Best offense in the league?
by Maddog on Feb 18, 2007 8:54 AM CST
Email
PrintBaseball Prospectus finally released their depth charts today and it has the Cubs team hitting .275/.337/.460. The depth charts are basically the intelligent look through the team's PECOTA projections and assigning playing time at each position. No more Tyler Colvin or Mark Pawelek's in the projections. For example, BP has Soriano projected to get 90% of the playing time in CF followed by Felix Pie and Angel Pagan each at 5%.
The Dodgers had the highest batting average in the NL in 2006 at .276 (2nd best team was at .270). The .275 mark would rank them 2nd in average.
A .337 OBP would have them tied with St. Louis and Atlanta for 5th in the National League.
Atlanta had the highest slugging in the league last season at .455. BP has the Cubs team projected at .460 (easily the best in the league for 2006 standards).
Philly had the highest OPS in the NL in 2006 at .794. If you can add, you know the projected team OPS for the Cubs is .797. OPS has a very high correlation with run scoring. Philly scored 865 runs in 2006; 16 more than runner-up Atlanta and they were one of only 5 NL teams to score more than 781 runs.
So if these projections prove to be true and 2006 was an average season in the NL, the 2007 Cubs will likely have the best offense in the National League.
But hold on. Their projected starting 5 for the Cubs is Carlos Zambrano, Ted Lilly, Rich Hill, Jason Marquis, and Sean Marshall with Prior, Miller and Guzman as spot starters making 11, 11, and 6 starts respectively. They have a total of 18 pitchers pitching for the Cubs in 2007 (Jae Kuk Ryu included -- obviously these were completed before the trade). These 18 pitchers (8 starters and 10 relievers) combine to go 74-67, which is about right on line with what many of us have been suggesting the Cubs are capable of going (7-10 games over .500). That's probably enough to win the Central in 2007.
Based on these depth charts and projections, the Cubs are going to have a powerful offense and a very mediocre pitching staff (if not slightly below average). From a quick glance through, though, it does look like the offense will be enough to carry the pitching.
I spent about 5 minutes writing this as I was looking at the depth charts and checking 2006 stats so if I made an error, I apologize.
This is a FanPost and does not necessarily reflect the views of SB Nation or Al Yellon, managing editor (unless it's a FanPost posted by Al). FanPost opinions are valued expressions of opinion by passionate and knowledgeable baseball fans.
61 comments
|
0 recs |
Do you like this story?
Comments
This summer at Wrigley
Or maybe the pitching staff will surprise. Crazier things have happened.
by nextyearcub on Feb 17, 2007 4:34 PM CST reply actions
Somehow I can see the Cubs figuring out...
But very nice analysis as always maddog. You are a giant amongst us.
by theprognosticator on Feb 17, 2007 4:34 PM CST reply actions
Thanks.
you're welcome...
You obviously enjoy the numbers and you make it enjoyable and engaging for the rest of us as well.
Thanks again.
by theprognosticator on Feb 17, 2007 4:49 PM CST up reply actions
I enjoy stats.
I know
No.
I've learned
Is that statement
I dig your optimism...
Only the Cubs could figure out a way to lead the league in OPS and be dead last in runs scored.
Probably by setting records for most times picked off, and most times getting thrown out at home plate.
by theprognosticator on Feb 18, 2007 3:33 AM CST up reply actions
See, I am the exact opposite..
It has to one day, right?
by TheBeerBaron on Feb 18, 2007 10:30 AM CST up reply actions
one would think so...
by theprognosticator on Feb 18, 2007 3:48 PM CST up reply actions
See...
That's not likely to happen this year, wouldn't you agree? Either Prior and Miller are going to be full-rotation starters, or be traded, or be on the DL all year.
Guzman likely spends the year at Iowa.
18 starters? In 2006, a terrible pitching staff, only 15 different pitchers started a game, and seven pitchers started 128 of them. I'd think the Cubs will have FEWER starters in 2007, not more.
18 pitchers.
I think those spot starts are not only based on last year's performances for the three of them, but the last several years. Furthermore, today in the Tribune, this was written: Piniella said he's looking for a long reliever, and the loser of the competition for the fifth spot involving Mark Prior, Wade Miller, Marshall and perhaps Neal Cotts could be that man.
There are obviously more choices to be had than pitchers for that 5th spot and there will be several losers. It's distinctly possible, Al, that both Prior and Miller fail to claim that 5th spot out of spring training and Guzman has a great spring. Then, with his injury history, it's not at all unlikely for him to go down after 6 starts leaving the 5th spot open for Prior, Miller, or Marshall.
Prior's stuff last season was just plain bad. He was worse than Jason Marquis and Glendon Rusch. Injuries or not, he was bad. It takes a leap of faith to believe he's capable of making 20 starts and actually earning that 5th starter's spot at this point.
I'm just glad to hear that Lou isn't going to give him the job if he doesn't earn it. I've been afraid that no matter how poorly he pitched, if he was healthy it was his job. According to that article, Lou has said he's going with the best pitcher.
5th starters for teams generally make about 20 starts and teams (healthy teams) have 3 to 4 guys who make a handful of spot starts.
I don't think those projected playing time numbers are that far off. They may end up being way off, but do you feel confident that either Prior, Miller, or Guzman will be good enough to earn that spot? I sure don't.
18 pitchers....
Everything else you say here makes sense, too, but generally, teams that are of contending caliber don't have more than perhaps 8 or 9 men start games during the season.
Guzman could have a great spring and Prior and Miller might suck, true enough. That doesn't seem all that likely, though, at least not all THREE of those things happening.
Finally, I agree. Piniella's going to do the right thing, make the guys earn their spots (well, except for the obvious one -- Z -- he'll be in the rotation regardless). I'm liking Piniella more and more with everything he does, and I was one of the ones saying loudest that he shouldn't have been the guy.
I'll be happy to be wrong about that.
I've always liked Piniella.
I don't see all 3 of those things happening either, but I also don't see Prior or Miller making more than 11 starts. And, yeah, I'm basing that off their past performance, which is all I have. If I had to give my predictions for number of starts it would be something like (this won't add up to 162 so don't bother)...
Zambrano -- 33
Hill -- 32
Lilly -- 30
Marquis -- 30
Prior -- 15 (I'm being generous)
Miller -- 10
Guzman -- 3
Gallagher -- 2 (I think we'll see him in September)
155, pretty close.
Z: 34
Hill: 33
Lilly: 33
Marquis: 30
Prior: 18
Miller: 10
Misc (Not predicting who): 4
starts
Oh, it's Kool-aid time!!!!
Lilly - 31
Hill - 31
Prior - 30
Marquis - 21
Miller - 20
Marshall - 9
Guzman - 4
by TheBeerBaron on Feb 18, 2007 10:35 AM CST up reply actions
I just realized that this would be
Except Prior making 30 starts!!!! It's happening this year is the year!!
by TheBeerBaron on Feb 18, 2007 10:37 AM CST up reply actions
i agree
i think i read
I have a question, Al.
I assume we could agree that past performance is the best indicator of future performance. Obviously you can't predict the future and in only one season there is a wide amount of variance, but that doesn't mean we can't use the tools to come to conclusions. These projections are basically giving us a realistic idea of what a player is capable of if the 2007 season was played thousands of times...what you may call the player's true ability. He may out-perform these projections, under-perform them, and he may be dead on. I don't believe anybody who lists these or does these believes they're going to be accurate for one individual player. As you know, the variance is too great in just one season.
I also assume you are basing many of your opinions about possible future production from players on their past performance. For example, I think you'd agree Juan Pierre isn't going to hit 50 home runs. We know this for a couple of reasons. The first being the years of data we have to look at where he's hardly hit any per season and his size, which is a more flawed way to take a look at future production (small guys have power too and some big guys don't).
When you say that you think Lee is capable of playing quite well (somewhat similar to his 2005 numbers), you are basing that opinion on past performance. You may not know it when you do it, but when you're talking about how good or how bad this team may be, your conclusions are the result of instantaneous projections based on your opinions, which are certainly based on past production.
So I guess my question is this: How are these projection systems so flawed, but your own predictions/projections aren't? You're doing the same thing without the mathematical analysis. You're even considering age (you've mentioned how Murton may develop power as he ages--people who do these projections consider a natural age progression in their formulas). You surely believe that as a player ages into his mid-30s that his production will likely continue to decline.
You're unconsciously doing the same thing that you are saying you dislike. Is it the relatively complex math being brought forward that you dislike? I doubt that's it. Batting average and slugging percentage was once complex math for this game and I'm sure you know that.
Obviously these are flawed. I don't think anybody has said they aren't. I believe I read a study recently on some Angels blog that compared the various projections for hitters who had over 500 plate appearances. PECOTA turned out to be the most reliable having ONLY a correlation coefficient of .51 (slightly higher I believe). Not a very strong correlation at all. But there is one and I'd think you'd agree that any available information that can make the team better is worthwhile information.
Bill James won't even bother projecting pitchers; says it can't be done. And the study showed that the correlation was much lower, but there was still a correlation (zips having correlation coefficient the highest with pitching, PECOTA 2nd).
It's information. It's flawed, but no more so than looking at a player's batting average and RBI's in one single season.
I'm not saying I don't...
I just think that you can't simply boil down projections or predictions to mathematical formulas (as you do above when you say PECOTA has a "correlation coefficient" -- when I hear that sort of thing, my eyes start glazing over).
Sure, I'll look at all of that. I do think that some people -- not you -- wind up being married to these projections, thinking that they are the be-all and end-all, and looking down their noses at those of us who don't bow down to the Great Stats God.
Like I said, this isn't directed at you. There's a middle ground, and I think I try to occupy it. If you think I'm one of those who just looks at the "TV stats", you are quite wrong.
Correlation coefficients and
A correlation coefficient of 1 implies, as you probably know, a linear relationship between two variables (home runs and at least one run scored in the above example). If it's 0, there's no relationship whatsoever (home runs to men landing on the moon). The closer it gets to 1, the greater the correlation between the two variables (OPS to runs scored, if memory serves, is about 0.88--pretty damn high). -1 is an inverse linear relationship (home runs to opposing team's runs prevented).
So, when a team like the Phils score 865 runs with an OPS .005 points lower than the CUbs projection for 2007, we have very good reason to believe that if these projections are accurate that the Cubs SHOULD score as many or more runs.
These kinds of things will become commonplace in baseball in a generation or two.
I believe you.
It used to do that to my eyes as well.
That's what's funny...
All the stats and correlation coefficients are really doing is confirming what they thought (or denying it).
0.88?
A correlation of 0.88
An r (correlation coefficient) is between -1 and 1. If r is negative, it's a negative slope (inverse relationship...one variable increasing causes the other variable to decrease). If it's positive, it has a positive sloped meaning that as one variable increases the other one does as well.
I'm not sure if this image will show up on here, but here's a scatter plot I recently did showing the correlation coefficient between negative VORP and wins:
If the image doesn't show up, check it out in this article. That image shows a correlation coefficient of 0.70 meaning that negative VORP and wins have a moderate correlation to one another, which is basically common sense. I wanted to know how much and since the Cubs had a bunch of negative VORP in 2006, this is one of the ways they can improve the most in 2007. Limit the negative production.
Sorry it's not bigger.
Hurray for stats class.
by sparkles721 on Feb 18, 2007 11:36 AM CST up reply actions
Something that was engrained into my mind
In other words, even if it was a perfect 1, OPS could never be a precise indicator for Runs Scored.
Statistics cannot be an end all be all for all of life's numerical questions.
Statistical Anomalies and Sperrious correlation would still make some of your findings irrelevant and misleading.
Now, as for this article, especially since you are using compound stats, which some consider impure, I ask you can negative VORP and Wins, be sperriously correlated.
You're right.
However, OPS (on-base plus slugging) creates runs. Runs can score without either (errors), but those situations are limited. You need men on base and the more bases you advance on a hit, the more runs you'll generally score (average situations considered).
Not only are those situations in which runs score without the help of OPS limited, it's the defense that is responsible; not the offense.
If the OBP of an entire team is .000, they'll score a minimal number of runs (the only runs that score will be the result of the opposing team's defense). If an entire team's OBP is 1.000, they'll score an infinite number of runs (ignoring caught stolen bases and pick offs, and other baserunning outs, of course).
OBP is but one part of OPS (I believe both parts are equal, and in all honesty would probably favor slugging over on-base skills if it were up to me, but as you can clearly see, as OBP rises, so do runs scored. OBP isn't the cause entirely, but the runs that do score is, generally speaking, are caused after someone gets on base (obp) and numerous bases (slugging) are advanced.
As for the 2nd question (negative VORP and wins), the answer is yes. There's not a lot of real value in the study that I did (unfortunately, perhaps fortunately, all my data has been destroyed since my hard drive crashed and I purchased a new computer). Most of it's common sense (limit negative production and you'll win more games). Most importantly, VORP is a stat with the intentions of comparing the performance of individual players and their impact in scoring runs to the team. I combined the numbers from the players on the team, which probably should never be done in the first place. But I did learn several things.
Dusty Baker has been very bad in his career as a manager at giving out playing time to players who are undeserving. This isn't a revelation as we're Cubs fans and have seen this for years, but the numbers back it up. Lou Piniella has been above average (very, very good in Seattle, but poor before that, and slightly below average in Tampa though better than the manager last year). He plays who deserves to be played and we can expect that from him all season long. If DeRosa isn't performing he's not going to hesitate to put Theriot in there or bring up Patterson or whatever.
Interestingly, only a few teams pay attention to this kind of thing. It was clear that Oakland has some understanding of negative production and so does Boston and Milwaukee under Ned Yost, but that's really it. Ned Yost is a very good manager and he's probably the best at limiting playing time to undeserving players in the National League and probably only bettered by whoever is managing the A's at the time (Beane runs that team...).
Okay, so maybe I misinterpreted
What IS VORP in statistical terms.. or in other words what stats go into VORP.
Same thing with WARP?
I know RC (or Runs Created) is a large part of what WARP is. but could you go into a little bit of depth on what these two stats ARE for dummies like myself who really haven't delved much into the sabermetrics of baseball.
VORP
So why Baseball Prospectus would turn around and do virtually the same thing that Bill James had complained and complained about 25 years prior is baffling. I like BP a lot and enjoy reading their site, but this is one thing that irritates me.
We know how Replacement Player (RP) is calculated (and sorry, but it's rather long--if I have time later tonight, I'll do my best) and we can get a rough idea of a player's value over RP, but we don't know the current formula BP is considering so our attempts at doing so would no be accurate.
Same goes with WARP except that it DOES consider defense.
I think that's one reason for people's skepticism of such stats and it's warranted. I've done as much reading as I can into how these are calculated and fell confident they are accurate stats, but I could not defend the method because I do not know what it is. And that's unfortunate because people who don't want to spend the time like I have reading and reading and reading to find the answers are rightfully going to dismiss such stats. BP is doing sabermetrics a disservice by not making this information public. They do so in the interest that it's THEIR stat and they want to be the only wants to publish that information. I understand that. They did the research and the hard work involved, but only when the public knows how these are calculated...and understands them can we make further necessary advances.
BP is better than The Hardball Times, but THT at least makes their stat formulas available.
VORP, though, is the cumulative effect the player has on run scoring compared to a replacement level player. I think this is what you were asking. I think you'd agree that last season, Ronny Cedeno, was worse than the average replacement player. He could have been replaced by nearly anyone (you, me, Al?) and they could have done better. His VORP on the season was, if I recall, -17.6 (3rd worst in baseball). He provided a negative contribution to the team. The average replacement player could have had a VORPof 0.0.
So it's the addition of all things positive and all things negative and then comparing it to replacement level.
WARP is similar except, as I said before, defense is included.
Does anyone else feel...
Thanks, Maddog, for your usual eloquent and thorough explanation.
Not that this COMPLETELY answers the question...
The question now simply becomes -- where does BP set replacement level? I'm at work, so I haven't the time or software necessary to toy with it, but given the data at hand, I'm pretty confident I could "reverse-engineer" VORP in an hour or so if I set myself to it.
RPOSMLV?
No wonder it was winnowed down to VORP, though that seems like it would be the sound a creature in a swamp would make.
by theprognosticator on Feb 18, 2007 7:09 PM CST up reply actions
So what stats are used stil
However, these stats seem only useful in retrospect, due to the lack of judging how well the replacement will play.
For example, Ronny Cedeno had a negative VORP. Everyone knew this. His primary replacement was Neifi Perez, who also would make a negative contribution.
Also, how does VORP work with Pitchers? Maddog stated that it only measures their offensive contributions, how do pitchers receive grades on their limiting contributions?
Replacement level
Over the last 40 to 50 years, the amount of playing time for "regular" players has remained rather constant. Essentially, 80% of the playing time is your regular players. The remaining 20% is the replacement level.
Basically, let's imagine one team. Let's imagine the Cubs and let's make it simple and imagine that Derrek Lee plays exactly 80% of the time at 1st base next season and Daryle Ward plays there 10% and Henry Blanco plays there's 10%. Replacement level is done at each position...the replacement level of a first baseman is going to be more productive than the the replacement level of a shortstop or catcher, for example.
Ok, Lee plays 80% and the other 2 make up replacement level. The average of Ward and Blanco's stats represent replacement level at 1st base.
If we now expand our imaginary league to what is reality, a simple way to remember replacement level is that the "regulars" (the regular 1st baseman for the Cubs in 2006 was John Mabry due to injury) make up 80% of the playing time.
If we take all the 1st basemen in the game and place them in order of plate appearances, we add them up down the line until we get to 80% of the playing time. The average of the remaining 20% is replacement level at 1st base.
Injuries to Lee last season change things somewhat from year to year. Mabry qualified as a regular 1B last season and Lee's stats were added into the replacement level pile.
For pitching, a number 1 through 5 starter is determined by who makes the most starts. Clemens would have been a number 3 or 4 for the Astros. Again, the top 80% of the pitchers (starters) make up the regulars while the average of the remaining 20% make up replacement level. The same is done for relievers.
So every year, no matter what...no matter how damn good the players may be, there will be players who are below replacement level. Stats that measure replacement level are intended to measure the value of an individual player across eras at the same position. While we can find numerous correlations between VORP and runs or win or whatever else we may be trying to find, replacement level is, in its purest sense, a way to evaluate a single player's contributions above what a typical replacement may be able to do. Your average replacement (fill-in) is not going to be average...he'll be below average, but hopefully above replacement level. The stats weren't created to evaluate team performance and certainly aren't the best mark for doing so, but that doesn't mean we can't do some evaluations of teams using them.
I'm pretty sure...
For example... Jeff Keppinger had 37 plate appearances playing third base for the Kansas City Royals last year, it would seem. He hit .267/.323/.400 -- a .723 OPS. He is, exactly, a replacement level player for third base. (The idea being that you can find some sad sack to hit the ball and somehow avoid being a total defensive travesty at FIRST base, but you'd better find someone who knows how to hold onto the ball at shortstop, meaning you're going to find... guys like Neifi and Tony Womack.)
But basically VORP is the sabermetric version of RBIs -- it measures run production as a counting stat, not a rate stat. It's really good at answering certain kinds of historical questions. It's not bad at producing reasonable, sane Silver Slugger and MVP nominees. It's alright comparing people at the same position, with some caveats. It's far less valuable for looking at guys like Soriano, who are changing position this year, and guys like Mark DeRosa, who have never HAD a real position.
One thing I was thinking...
It's not just about the negative VORP, as you mentioned, but more about what a team or manager does (or can do, really) to offset that negative VORP. This is common sense, of course, but just something I was thinking. Sure, the Yankees had a large negative VORP (injuries to LF, RF and 2B), but when they have Giambi, Posada, Damon, Jeter and ARod, the overall impact would be less. This would be the opposite of say having an injury to Derek Lee, but then only having it offset by Ramirez, Murton, Pierre, Cedeno, Barrett, etc.
I know you said a lot of data was lost and VORP is an individual anlysis, but I'd be curious to see if you had an "average VORP" for teams. Or if that would even make any sense.
Precisely.
I have the average negative VORP for a team since 1996, which is -72 and I believe the average PLAYOFF team in 2006 had a VORP of 420. The average playoff team had a negative VORP of only -50 (and the Yankees threw this off a bit).
I think the Cubs need to shoot for a negative VORP of -45 or lower in 2007. If they do that, they should be in good shape.
One thing to keep in mind about negative VORP is that youth plays a key role in this. It's one of the things I was going to look into and may still do it, but it's different when a player is 22 and providing a negative VORP than it is when a player is 36 and doing so. A player at 22 is worth giving some time to develop and the guy at 36 is only going downhill...get him out of there.
So teams who had a lot of young players (2006 Cubs) will likely be relatively poor in terms of negative VORP.
I should point out that the 2001 Mariners (Lou Piniella...a billion-win team) had a negative VORP of -3.7 (that wasn't the lowest in the 11-year stretch I looked at, but it was top 2 or 3), but that team, pitching, had only one player provide negative VORP....and he was a position player (John Mabry).
I have to agree...
What the heck makes anyone think this guy can start 25 games this year?
(adding the needed disclaimer: of course, it can happen)
BUT, this guys hasn't had a full season since 2003. And that was his only one.
He's pitched:
2002 - 116
2003 - 211
2004 - 118
2005 - 166
2006 - 43
We all know the troubles with increasing a rookie's innings pitched by a certain percentage: I don't remember the percentage, but there's one out there where it's pretty ridiculous to EXPECT someone to pitch a certain percentage of innings more than he did the previous year. Anyone know what I'm talking about?
Anyway, he hasn't pitched a lot of innings for four years, and at that it was only one time. It'd be nice to think that if he's healthy out of Spring Training, he could stay healthy all year, but I would put the chances (just a guess) at 10-20% that he can start more than 20 games and pitch more than 165 innings.
i believe
if a pitcher exceeds their previous season's total, or maybe its career high, by more than 30 innings there's an increased risk of injury
by DartmouthCubsFan on Feb 17, 2007 6:10 PM CST up reply actions
Okay...
Now this is why I read BCB
division crown?
by cubz409 on Feb 18, 2007 12:00 AM CST reply actions
give me a taste
WOW
For what it's worth...
If you go based on runs scored, that depth chart projects the Cubs to score 1,242 runs. If you go based on runs batted in, the Cubs score 1085 runs. (For reference, the Yankees scored 930 runs last year.)
So obviously something's not quite right here. (Also not quite right? As in not right at all? Daryl Ward getting bats at almost EVERYWHERE but at pinch hitter).
I'm pretty sure that this thing is double-counting some things for people who play more than one position, which is throwing a monkey's wrench in it.
Does this affect the veracity of the averaged averages? No idea.
Bench players
Sorry...
That said... I get 838 runs from totaling runs scored, 769 runs from totaling RBIs. Note that this doesn't cover offensive contributions from pitching -- and our pitching staff, if nothing else, can hit, as far as the meager standards of their compatriots go.
Now, let's go ahead and do the flip side of this... project out runs allowed, from the ERAs listed. Take the innings pitched, divide by nine, and multiply by the ERAs, and you come out with 716 earned runs allowed. Unfortunately, you can still lose games from unearned runs, so that, again, is an incomplete figure. So let's compound inaccuracies and introduce another level of abstraction, and derive a Pythagorean win expectation. It's late, and I'm tired, so we'll do the quick and dirty version, no fancy Pythagenport exponents.
Average it out to 803 runs scored... and you get a .557 winning percentage, or a 90-72 team.
Yay, cocktail napkin math.
1,242 runs scored?
Somehow, we'll figure out a way to lose 90 of them.
by theprognosticator on Feb 18, 2007 5:38 PM CST up reply actions
90?
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
Cubs By The Numbers
Cubs By The Numbers is a history of the ballclub by uniform number, but the biographies help trace the history of our beloved team in a new way. For everyone who's a Cubs fan, anyone who ever wore the uniform is like family. Cubs By The Numbers reintroduces readers to some of their long-lost ancestors, even ones they think they already know.
Click here to order your copy, available now!
Recent Stories in Ticket Exchanges
Baseball Nation Recent Stories
In Association With
Big League Stew - MLB - Yahoo! Sports
- Podcast: Dirk Hayhurst talks about Italy, the reality of the big leagues and his new book
- Dennis ‘Oil Can’ Boyd claims he used cocaine before a majority of his games
- Does Billy Beane’s contract extension signal an optimistic dawn for Oakland fans?
- Ask Alex: Will Alex Avila be an All-Star again?
- Create-a-Caption: Matt Cain’s great pair of golf pants
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.
