2010 Central Division Projected Standings
by Jon Kammerer on Dec 13, 2009 1:02 PM EST
Email
PrintBoth defensive and offensive projections were obtained at baseballprojections.com. CHONE is the name of the projection system. I then added in the team baserunning totals found at BP from 2009 to the WAR for position players. I gave starters 600 AB and reserves 100. I used innings projections from CHONE for starting pitchers.
This is only an estimate of each teams true talent. Teams are usually within 7 games, plus or minus, of their true talent. I also realize that the offseason has just begun so every team has the opportunity to improve. I'd have posted the excel documents but I can never get them to format correctly.
Twins 87.9 Wins. Position players: 22.4 WAR, Starting Pitchers: 12.3 WAR, Bullpen 4.8 WAR. Biggest hole: 3B/LF at 0.6 WAR.
White Sox 84.4 Wins. Position players: 17.0 WAR, Starting Pitchers: 13.5 WAR, Bullpen 5.7 WAR. Biggest hole DH at 0 WAR.
Indians 82.7 Wins. Position players: 20.4 WAR, Starting Pitchers: 9.2 WAR, Bullpen 4.8 WAR. Biggest hole 1B at 1 WAR.
Royals 78.3 Wins. Position players: 14.2 WAR, Staring Pitchers: 11.1 WAR, Bullpen 4.6 WAR. Biggest hole RF at 0.4 WAR.
Tigers 78.0 Wins. Position players: 16.1 WAR, Starting Pitchers: 10.1 WAR, Bullpen 3.3 WAR. Biggest hole at CF at 0 WAR.
66 comments
|
0 recs |
Do you like this story?
Comments
He’s pretty optimistic about Delmon. I’d be pretty happy with 0.6 WAR at this point. Did you use his defensive projections or did you do the Marcels+fans scouting report thing?
Kevin Kouzmanoff and a decent 4th OF (Ryan Church?) would put us over the 90 win TT mark. I doubt the tigers, Indians, or royals do much upgrading, but the White Sox will sign a DH. If its a guy like Matsui they’d be right there with us.
One thing that’s probably not accurately acounted for here is depth. I think the Twins can handle an injury or two on the pitching staff a lot better than the Sox. It will be interesting to se what these numbers are in march after the rosters are set, with more accurate playing time projections.
by lookatthosetwins on Dec 13, 2009 1:38 PM EST reply actions
CHONE
I used CHONE for the defensive projections. It didn’t make much difference on the team level. Using weighted averages of UZR, good defenders project better and poor defenders project worse but on the team level it was a wash. Delmon projects as 0 WAR using my numbers.
Cust projects as a 2.1 WAR DH, Guerrero 1..3, Matusi 1.2, Thome 0.8 WAR. A 3 WAR DH is a Justin Morneau caliber hitter. The only free agent hitter at that level is Matt Holiday.
Chone projects Pridie as a +7 CF. That makes him worth 0.4 WAR over 300 ABs. I’d much rather see them pick up a Kouz, Glaus, Hudson, Lopez, KJ than a 4th outfielder. Pridie isn’t great but I don’t think he’ll kill us even if he had to play regularly.
by Jon Kammerer on Dec 13, 2009 2:49 PM EST up reply actions
Yeah, Kouz and KJ would be ideal
really solves both problems there. KJ can backup LF and RF and Pridie will backup CF once in a blue moon. Those two and we’re in the 91-92 win region.
I guess I just think that what is most likely to happen is we sign/trade for a league average or so 3b, and sign a guy like Quinlan to backup the corner OF spots and pinch hit. That probably puts us at about 90 wins.
by lookatthosetwins on Dec 13, 2009 4:50 PM EST up reply actions
The Bullpen seems light to me. 4.8
Neshek, Delaney, Liriano,Duensing all unknow, but that looks strong to me.
Duensing and Liriano
Both are included as starters. Due to injury histories Pavano and Slowey have low innings projections. Duensing and Liriano are probably the 5th and 6th starters. I used Manship as our 7th starter.
by Jon Kammerer on Dec 13, 2009 2:22 PM EST up reply actions
I was thinking the bullpen would be stronger also.
Are you using leverage index to calculate their WAR? And if so, what numbers are you using for it?
by lookatthosetwins on Dec 13, 2009 8:36 PM EST up reply actions
leverage index
WAR uses half of the leverage index between expected and 1. So Nathan gets 1.5. Guerrier gets 1.3 and so on. This is because Nathan’s replacement is not a AAAA guy but the next best guy in the bullpen. Here’s a full explanation. http://www.beyondtheboxscore.com/2009/4/29/856308/bullpen-chaining-and-reliever-war. Last year’s bullpen was worth 3.2 WAR. So projecting 4.8 WAR is a big improvement. This is mostly accomplished by getting rid of the negative contributors like Dickey and Henn.
by Jon Kammerer on Dec 13, 2009 10:52 PM EST up reply actions
That's interesting... I've always used 1.8 for closers
mainly because it says to in Sky’s team WAR worksheet. It looks like he wrote this after he posted that last year though. As far as thinking 4.8 was low, I was more looking at compared to other teams. I spose guys like Juan Cruz and Kerry Wood are due to bounce back.
I do think that we have a lot more of a cushion to withstand a major injury than the other teams though. I’d put my money on our pitching staff being better than the White Sox as a whole when its all said and done.
by lookatthosetwins on Dec 14, 2009 12:20 AM EST up reply actions
Upside
Slowey only projects for 135 innings and 2.0 WAR. He’s a 3WAR player if he hits 200. Pavano projects for 1.4 WAR in 116 innings. If he hits 200 that adds another 1 WAR. The team projection is already counting 200 innings of Duensing/Manship. The White Sox have a higher WAR for their starting pitchers because 1-4 they project for more innings.
I was mistaken above. I used 1.8 for Nathan’s LI. I am also working of Sky’s WAR worksheet. I had assumed he’d used the numbers in the above noted discussion. It shouldn’t make too much difference for the divisional projections because I used the same LI for each team.
by Jon Kammerer on Dec 14, 2009 12:59 AM EST up reply actions
The Tigers last?
I know they gave up Polanco and Granderson, but still…
"Is it normal to wake up in the morning in a sweat because you can't wait to beat another human's guts out?" -Joe Kapp
by less cowbell, more 'neau on Dec 13, 2009 5:13 PM EST reply actions
Especially with the Royals in the division
"So [Kouzmanoff]’s going to make decent monkey, but he’d be affordable." - Jesse
by what_would_gil_thorp_do on Dec 13, 2009 5:18 PM EST up reply actions
Yeah, I was kinda wondering about that...
but then I realized that last year, a lot of sports writers and analysts picked the Royals to take the division. Who knows.
"You can't sit on a lead and run a few plays into the line and just kill the clock. You've got to throw the ball over the damn plate and give the other man his chance. That's why baseball is the greatest game of them all." -Earl Weaver
And Rodney, and Edwin Jackson, and Lyon
Polanco and Granderson alone account for about 7 WAR, so that would take them from 85 to 78. As it stands, thats a pretty poor team.
by lookatthosetwins on Dec 13, 2009 5:48 PM EST up reply actions
Well, they are having quite the fire sale
Even if they do get Phil Coke. I guess we’ll see how everybody looks this spring.
"Is it normal to wake up in the morning in a sweat because you can't wait to beat another human's guts out?" -Joe Kapp
by less cowbell, more 'neau on Dec 13, 2009 6:21 PM EST up reply actions
Another way to look at it is that Polanco and Granderson together had similar production to Joe Mauer
how many games would we have won last year without Mauer?
by lookatthosetwins on Dec 13, 2009 5:51 PM EST up reply actions
Look back to April
We did’t have Joe for a month.
by b1 on Dec 13, 2009 7:17 PM EST up reply actions
Poor Detroit
That big offseason with the Cabrera/Dontrelle deal is looking pretty awful right now – they’re now so strapped for cash that they’re holding a fire sale while still on the hook for eleventy billion dollars to Cabrera. I know that it doesn’t take much to be competitive in the AL Central, but when your finances dry up, that makes it hard to rebuild very quickly.
"There are only two things that are infinite, the universe and human stupidity, and I'm not sure about the former." - Albert Einstein
let that be a reminder
for everyone screaming about paying Joe Nathan “whatever he wants” or shelling out big bucks/multiyear Ks for risks like Harden or Sheets or even Beltre.
by montanatwinsfan on Dec 14, 2009 12:44 PM EST up reply actions
Cabrera near MVP
The deal isn’t looking that bad. One guy even thought Cabrera had a better year than Mauer ?!?!?!?!?!
No, wait, that deal was terrible.
The trade wasn't that bad
Signing both to long-term deals was a disaster.
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
Well...
Even signing Cabrera to a long-term deal wasn’t a huge mistake, in and of itself – he’s likely to be a productive player for the length of the deal and basically earn that money (I’m not going to try and defend the Dontrelle signing – he was extremely high-risk when they acquired him). The problem was that their financial situation got so much worse than they expected that they now can’t afford to surround him with anything – the combination of the economy (especially in Michigan) and Cabrera’s contract is the real issue.
"There are only two things that are infinite, the universe and human stupidity, and I'm not sure about the former." - Albert Einstein
Yeah, but that's on Dombrowski
You don’t increase payroll by 40% without any new revenue streams and expect to sustain a competitive team. Even if the Michigan economy had held steady (which few economists were predicting), they couldn’t afford those contracts and guys like Granderson, Verlander, and Porcello long term.
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
87 wins
We’ve added Hardy and will have Pavano for a whole season… I’d imagine this team should be a low 90s team if we do nothing else…
by diehardtwinsfan on Dec 13, 2009 10:21 PM EST reply actions
Regression to the mean
Projection systems use 3 year weighted averages and then regress to the mean. The amount of regression depends on the sample size. So Joe projects to be all-world next year and not all-universe. That’s a loss of 2 WAR. Cuddyer and Kubel also project to be less valuable after career years. I’ll post all the WAR projections for Twins when I get a chance.
Out of curiousity
How much of the Royals’ rotation’s 11.1 WAR is from Greinke?
"So [Kouzmanoff]’s going to make decent monkey, but he’d be affordable." - Jesse
by what_would_gil_thorp_do on Dec 14, 2009 1:49 AM EST reply actions
Mauer 6.5
Morales 0.4
Morneau 2.8
Cuddyer 2.2
Punto 0.9
Tolbert 0.3
Hardy 2.9
Harris 0.6
Young 0.5
Pridie 0.4
Span 3.5
Kubel 1.3
Baker 3.6 in 183
Blackburn 2.0 in 173
Slowey 2.0 in 135
Pavano 1.4 in 116
Liriano 1.6 in 135
Duensing 1.4 in 159
Manship 0.3 in 39
Nathan 2.5
MattG 0.5
Mijares 0.3
Neshek 0.3
Rauch 0.5
Delaney 0.2
Crain 0.3
Keppel 0.1
Baserunning is included on the individual level for Twins. This gives Punto, Span, Hardy and Tolbert a boost. It knocks Kubel down a notch.
Brendan Harris projects > Delmon Young
Has anyone here suggested not tendering Delmon Young a contract?
ok a brief explanation is needed
how can every one of our players offer a positive contribution. This obviously is not weighted against a mean, or against an “average” player.
I thought WAR is Wins Above Replacement, but your list suggests that EVERYONE on our team will be better than an replacement level player at that position…Yankees look out here we come!
by montanatwinsfan on Dec 14, 2009 12:48 PM EST up reply actions
Flaws of a projection system
Projection systems tend to project players to be better than replacment or they assume the player will be replaced quickly. It’s sort of “if he plays the whole season, he’s not going to suck that bad”.
Another comment on projection systems
What is left out of projectors is also important. Catcher defense is one and our C defense is WAY better than the Whiteys. Managing the bullpen is another one (using players at max leverage).
CHONE projects catcher defense
Joe +5 runs
Morales -11 runs
by Jon Kammerer on Dec 14, 2009 2:31 PM EST up reply actions
Well
I’m not sure that projecting all of these players at above replacement is a flaw. While some of them may perform below that level, presumably they will be replaced pretty quickly.
Further, projection systems projections are always, intentionally, bunched away from the real life extremes that will inevitably occur. We complain about that when systems project big regressions from players who are coming off big years, but it works at the low end as well.
by Eric in Madison on Dec 14, 2009 1:40 PM EST up reply actions
Replacement Level
A team that consisted soley of 0 WAR players would be expected to win roughly 48 games. A replacement level team would be historically bad. A 0 WAR player should be freely available for the league minimum. Either as a AAA callup or minor league free agent.
by Jon Kammerer on Dec 14, 2009 2:37 PM EST up reply actions
True, not all players can perform above replacement level
In fact, in 2009 we had our share of overall below replacement performances, according to Fangraphs:
Tolbert: -0.2 WAR
Harris: -0.2
Young: -1.3
Casilla: -1.4
Humber, Gabino, Morillo, Dickey, Henn, Breslow all between -0.1 and -0.3 WAR as pitchers.
The projections somewhat account for this, as the best players’ projections are also tempered a bit. But in the end, you need to account for this “inflation” when projecting WAR for teams. Last offseason, I projected all major league clubs, then normalized (about a 2-3 win adjustment for each team by adjusting replacement level wins) to ensure that across the major leagues there was a .500 record.
by Adam Peterson on Dec 14, 2009 3:39 PM EST up reply actions
Good point
I only had the time and inclination to run the central divisions teams. I do think it gives us a good idea of where the Twins are in relation to the competition.
by Jon Kammerer on Dec 14, 2009 3:53 PM EST up reply actions
Very true
and I totally understand the time/inclination. At this point, I’d probably take 2-3 wins off every team’s projection to balance. As we all fill in pieces, I expect the division to shake out not too far from where you have it.
by Adam Peterson on Dec 14, 2009 7:09 PM EST up reply actions
Replacement is not average
Like Jon said, a “replacement-level” player is a player who is considered freely available, like a minor-league callup or street free agent that would likely sign for the minimum – usually, AAAA guys like Jason Tyner, Randy Ruiz, and Matt Tolbert. League average players are generally worth somewhere in the neighborhood of 2 WAR (I think), as an average player is better than a replacement player. In theory, a team of replacement-level players would go 48-114, while a team of average players would go 81-81.
In the projections, about half the Twins’ starting lineup is projected as above-average.
"There are only two things that are infinite, the universe and human stupidity, and I'm not sure about the former." - Albert Einstein
Put another way
Every player we project should be above replacement, otherwise we shouldn’t be including them in the projections.
by Adam Peterson on Dec 15, 2009 2:03 PM EST up reply actions
Does anyone check projection systems for accuracy after the fact?
My sense is projections systems have a lot of room for improvement. One thing that seems to be missing from them is upside. Perhaps the average player at a given age with a certain performance record will tend to play to his projections. But a lot of very talented players outshine their projections. And a lot of guys who play over their heads for a period of years regress more. I’m thinking if the projection systems also factored in scouting reports on what the player is likely to do in the future, they might be more accurate. Just a thought.
"You're thinking too much. Just have fun." -- Bennie "The Jet" Rodriguez in Sandlot
Doublechecking
Yes, the projection systems are regression testing their performance in order to improve the model. However, what they don’t all do is test whether they’re predicting team wins. The correlation between team WAR and team wins isn’t perfect.
players vs teams
The projection systems are going to be more accurate on the team level than for any one particular player. You can compare it to a mutual fund. Some individual stocks will overperform, some will underperform, but in the aggregate you generally get the expected rate of return.
Projection systems are really just an estimate of the true talent of a team. Which is important to make good offseason decisions. The Mariners traded their future(Jones, Sherrill, Tillman) away because they misjudged how good their team was comming off an 87 win season.
by Jon Kammerer on Dec 14, 2009 2:59 PM EST up reply actions
Comparing my own projections a year later
In fact, at some point during last year’s offseason I had a Fan Post with pretty similar analysis of the AL Central as of January. Going back to my spreadsheet (all numbers rounded):
1. Cleveland (90 wins)
2. Minnesota (87 wins)
3. Detroit (85 wins)
4. Kansas City (79 wins)
5. Chicago (78 wins)
In general, I think it turned out fairly well. Minnesota, Detroit and Chicago were all within one win of their real-world performance, KC underperformed, and I was nowhere close to projecting Cleveland correctly.
by Adam Peterson on Dec 14, 2009 3:33 PM EST up reply actions
Cleveland was understandable. They ran into some bad luck and then traded away their best player
by lookatthosetwins on Dec 14, 2009 3:42 PM EST up reply actions
Sizemore
he was ineffective or on the DL most of the year. Losing those 3 would be like having Mauer hurt all year and then trading Morneau and Span.
by Jon Kammerer on Dec 14, 2009 6:00 PM EST up reply actions
All 3 were hurt last year
Span played 145 games,Mauer 138 and Morneau 135 for 7th on the team.
by b1 on Dec 14, 2009 6:22 PM EST up reply actions
Also the loss of Slowey
I’m just noting this because I think we could be much better this year.
by b1 on Dec 14, 2009 6:25 PM EST up reply actions
Traded away Lee and Martinez
and Sizemore spent a lot of time on the DL or at less than 50%.
Who else am I missing? Major problems at the back end of the bullpen during the highest leverage innings? Yep.
by Adam Peterson on Dec 14, 2009 7:11 PM EST up reply actions
Not really
more of combination\ of multiple projection systems (CHONE, etc) and weighted UZR. I focused quite a bit on playing time, who I thought would get the AB/IP given the current roster.
I’ve been planning to post Twins WAR projections in the next week in an attempt to give us a baseline from which to look at trades, signings, etc. You beat me to the division WAR projections, which is great.
by Adam Peterson on Dec 15, 2009 6:53 AM EST up reply actions
We definitely still need one with better playing time estimates
One of the key things to remember is who’s playing time will be affected by any signing. If we signed both a 3b and 2b (unlikely) Punto would still probably end up with 500 pas. Tolbert and Casilla (replacement level or worse) would be the ones who’s playing time would get completely taken away. Because of that, any infielder (and probably any 2 infielders) is almost completely worth their projected WAR, which wouldn’t be true if, say, we signed a starting pitcher or reliever who would be replacing an above replacement level player.
by lookatthosetwins on Dec 15, 2009 3:44 PM EST up reply actions
I'll post in a couple days
but I’ve actually done four separate WAR projections for the Twins. First with actual results from the 2009 season, then with:
1. The roster right at the end of the season, i.e. no free agents
2. After the Gomez-Hardy trade and releasing Buscher
3. Keeping Pavano and letting Boof go.
Gives me a good idea of how the moves have helped the ballclub, including a breakout by position. More to come!
by Adam Peterson on Dec 15, 2009 7:57 PM EST up reply actions
That sounds great
I think WAR projections work better for looking at the differences between a team before/after a trade and for comparing teams than they do for projecting an actual number of wins. I have a pretty flawed worksheet with a WAR projection for the twins. It might not be that accurate overall, but I think it works pretty well for assessing trades. After hearing rumors of an Uggla/Nolasco trade, I plugged them in and had us at 95 wins! If only…
Anyway, I’m looking forward to your posts. I’ve heard that combining different projection systems works well – getting rid of the biases each has – so it will be interesting to see.
What I’d really like to see, is taking all of this and relating it too playoff odds. That might be pointless to do this early in the offseason, but it would be fun to see how it changes as players are added and subtracted throughout the division. Anyway, I don’t know if you have the necessary information to do something like that… just a thought.
by lookatthosetwins on Dec 15, 2009 9:35 PM EST up reply actions
When you say, usually within 7 games
You’re talking about actual true talent, not projected true talent, correct? I have to think they vary quite a bit from the projections.
by lookatthosetwins on Dec 14, 2009 3:43 PM EST reply actions
A bet?
What odds would you need to take the under on 81 wins for the 2010 Twins as currently constructed? 2 to 1? 5 to 1? 10 to 1? How about the over on 95 wins?
Projections are just an estimate of a team’s true talent. We can never know precisely because of sample size issues and the fact that players can become better or worse than their track records suggest.
As currently constructed
I’d say 3 to 1, maybe 2 to 1 odds. There’s always the chance of an injury to a major player, Mauer/Morneau would be obvious, but with the team as currently constructed, what happens if Span or Hardy go down? Pridie every day in CF? An infield of Harris-Punto-Casilla/Tolbert?
If we acquire one or two additional solid position players, it’s much tougher to reach the downside and I’d say at least 5 to 1 odds.
by Adam Peterson on Dec 14, 2009 7:14 PM EST up reply actions
Their's a couple of unknows that will help determine
the season. Hardy? Slowey? Liriano? Neshek? and 3B. Their’s no way to know how any of them will do. We’ll have to wait for Spring games to get a hint.
White Sox will fill that DH hole
Rumors are Hideki Matsui
Matsui is an Angel 1yr/$6.5M
Agreed that they’ll find someone. probably in the 1-2 WAR range. CHONE projects Matsui for 1.2WAR. Vlad, Thome, Cust would all be roughly 1-2 WAR DHs.
by Jon Kammerer on Dec 14, 2009 10:44 PM EST up reply actions
Juan Pierre
Sox pick up Pierre. Sliding Andruw Jones to the bench/DH. Net gain 0.6 WAR. Sox still could use a DH. Jones is replacement level for DH.
If that's all they do, I'll like our chances going in.
I have to think they’ll end up signing a lefty, probably Thome or Cust to platoon DH. Jones/Thome should be close to league average if used as a straight platoon, don’t you think? So they’ll end up with an 86-87 win projection in that scenario.
For us, probably just trade for Kouzmanoff or sign somone in the 2 WAR range, putting us in the 90 range. What are our odds of finishing better than the WS in that scenario? 55% 60%?
by lookatthosetwins on Dec 15, 2009 10:09 PM EST up reply actions
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recent FanPosts
In Association With
Big League Stew - MLB - Yahoo! Sports
- Photo: Padres fan regrets letting mother capture moment with Trevor Hoffman
- Marlins agree to unretire No. 5 for Logan Morrison, who will honor his father’s favorite player
- Weatherman Derek Holland does Dallas TV
- Ask Alex: Will Julio Teheran or Matt Moore win rookie of the year?
- Ku Klux Klan, umpires and bowling: Why the minor leagues tossed out Jim Evans academy
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.
