and clearly the ERA projection wasn't a component-ERA or anything like it
Must have been a straight projection based on past ERA’s.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 11:19 PM EST up reply actions
I don’t think ERA over a not huge sample is the best predictor of future ERA or RA.
You may know me as NYRoyal.
by Scott McKinney on Dec 15, 2011 11:20 PM EST up reply actions
I've got a hard time seeing a 21 year old who put up 1.6 WAR in about 3/4ths of a season drop down THAT badly,
I don’t understand what leads to a projection system having Hosmer being a SIGNIFICANTLY worse player than he was last year. Is there something in the numbers that says he was an extremely lucky player last year? Doesn’t make sense to me how Hosmer is projected to be that poor of a hitter going into his 2nd year.
Killing time until time kills me
I do think ZiPS will be better
They perfectly predicted Butler’s OPS+ last year. So maybe they got their 1B/DH locked in.
The WAR for our relivers seems a little low as well.
The Royals will most likely be a couple of wins better overall there as well.
Go Royals!
Everett Teaford will not be a -0.5 WAR pitcher. He won’t get that opportunity.
by WURoyal on Dec 16, 2011 2:17 AM EST via mobile up reply actions
Well, not just that, for example Blake Wood has War totals of
2010 -0.2
2011 +0.6
So when you use the 5-4-3 weight, you would get a projection of
+ 0.24 WAR and he is listed as .03 WAR
The numbers just don’t seem right.
Go Royals!
But there will be some negative WAR pitchers at the bottom of the pitching staff
There always are.
You may know me as NYRoyal.
by Scott McKinney on Dec 16, 2011 12:42 PM EST up reply actions
Maybe with the lefties in the staff.
The top right handers are Soria, Holland, Crow, Broxton, Coleman, Wood, Herrera and Jeffress. Three of them won’t make the team and can be brought up as needed. That is a crazy good set of arms for only 3 WAR, and they will most likely get more innings then an average staff as well. There is no way you can justify only 3 WAR as the 50% line for the Royals pen.
Go Royals!
Scott, did you use FIP for pitcher WAR?
RA (or ERA) would be a better way to go. FIP is intended to take the luck out of pitching numbers to get at a player’s true talent, but a projection run average is supposed to already give its best estimate of a pitcher’s true talent.
I ask because CAIRO thinks Paulino will be awful (5.45 RA), but it thinks his peripherals will be much better than his actual run prevention (3.78 FIP). While I’d love to give Paulino credit for his peripherals, what matters to the team is how many runs he allows, and CAIRO thinks he’ll allow a ton.
Yes, I used FIP for pitcher WAR
RA (or ERA) would be a better way to go. FIP is intended to take the luck out of pitching numbers to get at a player’s true talent, but a projection run average is supposed to already give its best estimate of a pitcher’s true talent.
It is clear from looking at the data that CAIRO projected ERA based on past ERA. The ERA in the projection is clearly not a component ERA. And I don’t think a 3-year weighted average of ERA provides a good projection of future ERA or RA (unless the pitcher has 180+ innings in each of the 3 seasons), and the research supports that. So I think when you have to choose between ERA and FIP (which are the only choices from this data set), FIP gives you something much closer to true talent level.
I ask because CAIRO thinks Paulino will be awful (5.45 RA), but it thinks his peripherals will be much better than his actual run prevention (3.78 FIP). While I’d love to give Paulino credit for his peripherals, what matters to the team is how many runs he allows, and CAIRO thinks he’ll allow a ton.
CAIRO thinks he’ll have a high ERA because he’s had a higher ERA in the past (over a not huge sample of data). And I don’t think that data is the best data to project how many runs he’ll give up in the future.
FWIW, if we project this pitching staff’s RA, based on their past ERA or RA, they are going to look pretty awful, probably in every projection system. Using ERA converted to RA, the pitching staff only totals 10.7 WAR, as opposed to 13.32 using FIP. The only hope we have that this pitching staff might be even decent is if their TTL is more represented by their DIPS stats than their ERA or RA.
You may know me as NYRoyal.
by Scott McKinney on Dec 16, 2011 12:51 PM EST up reply actions
I guess you can sum it up however you want, but I think if we’re trying to figure out how a projection system views a team, we should look at the projected RA’s for the team’s pitchers rather than their FIPs. We should take the numbers the system spits out as its best guess at the players’ true talent levels. If it spits out numbers that don’t square with our expectations, we should question what the projection system (or our expectations) is doing wrong rather than finding other numbers that do fit our expectations.
As it happens, I think the FIPs look more realistic than the ERA’s, and I think you’re right that it’s because CAIRO doesn’t regress out as much of the “luck” components out of ERA as we would like, but we should remember that this might be by design.
I’m not trying to pick the stats that meet my expectations. The ERA-based WAR actually comes up with a result closer to my expectation. But I use FIP for WAR because over a smaller sample of data (such as what is used for projections), FIP shows true talent level considerably better than ERA. That’s why Fangraphs uses FIP for its pitching WAR.
As it happens, I think the FIPs look more realistic than the ERA’s, and I think you’re right that it’s because CAIRO doesn’t regress out as much of the "luck" components out of ERA as we would like, but we should remember that this might be by design.
I think CAIRO projects ERA from ERA. I don’t think CAIRO is trying to make ERA into CERA (Component ERA). I don’t think CAIRO is trying to make ERA into a single true talent measure. ERA doesn’t work well for that under most circumstances anyway, so I can’t fault them there. FIP is inherently measure of meaningful defense independent components, so it projects much better and more accurately than ERA.
When the other projections come out, I’m going to continue to make team projections from FIP (or other defense independent metrics, if they are included in the projection) as it is a better measure of true talent than ERA. And I don’t think we can just guess or assume that the ZiPS or PECOTA ERA projections are actually component ERA’s which should be better TTL measures than FIP. I would guess that PECOTA will have a SIERA projection, which I’m happy to use to project team RA.
You may know me as NYRoyal.
by Scott McKinney on Dec 16, 2011 8:03 PM EST up reply actions
Gonna say you're wrong on this one
Basically, when projecting the Royals future performance, we care about runs scored and runs allowed. The projection systems are kind enough to tell us straight up how many runs they think our pitchers will allow. We should use those numbers.
It seems like you want to use FIP because you don’t think CAIRO did a good enough job projecting ERA. I’m sympathetic to that view, but for whatever reason, CAIRO thinks Royals pitchers will allow a lot more runs than their peripherals will suggest that they should have. It seems like a bizarre position for CAIRO to take, but that’s what their numbers say.
Basically, when projecting the Royals future performance, we care about runs scored and runs allowed. The projection systems are kind enough to tell us straight up how many runs they think our pitchers will allow. We should use those numbers.
And projected ERA doesn’t do a good job of projecting actual runs allowed. Why? Because past ERA doesn’t project future runs allowed well (again, unless you have a ton of data from the pitcher).
It seems like you want to use FIP because you don’t think CAIRO did a good enough job projecting ERA. I’m sympathetic to that view, but for whatever reason, CAIRO thinks Royals pitchers will allow a lot more runs than their peripherals will suggest that they should have
Because that system clearly is projecting future ERA based on past ERA. And these pitchers have given up a lot of earned runs in recent seasons. I don’t think that’s the best way to project how many runs they will give up in the future, and the research supports that.
For team RS, would you just total up the projected “R” for each player (with appropriately adjusted playing time)? As opposed to a more meaningful run estimator, like Base Runs? Again, I think a simple stat like “Runs” is projected based on past Runs scored by the player. But Base Runs counts up a variety of meaningful projected events (1B, 2B, BB, SB, etc.) and gives them their proper weight to determine how many runs certain hitting should give rise to. So you can go with an ultra simple measure (ERA for RA and R for RS), or a more complex, meaningful and very likely more accurate measure. I prefer the latter.
You may know me as NYRoyal.
by Scott McKinney on Dec 16, 2011 9:08 PM EST up reply actions
I think you’re missing the step where the projections already attempt to output true talent level. They’re supposed to tell us their best guess at true talent RA, ERA, FIP, etc. For purposes of estimating wins, we care about true talent RA.
But some stats don't do a good job of showing true talent level
No matter what machinations you go through, you can’t show a hitter’s true talent level by projecting his batting average. That is, unless you throw a hell of a lot more into it and make it into wOBA. But when you do that, you’re not talking about batting average at all.
What is it that you think these projection systems do to project ERA which makes it what we should rely on as the player’s RA true talent level? Do they only use defense independent components to project ERA (I don’t think they do). Do they find the pitcher’s personal BABIP and use that, along with defense independent components and batted ball data to come up with a TTL ERA? I don’t think they do that either, and you can’t come up with anywhere near a reliable personal BABIP for a pitcher unless you have a massive amount of data for the pitcher. So why is it that you assume that projected ERA hits close to TTL of runs allowed for a pitcher when ERA does not?
You may know me as NYRoyal.
by Scott McKinney on Dec 17, 2011 12:33 AM EST up reply actions
We’ll stick to RA—no need to bring errors into this.
I can’t tell you what CAIRO does to project RA. It looks like it just regresses recent seasons of RA which is a bad way to do it.
A better way would be to do something like what FIP, xFIP and SIERA do which is to look at K%, BB%, GB%, and have those components make up the bulk of your weighting, and then splash in a little weight for recent BABIP and RA data.
But I don’t think we should try to hot-fix what projection systems output—or at least if we do that, we should make it clear that we’re doing so. CAIRO says Paulino is going to have good peripherals and give up a ton of runs. That seems to me to point to a flaw in the CAIRO methodology. I expect that ZiPS will have Paulino’s RA more in line with his peripherals.
But I don’t think we should try to hot-fix what projection systems output
The projection system’s output included ERA, FIP and much more. If you’re arguing that the only good way to get to projected team RA is to use ERA, then I categorically reject that. We’re projecting team RA from TTL, right? Then let’s use TTL. FIP captures it better than ERA, period. This isn’t that complicated.
I expect that ZiPS will have Paulino’s RA more in line with his peripherals.
Based on past ZiPS projections I’ve seen, I bet his projected ERA will be pretty high. Perhaps more in line with his peripherals than CAIRO, but probably not by a lot. We’ll see.
You may know me as NYRoyal.
by Scott McKinney on Dec 17, 2011 3:30 AM EST up reply actions
I don’t have the spreadsheet downloaded to this comp, but I’m pretty sure it included RA.
And every number that the projection system outputs should be its estimate of that players TTL. So we have estimated TTL RA, ERA, FIP, K-rate, BB-rate, etc. All of those numbers have meaning, but for purposes of calculating wins, we care about runs allowed.
You have to remember that FIP is a means for expressing peripherals, and it doesn’t include every aspect of a pitchers run-prevention TTL.
Ok, we’ll agree to disagree. And when the ZiPS and PECOTA projections some out, I’ll run the numbers based on straight RA and RA from a DIPS metric like FIP.
I asked above how you think a teams projected total RS should be calculated. Should it we just use the projected “R” for each player, with playing time adjusted to what we think is the best estimate? Or should we use something like Base Runs?
And putting all of that aside, given the numbers from the CAIRO projections, which project the Royals to have somewhere between 73 and 78 wins (depending on how you crunch the individual projections and turn them into team W/L), does that affect your opinion on how good this team should be in 2012, and how close they are to being contenders? If I remember correctly, your estimate (pre-Sanchez trade) was somewhere in the mid-80’s. Has your estimate changed with this new data?
You may know me as NYRoyal.
by Scott McKinney on Dec 17, 2011 12:37 PM EST up reply actions
For offense, I’d do what Fangraphs does which is to take the projected singles, doubles, walks, etc, and from those, calculate each player’s wOBA. You can use the wOBA and playing time to get batting runs.
For pitching, I’d just do RA times IP divided by 9. The problem then is what to do with defense—does the projection system account for defense in the RA or do you need to add team defense on top of that value? I’d add defense on top, but it may depend on the particulars of the projection system.
For offense, I’d do what Fangraphs does which is to take the projected singles, doubles, walks, etc, and from those, calculate each player’s wOBA. You can use the wOBA and playing time to get batting runs.
But wait, they give you the runs scored for each player. Why go through some complicated machinations? Why would you be in favor of “hot-fixing what projection systems output”? For hitters, you don’t just accept the Runs you are given. You use the best, most meaningful metrics to come up with RS. But with pitching, you close your eyes and assume the RA represents TTL and is based on all meaningful inputs properly weighted. Why not for RS? This contradiction makes no sense.
And how has this data affected your opinion of how good well this team should perform in 2012?
You may know me as NYRoyal.
by Scott McKinney on Dec 17, 2011 6:48 PM EST up reply actions
So sassy
I don’t know how they project runs scored, but my thinking was that the individual players would score more or less runs based on the playing time of the players around them, so if we want to manipulate playing time to suit our needs, we should use rate statistics that aren’t playing-time dependent.
And how has this data affected your opinion of how good well this team should perform in 2012?
You may know me as NYRoyal.
by Scott McKinney on Dec 17, 2011 7:38 PM EST up reply actions
we should use rate statistics that aren’t playing-time dependent.
You’re relying on a distinction without a difference. The RA in the CAIRO spreadsheet is RA/9. To get the RA for the pitcher, you just multiply that by 9, and then by IP. The fact that it is shown broken down into RA/9 does not provide any meaningful difference. The completely different ways that you would derive RS and RA are entirely contradictory. It makes no sense. But I’m done.
You may know me as NYRoyal.
by Scott McKinney on Dec 17, 2011 8:06 PM EST up reply actions
You misunderstood my point, so I'll try to explain more clearly
We want to use stats that capture each player’s individual contributions as fully as possible. Runs scored don’t do that for two reasons:
-Runs scored are highly context-dependent. If Hosmer and Butler are injured all year, Alex Gordons runs scored will be much lower than projected even if he bats just as well. In a similar sense, since projected runs scored are based on recent seasons of data, had Gordon spent the last 4 years on the Red Sox and batted exactly as he has for the Royals, his 2012 projected runs scored would be significantly higher even tho his projected wOBA would be the same.
-Runs scored only capture a fraction of a player’s offensive contributions. A player can help his team by getting on base and scoring runs, but he can also help his team by advancing runners and helping to drive his teammates in. Taking runs scored as a player’s full contribution is akin to taking OBP as his full contribution while ignoring his slugging.
These considerations don’t factor in for pitching in the same way. While it’s true that RA is somewhat context dependent (if Escobar goes down, Hoch will give up more runs) the alternative, FIP, ignores so much of a pitcher’s run-prevention contributions (GB/FB, BABIP, holding baserunners, etc) that the lesser of the two evils is to accept the context-dependence.
The bigger consideration with using runs scored for offense is that it doesn’t include half of a player’s offensive contributions. The same obviously cannot be said for a pitcher’s RA. Every aspect of run prevention will show up in his runs allowed.
Any chance you’re going to answer the question I’ve asked a few times in this thread about how has this data affected your opinion of how good well this team should perform in 2012? Or are you punting?
You may know me as NYRoyal.
by Scott McKinney on Dec 18, 2011 10:31 AM EST up reply actions
These CAIRO projections? Not at all. I think several of the players will out-play these projections.
So the data from the first projection system with a pretty good track record comes out and it doesn’t affect your opinion of how good the team is at all because you disagree with some of the individual player projections? Don’t you think you’re going to disagree with some of the player projections from ZiPS and PECOTA too? That’s how it goes every year for every team and every projection system. Trust me, you’re going to think that several players will out-play their ZiPS and PECOTA projections too. You’re very optimistic about the Royals and their young players. That’s kind of why we should look at the data from projection systems. They are objective and unbiased.
This is starting to look like you’re not really interested in the data from projection systems. You think they are an 85-win team (or something like that) and you’re not going to let the data get in the way of your opinion. Or I guess you’ll just rely much less telling data (like projecting the team based on a baseline of the team’s 2011 pythag) because you like the results of that method better.
You may know me as NYRoyal.
by Scott McKinney on Dec 18, 2011 5:20 PM EST up reply actions
yawn
Keep beating that horse. We’ll talk when we have a group of projections that isn’t obviously flawed.
by kcdc1 on Dec 18, 2011 6:25 PM EST via mobile up reply actions
I can’t wait to hear about the “obvious flaws” of many of the player projections from ZiPS and PECOTA for the Royals. Outside of the Hosmer projection, what “obvious flaws” do you see in the CAIRO projections? Does “obvious flaw” = “I think he’ll perform better than that”? If so, then I think you need to re-think your analysis. Hopes, dreams and fan optimism are no substitute for solid analysis. It’s fun though.
You may know me as NYRoyal.
by Scott McKinney on Dec 18, 2011 7:34 PM EST up reply actions
Not interested in arguing over these CAIRO projects
I’m sure we both see a lot of things we disagree with. I can handle waiting a couple weeks for ZiPS.
Also, I’m not exactly sure how our discussion of how FIP and ERA should be interpreted differently for past statistics and forward projections turned into you taunting me about my disregard for numbers. It’s almost like you just want to argue.
I’m sure we both see a lot of things we disagree with. I can handle waiting a couple weeks for ZiPS.
You’re missing my point. In every set of projections for a team, you’re going to see several that you or I or anyone disagrees with. If that leads one to toss out the whole set of projections (as data which doesn’t affect one’s opinion of how good the team will be next year), then there won’t be any projection system which you will buy. It’s not like you’re going to agree with all of the ZiPS projections (nor will I, or anybody)
Also, I’m not exactly sure how our discussion of how FIP and ERA should be interpreted differently for past statistics and forward projections turned into you taunting me about my disregard for numbers.
Because I asked a question in this thread, unrelated to the FIP/ERA discussion. I asked about how this data has affected your opinion of how good the team will be. And you basically said that because you disagree with some of the individual player projections, you’re disregarding the whole thing. As you said, this data has affected your opinion “not at all.” So it sounds like you’re ignoring the CAIRO projections because you don’t like the negative data. I have a strong feeling that if it had some really optimistic projections instead of some very pessimistic ones, you wouldn’t be throwing out the data. You’d be relying on it.
It seems like you’re saying that there are several CAIRO player projections that seem way off to you. Other than Hosmer, which ones are those? I asked this question above. Will you answer it now?
You may know me as NYRoyal.
by Scott McKinney on Dec 19, 2011 10:51 AM EST up reply actions
First, I try not to respond to your positivity thesis comments, but I’ll note that you’re getting off-track here. Debate the ideas, not the person.
Second, I don’t want to go line-by-line on every projection in this set because every projection I cite will turn into a never-ending argument. Let’s just say I’m unconvinced by CAIRO’s methodology (which is apparently just Marcel with a couple extra inputs, tho I’m not sure it’s even that good) and I find several of the results questionable.
If we’re going to have a long drawn-out argument over projections, I’d rather wait until wait until we have a better product to argue about. Just wait a couple weeks and then you can tell me about how I don’t understand statistics and that I ignore data that doesn’t suit my emotional needs.
Just wait a couple weeks and then you can tell me about how I don’t understand statistics and that I ignore data that doesn’t suit my emotional needs.
I don’t think I’ll be saying the former. We will likely debate the meaning of some stats and which should be used for this and that. With regard to the former, we’ll see how you react to ZiPS and PECOTA, which will undoubtedly have some projections you disagree with, because no one will agree with every projection. The important part is where one goes from there. Does one just recognize that projections will have some which are somewhat high and some which are somewhat low and recognize their overall value, or does one just see some projections they disagree and then throw the whole thing out.
With regard to my “positivity thesis,” I think pointing out bias is both fair and relevant.
You may know me as NYRoyal.
by Scott McKinney on Dec 19, 2011 12:08 PM EST up reply actions
With regard to my "positivity thesis," I think pointing out bias is both fair and relevant.
Will a prosecutor make headway in a trial by arguing that the defendant’s attorney is biased to argue in defense of his client?
It’s not like I’m a witness that’s supplying evidence here. I just formulate arguments with the evidence on record, so even if I admitted a bias (which I don’t), I’m not sure that it would have any bearing on the legitimacy of my points.
It’s not like I’m a witness that’s supplying evidence here. I just formulate arguments with the evidence on record, so even if I admitted a bias (which I don’t), I’m not sure that it would have any bearing on the legitimacy of my points.
I think that the various choices you make in your analysis of the Royals are strongly colored by your fan bias. It weakens your analysis, in my opinion.
For instance, made your fanpost with analysis which projected 2012 by using 2011 as a baseline (and the only data point) and extrapolating from there, and only using regulars and not the whole roster, you were confident that it was “pretty close to accurate” as a projection for the whole team. It had a positive result for the Royals. But when projection system with a good recent track record came out and a team projection was created from that using the entire roster with realistic playing time estimates for each, you rejected it completely because you didn’t agree with some projections. It had a not so positive outcome for the Royals. Picking the methodology of your choosing for coming up with projected team record, this is a clearly better projection of the 2012 Royals. And yet you prefer your very simplistic analysis as think it is “pretty close to accurate.” You really don’t think that has anything to do with the results? You really prefer your slip-shod methodology and think it provides a more accurate result? You can’t even see the clear bias that weakens your analysis?
I doubt you consciously do it, but I think your analysis strains to come up with a positive result for the Royals. You are free to ignore any such comments from me. But I’m going to continue to point out fan bias when I see it. If it affects one’s analysis, it is worthy of note.
You may know me as NYRoyal.
by Scott McKinney on Dec 19, 2011 1:15 PM EST up reply actions
My point is that you where you disagree with me, you should simply argue for the opposing position.
Arguing about my motivations doesn’t strengthen your position—it simply draws attention away from the initial topic and attempts to put me on trial instead.
As much as I’d love to directly respond to your arguments about my capacity for analysis, they’re completely irrelevant to the topics we’ve discussed in this thread, and we’re already cramped on the right side of the page.
???
“And putting all of that aside, given the numbers from the CAIRO projections, which project the Royals to have somewhere between 73 and 78 wins "
How do you figure that? Your point estimate for the average was 78, so it could vary equally likely on either side of that number.
by BlueEyes_Austin on Dec 20, 2011 1:45 PM EST up reply actions
There are multiple ways to process the CAIRO projections into a XWL for the Royals
The way Scott did it in the fanpost, it came out to 78. We did it another way in the comments and it came to 73. The way I’d run the numbers (using RA instead of FIP), it might be below 70.
CAIRO thinks the Royals have Alex Gordon, Billy Butler and a heaping pile of shit.
Side note on Marcel/CAIRO pitcher ERA projections
I don’t know how SG does it with CAIRO, but since he says that it is basically Marcel with Park adjustments and MLEs, it is worth noting that Marcel’s ERA (or at least one of the ERA fields) is a bit more sophisticated than just going off of past ERA — it partly done on a basic component ERA (I think using BaseRuns) according to Tango.
Making watching baseball as fun as doing your taxes.
My Twitter feed.
Before getting tweaked, read up on regression.
by Matt Klaassen on Dec 17, 2011 3:29 PM EST up reply actions
I dont know if this has been talked about before
but my buddy and I were just throwing things around the other day.(over a few beers)
I would love to see what people think about this idea also I wounder what it would do to our win total projection.
-
The Royals trade for Hanley Ramirez. I know right… WTH is this guy talking about!
-
The Marlins really have an issue on their hands right now. I know he is owed plenty of money in the next 3 years($46.5mil) but adding him at SS would do plenty to our roster. I dont know how Escobar would grade out at 2nd but can you imagine the defensive value of having Ramirez and Escobar up the middle? You also gain plenyy in the hitting department with Ramirez/Escobar over Escobar/Giavotella. I really wonder what the Royals would have to give up acquire him. That might be the biggest issue. You are getting 3 years of an All-Star SS who had a bit of an injury bug last year…
-
Thoughts?
What a year! At least we have the Royals... Oh Wait!
Who would you trade for him?
Also, there’s the deal with Gio at 2nd, who might turn into a good offensive player who is a heck of a lot cheaper and under control for longer. Gio’s total minor league slash in his career is .305/.375/.437 and if he can put up major league numbers like that, I’d take that at the league minimum over Ramirez’s $46.5 million. I know it’s a gigantic if, but I think you gotta give him the chance.
It’s an interesting idea for sure. If Escobar was still on the team, he could be the super-utility guy. He would be the rare utility guy whose defense at each infield position would be better than the starter at each…hmm. And his bat might be such a liability that he is forced into utility status.
I like Gio
but I really dont see those stat lines anywhere near that in his future. I dont know the actual number on his advanced defensive stats but the eyeball test is not all that great. That is not just about his size but his range, reaction time, etc. I will just use his ESPN stats in the field this year(dont forget to account that it was his first 1/3 season in the MLB)
-
Fld %- .972 This would be good for 18th in qualifying 2B out of 19.
RF- 4.04 This would be good for dead last by a large margin.
-
As far as the players involved in the trade. I just visited the Marlins site and they are talking about a front end pitcher prospect and a power bat(any position really). This is where it gets real tricky for most Royals fans. We have been ingrained not to let prospects go…
What a year! At least we have the Royals... Oh Wait!
Hanley hit .243/.333/.379 last year
I know he would certainly bounce back some, but unless their is a “buy-low” concession in any trade, Its probly best to save the prospects and cash and go with the status quo.
This is the kind of deal you might make a year from now when the youngsters have fleshed out. I would rather see what we got and save that cash for a….(wait for it)….Zack Attack Comback.
Trademark pending.
Yeah...
I am not going to look too much into a season like that when he was injured for most of the season. He is a legit .300/.375/.500 guy. Not to mention he gets you anywhere from 20-35 steals. I dont think there is a better time to buy on him. If he really is mad about moving to 3rd you have multiple things going for us on his trade value.
1. Coming off an injury down year
2. He would be a cancer in the clubhouse if the reports are true
3. He is coming into the highest part of his 6 year contract.
Personally I would rather have a sure thing over the chance our prospects are going to pan out. Another year like the one Montgomery had in AAA or the MLB and his value would take a HUGE dive.
What a year! At least we have the Royals... Oh Wait!
The guy played in every game until Aug 2.
I suppose Carl Crawford was “injured” all season as well yes?
Moot point. He will never play here. Nor should KC pursue him.
Really?
Do you know to do research? There is an amazing thing called ESPN! It breaks down EVERY game for you!
He missed 2 sconsecutive games in April, missed 1 game in May played in 2 then missed 14 consecutive games played 1 then sat out another game. He then missed ever game after Aug 2. He was on the DL TWICE last year…
Are you really trying to throw away 5 straight years of production because of a season he played 92 games?
As far as your “moot” point. THIS IS WHY IT IS CALLED A CONVERSATION!!! If you dont like talking about stuff then wait for the news to come out on ESPN. So all of the post on this site talking about pitchers or any players are “moot” points because I am pretty sure 99% of those will never happen? You talk about this being a “moot” point yet you throw out Zach! Pot Meet Kettle
What a year! At least we have the Royals... Oh Wait!
No..No..No..please...I like Eco on short stop with Gio at 2nd!
Last I heard, Hanley will not be traded..so let him stay with the Marlins this year! Would love to keep our regulars for this new season. I was heart broken when KC broke my Trio in the outfield. I know it was for a good cause and because we need some more pitchers. But still sadden, we had to trade “Melky”…..and wish Melky well! He will be miss not having him with KC!
by Fans-of Eric Hosmer on Dec 17, 2011 4:15 PM EST up reply actions
If Hosmer is only going to give us 0.59 WAR, then why don’t we just get rid of him? Clint Robinson would give us more WAR, for goodness sake.
This is ridiculous. If Hosmer has 0.59 WAR in 2012, we might as well just give up and start following some other team.
I'm sure the Yankees would take HOZ within a heartbeat!
by Fans-of Eric Hosmer on Dec 19, 2011 1:23 AM EST up reply actions
In addition!
Time will tell and if some feel Hoz should be traded based on perdictions or data used to evaluate players! All I can say is….the eyes can see very well and he played and awesome Rookie Year! If KC Royals don’t want me…I know plenty of teams out there now…that would pick HOZ up! He had many walk-off…did not so bad with home runs and started late in the season..ended up pretty well with his batting average and still did great work in defense! In addition, I recall many bad throws he scopped up or some that I thought would be over his head..and those plays were caught and the out was made…on the other hand if he were to miss those bad throws…they would of been error on the player who made the throw. Sorry, but perdictions are just perdictions..and the data…I don’t take them seriously..because they are off and not accurate! I believe most of the players above are going to do much better then the perdictions provided. Article is good…but I disagree on the perdictions.
by Fans-of Eric Hosmer on Dec 19, 2011 1:31 AM EST up reply actions
I would say that Butler, Hosmer, Francoeur, Cain, Moustakas, Perez, and Giavotella
All out-perform those projections to varying degrees.
The pitching projections seem much more realistic to me.
We should trade for Vance Worley.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Royals Review
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
