The Other Fifteen

It's a slow news day, so Fred Mitchell talks about Lee Elia's famous rant. This site is named after that rant, so I feel obligated to mention it.

The Cub Reporter has moved. Apparently they felt like they weren't able to swear enough as members of MVN. This is their third location in the past year or so - maybe one day they'll get blog-wives and settle down or something, maybe hook up with a nice Chicago Bandits blog and have some blog-children.

An object lesson as to why you should avoid forums at all costs. "Your linear weights are too generic for me. I'm just going to subtract CS from the numerator of OBP and add SB to the numerator of SLG!" Help me I am in hell.

The box score for today's game is here. It should update live. Or be a true rebel, and visit the Secret Gameday Link. I make no warranty, guaranteed or implied, about what should happen if you follow that link. Gameday isn't officially operating yet.

And a warning to commenters - anyone caught complaining about the lineup of a spring training game on this blog will be summarily hung without trial.

Labels: Baseball, Chicago Cubs, Linear Weights

People often say, "You can't have an All Star at every position!" Know this (tattoo it on yourself where you'll see it if you think you'll forget): any time this is brought up, you're talking about a crappy ballplayer. I mean, seriously. Nobody starts off conversations about good ballplayers that way. It's some sort of red flag that people want to keep a baseball player on the field for non-baseball reasons.

Apply that sort of thinking to your everyday life. Say your kid brings home a report card with an F on it. How would you react if he said, "You can't have an A+ in every class!" Or say your spouse is responsible for making dinner and sets it (and the kitchen) ablaze. "You can't have filet mignon at every meal!"

I'm not asking for straight As, and I'm not asking for fancy French cooking. I'm asking for the shortstop equivalent of a C, or some chili macaroni Hamburger Helper. I'm looking for average. It's a total straw man argument.

To go ahead and illustrate my point, I've compiled a list of what I've supposed to be the starting shortstops for every team in the majors, and calculated Wins Above Replacement using Sean Smith's projections. [In the case of the Angels and the Nationals, I've used two shortstops.] I've also included perpetual Cubs fan favorite, the Great Destroyer himself, Ronny Cedeno.

I held playing time constant for all players, and have not used any park adjustments. Both of those (false) assumptions would tend to favor Ryan Theriot in comparison to other shortstops.

Cubs players are in bold.

Name	Team	League	wOBA	Defense	WAR
Troy Tulowitzki	COL	NL	0.356	14.00	4.06
Miguel Tejada	HOU	NL	0.373	0.00	3.74
Jose Reyes	NYM	NL	0.349	6.00	2.98
Jimmy Rollins	PHI	NL	0.355	2.00	2.95
Jason Bartlett	TBA	AL	0.32	13.00	2.50
Hanley Ramirez	FLA	NL	0.375	-17.00	2.35
JJ Hardy	MIL	NL	0.348	-1.00	2.31
Adam Everett	MIN	AL	0.283	31.00	2.10
Derek Jeter	NYA	AL	0.358	-15.00	2.07
Khalil Greene	SDN	NL	0.33	7.00	2.05
Jhonny Peralta	CLE	AL	0.345	-8.00	1.99
Jack Wilson	PIT	NL	0.327	7.00	1.88
Michael Young	TEX	AL	0.352	-14.00	1.84
David Eckstein	TOR	AL	0.323	3.00	1.78
Edgar Renteria	DET	AL	0.33	-2.00	1.71
Rafael Furcal	LAN	NL	0.331	1.00	1.57
Macier Izturis	LAA	AL	0.333	-6.00	1.52
Yunel Escobar	ATL	NL	0.34	-6.00	1.43
Orlando Cabrera	CHA	AL	0.323	-1.00	1.43
Alex Gonzalez	CIN	NL	0.323	4.00	1.40
Ronny Cedeno	CHN	NL	0.331	-1.00	1.40
Stephen Drew	ARI	NL	0.339	-6.00	1.38
Julio Lugo	BOS	AL	0.321	-3.00	1.14
Bobby Crosby	OAK	AL	0.306	6.00	1.13
Omar Vizquel	SFN	NL	0.302	10.00	0.80
Ryan Theriot	CHN	NL	0.312	1.00	0.55
Yuniesky Betancourt	SEA	AL	0.312	-5.00	0.48
Tony Pena	KCA	AL	0.279	10.00	0.03
Cesar Izturis	SLN	NL	0.295	2.00	-0.28
Christian Guzman	WAS	NL	0.31	-9.00	-0.45
Felipe Lopez	WAS	NL	0.316	-13.00	-0.48
Erick Aybar	LAA	AL	0.288	-10.00	-1.25
Luis Hernandez	BAL	AL	0.268	1.00	-1.36

So, when I say that almost anybody would be an improvement on Ryan Theriot, I'm not exaggerating or showing some sort of bias against scrappy white guys. He's not the worst starting shortstop in the majors, but he's not too far away.

Now, obviously this is based upon projections of performance, and those projections could be wrong. The projections on offense are probably more reliable than the projections on defense. But that's as true for Troy Tulowitzki as it is for Luis Hernandez. Unless you have a specific reason that the projections are underrating Ryan Theriot relative to the other shortstops in baseball, I don't see a reason to think he'll be very good for the Cubs next year.

He's an average defender at shortstop - he's got sure hands, even if his range isn't very good; not a butcher like Michael Young or Hanley Ramirez, but not a solid defender like Troy Tulowitzki or Omar Vizquel. At the same time he's not a very good hitter - he's not as bad as the Felipe Lopez/Christian Guzman contingent, but he's certainly not even in the vicinity of the Orlando Cabrerra/Alex Gonzlez "respectable but not spectacular" benchmark. He does nothing particularly well.

The worst part of it is, you do not need a lot of advanced metrics to figure this out. The fact that his defense is good but not spectacular should be readily obvious to the armchair scouts out there; Cubs fans voting in the Fan's Scouting Report pretty much came to that conclusion. Cubs fans are equally able to figure out his deficiencies on offense; the Community Projections over at Bleed Cubbie Blue were very much in line with what other projection systems were saying.

In fact, let's rerun his WAR, this time using nothing more than the collected wisdom of Cubs fans. Converting the Fans' Scouting Report to runs using Tango's method puts Theriot at plus 7 runs; for a projection we really should factor in aging and regression to the mean, but screw it. Fans project Theriot to have a .316 wOBA, a whole four points above CHONE. That works out to a WAR of 1.29 - certainly more optimistic than what I have here, but far from outstanding.

As a group, Cubs fans seem very able to figure out Ryan Theriot's absolute value as a player. So why do they seem so unable to grasp his relative value? Maybe it's just a case of the vocal minority skewing my perception of Cubs fans, I dunno.

As a bonus, some quick, largely non-Cubs related, thoughts:

1. Troy Tulowitzki is an absolute monster. The second-best defensive shortstop in the game, and a solidly above-average hitter? There's a Coors Field effect in there, but damn.
2. Plus 31 runs for Adam Everett? Holy crap. If you can think of a more underrated player in all of baseball, I'd love to hear your thoughts.
3. Tejada shows up rather better than I thought he would. Without knowing the precise aging curve that Sean Smith used for his defensive projections, I'd have to be tempted to take the under on that projection, though.
4. Erick Aybar is just bad. Just... bad. I really wonder what the Angels are planning to do about that.
5. The Red Sox can't be happy about the Julio Lugo signing right now.
6. What the hell, Cardinals? Cesar the Wonder Out is making $2.85 million on a one-year deal; Eckstein is making $4.5 million on a one-year deal. There's almost no way that the difference in performance between the two is worth $1.5 million - and it's possible that Eckstein would have given the Cardinals a hometown discount. Especially given Eckstein's marketing potential for the Cards (he's absolutely beloved by them), this is just stupid on their part.

Labels: Baseball, Chicago Cubs, Defense, Linear Weights, Projections, Ronny Cedeno, Ryan Theriot, WAR

Reader DeRoMyHero asked:

Is there any way to judge a "good walk rate" (a.k.a. "plate discipline)? After all, not all walks are equally good for hitters or bad for pitchers.

1. For Rickey Henderson or Tim Raines, a walk is almost always good. Even though both had some power (especially Henderson), they created more problems by just getting on base.

2. For Barry Bonds, a walk (particularly with RISP) is usually bad -- especially with Pedro Feliz behind him. Maybe he should have expanded his SZ a bit, trading a few more outs for more RBI potential. Was he selfish to guard his stats at the expense of the team, or would his performance have slipped too much?

3. For some aggressive hitters (Aramis Ramirez, Michael Young), trying to get them to draw more walks would diminish their overall effectiveness. They do so much damage with fastballs early in the count that a walk would be a consolation prize.

4. A cleanup hitter like Adam Dunn accomplishes less with a walk than a leadoff hitter. The guy behind him likely is not as good of a hitter, and it takes at least 4 singles to score Dunn from 1B. Would he do more damage by being more aggressive?

5. A #8 hitter has to adjust based on the situation, i.e., he must be capable of being patient or aggressive. I ask this because I have seen reviews of Tyler Colvin that say he "needs more plate discipline". If he can produce like Michael Young when he swings, who cares?

Conversely, the Rangers have a young DH named Jason Botts, who has made a living in AAA of taking pitchers 3-2, then hitting the cripple fastball. (He is Adam Dunn with more speed -- BB, K, HR.) The jury seems to be out on whether he can hit anything other than a fastball down the middle -- a requirement in the ML, especially as a DH.

Has anyone come up with a way to analyze good walk rates based on type of hitter or spot in the batting order?

Nothing quite like a simple question, right?

First, a quick aside. When it comes down to winning games, we really don't care where a player's value comes from, just so long as he provides that value. You can have an all-glove, no-hit shortstop and a stone-handed Silver Slugger at the position - so long as their overall value is equivalent there's no difference in value between them.

Where the how becomes valuable is in player development and forecasting. You worry about how a player creates his value when you want to try and figure out what value he'll provide in the future, and how you can maximize his future value.

Okay, now back to bidness. First, we need to see if the premise is true - are some walks more valuable than others? To figure that we need a tool to value walks in relation to other events on offense.

I've made no secret of my deep and abiding love for linear weights. Simply put, linear weights are simply the average expected run value of an event. But we know that the run expectancy value of a triple is very different depending on whether or not the bases are loaded or empty - the linear weight tables used in, say, wOBA or Batting Runs don't look at the base/out state information when determining value. But nothing stops us from doing so.

[If you want to get more deeply into these things, Win Probability Added is the stat for you. Suffice it to say that there is a sabermetric holy war over WPA that is well beyond the scope of this article.]

Let's ignore out state for a moment and just look at run values based on men on base, which is what we're talking about here - the question is distinctly asking about the difference between Runs and RBIs, essentially.

Baserunners	1B	2B	3B	HR	BB	K	Out
---	.29	.49	.68	1.00	.29	-.20	-.20
x--	.49	.97	1.36	1.74	.43	-.32	-.36
-x-	.72	1.00	1.16	1.60	.23	-.39	-.34
--x	.72	.86	1.00	1.51	.21	-.48	-.29
xx-	.93	1.54	1.94	2.38	.56	-.52	-.48
x-x	.88	.93	1.77	2.22	.38	-.61	-.46
-xx	1.17	1.46	1.62	2.07	.23	-.70	-.56
xxx	1.38	2.00	2.40	2.86	1.00	-.82	-.68

And, yes, the relative value of a walk compared to the other positive offensive events decreases in "RBI situations;" this is why a single is generally more valuable than a walk - a single can advance the runner, while the walk can't.

But a walk does still have positive run value, especially compared to the negative run value of the out. And here is the cold, hard truth:

72% of all balls put into play are turned into outs.

There are, of course, things you can do as a hitter to help a particular batted ball beat those odds - you can hit a line drive, which is much harder to field than a ground ball or a fly ball, or you could avoid putting the ball in play together, and just knock it out of the yard. Both of those require making solid contact with the ball. And, I'm sure it won't surprise you to hear that the further a ball is from the center of the strike zone, the harder it is to turn it into a hit. You do not do yourself any favors by swinging at pitchers pitches - just because it would be more valuable for Barry Bonds to hit a home run than take a walk, doesn't mean that he's going to have much success trying to hit curveballs aimed at his shoelaces.

This is why plate discipline is so important, and it's not just because of walks and OBP. Everyone, repeat after me:

OBP is not plate discipline.

OBP is not plate discipline.

OBP is not plate discipline.

And, just to add on:

Walks are not plate discipline.

Walks are not plate discipline.

Walks are not plate discipline.

Both of them correlate well with plate discipline and are crude proxies of plate discipline, but they are not plate discipline.

The two best measures of plate discipline are IsoD and Pitches per Plate Appearance. Pitches per plate appearance is a wonderful, oft-overlooked stat - shocking, since it's so dirt simple that even ESPN provides it. IsoD is just a fancy name for walks over plate appearances; essentially OBP without the hitting. Anyways, here's all Cubs with 100 at-bats or more, sorted by IsoD:

NAME	AB	BB/PA	P/PA
Daryle Ward	110	0.165	3.64
Cliff Floyd	282	0.109	3.58
Derrek Lee	567	0.109	4.02
Mark DeRosa	502	0.101	3.96
Matt Murton	235	0.1	3.75
Jason Kendall	174	0.094	4.23
Mike Fontenot	234	0.085	3.93
Ryan Theriot	537	0.082	3.53
Aramis Ramirez	506	0.077	3.67
Michael Barrett	211	0.074	3.54
Felix Pie	177	0.072	3.73
Jacque Jones	453	0.069	3.64
Cesar Izturis	191	0.063	3.42
Angel Pagan	148	0.062	3.87
Alfonso Soriano	579	0.05	3.67

And here's the same table, sorted by pitches per plate appearance:

NAME	AB	BB/PA	P/PA
Jason Kendall	174	0.094	4.23
Derrek Lee	567	0.109	4.02
Mark DeRosa	502	0.101	3.96
Mike Fontenot	234	0.085	3.93
Angel Pagan	148	0.062	3.87
Matt Murton	235	0.1	3.75
Felix Pie	177	0.072	3.73
Aramis Ramirez	506	0.077	3.67
Alfonso Soriano	579	0.05	3.67
Daryle Ward	110	0.165	3.64
Jacque Jones	453	0.069	3.64
Cliff Floyd	282	0.109	3.58
Michael Barrett	211	0.074	3.54
Ryan Theriot	537	0.082	3.53
Cesar Izturis	191	0.063	3.42

First off, just because it's something I'm dying to point out: Ryan Theriot? Not a very patient hitter. (In fact, he was the WORST starter on the team when it came to pitches per plate appearance. True story!)

But now, let me get over that and try to provide some analysis. Remember? Walks aren't plate discipline - or rather, they aren't all of plate discipline.

Plate discipline, simply defined, is only swinging at your pitches, not the pitcher's pitches. Remember: balls are harder to hit than strikes. So you want to do a good job of making pitchers throw you strikes, and recognizing which pitches are good to hit and which aren't.

I won't belabor the whole idea of "pitchers counts" and "hitters counts;" anything that you're likely to hear about from TV announcers hardly needs my help in becoming common knowledge. Suffice it to say - you get better results as a hitter if you learn to take bad pitches, work yourself into favorable counts, and make the pitcher throw you strikes. [If you're interested in looking at the specific dynamics of how counts effect a hitter, I'm happy to oblige, though.]

So when we say we want Colvin to develop more plate discipline, we do want him to walk more. But we want him to walk more as a byproduct of swinging at pitches he can hit and taking pitches he can't. Better plate discipline makes you a better hitter - it doesn't just mean walks, it means more hits and more hits for extra bases.

Labels: Baseball, Linear Weights

I decided to do something a little different this time and break down the chart by positions. It helped me sort out some of the playing time issues, at any rate, and I think it's rather nifty to look at. Guys like DeRosa, Cedeno and Murton who are expected to play multiple positions get the most milage out of it.

Name	PA	wOBA	Defense	WAR
PITCHER	370			0
CATCHER
Geovanny Soto	512	0.355	0.50	3.28
Henry Blanco	188	0.301	0.50	0.36
JD Closser	48	0.323	0.00	0.15
FIRST BASE
Derrek Lee	632	0.396	0.50	4.39
Daryle Ward	50	0.354	-1.43	0.04
Mark DeRosa	22	0.342	0.00	0.04
SECOND BASE
Mark DeRosa	422	0.342	-0.57	1.00
Mike Fontenot	225	0.317	-0.29	0.16
SHORTSTOP
Ryan Theriot	542	0.312	-0.10	0.69
Ronny Cedeno	130	0.331	-0.10	0.37
THIRD BASE
Aramis Ramirez	572	0.385	-0.19	3.71
Mark DeRosa	112	0.342	0.00	0.36
Ronny Cedeno	20	0.331	-0.29	0.04
LEFT FIELD
Alfonso Soriano	621	0.371	0.57	3.53
Matt Murton	110	0.357	0.29	0.45
Daryle Ward	60	.354	-1	0.12
CENTER FIELD
Felix Pie	556	0.328	1.43	2.66
Ronny Cedeno	68	0.331	-0.67	0.14
Sam Fuld	65	0.307	-0.10	0.06
RIGHT FIELD
Kosuke Fukudome	560	0.381	0.95	3.96
Matt Murton	110	0.357	0.12	0.43
Sam Fuld	38	0.307	-0.10	0.03
DH/PH
Matt Murton	60	0.328	0.00	0.08
Daryle Ward	70	0.344	-1.43	0.04
	6103			26.08

I plan on cleaning up the presentation of this and doing some other things with it, but I'd like to get some feedback on it for the time being. wOBA is still pretty much entirely driven by CHONE, and I plan on doing some minor revising in that department. Pitcher hitting is entirely accounted for, and there's about 200-300 replacement-level at-bats included in there, so I don't think I'm being unreasonable about playing time.

Labels: Baseball, Chicago Cubs, Linear Weights, Projections, WAR

A few additional notes on the WAR depth chart I posted the other day.

If you'd like to go ahead and calculate WAR yourself, then go to Sean Smith's stat site; my WAR calculation spreadsheet is available for download there. I'd like to thank Sean for making that available.

Unlike with the version on EditGrid, it's not autopopulated with Cubs data - it is, however, set up to make it scaldingly easy to figure out the WAR for any player, NL or AL. Download the CHONE projections and look up the defense projections available there for all the data you need.

Maddog commented on the WAR chart, and I thought I should go ahead and clarify a few things.

First, like I said there, pitcher hitting is already accounted for in the replacement level - all NL teams have pitchers hitting, so unless you think the Cubs pitchers are going to hit significantly worse than other teams its a nonissue.

The other issue is the total amount of plate appearances. I projected 6145 plate appearances for the Cubs; last season the Cubs; last year the Cubs had 6268 plate appearances. So by that standard it would appear I'm not leaving enough room for the 400 or so PAs by pitchers, or for replacement level hitting.

And... I didn't. But it's not as bad as you might think, because plate appearances are a function of On-Base Percentage - you can "buy" more plate appearances for your team by making fewer outs per plate appearance. I'm expecting rougly 2,400 plate appearances out of the Cubs for next season, based on their increased OBP.

So, I still need to take 200-300 PAs off the depth chart, and I'll work on that for tomorrow.

And maddog is right - someone is going to provide stats that look below replacement level, at some point, for the Cubs next season. But that's completely irrelevant, because remember, replacement level is an average of a group of players - some are going to fall above/below the actual replacement line, both as a function of sample size and of distribution.

Put another way - the Cubs gave playing time to several players last season that could be described as "replacement level," given the definition I'm using:

• Ryan Theriot
• Mike Fontenot
• Angel Pagan
• Koyie Hill
• Rob Bowen
• Ronny Cedeno
• Sam Fuld

You can add or subtract from that list as you please; doesn't really bother me any. Some of them outperformed the replacement baseline, and some of them underperformed the replacement baseline. When you average that out, you get replacement level production. You don't need to specifically account for below replacement player production.

But you do need to give below-replacement players a share of the plate appearances, which again, I didn't do well enough. So I will revise the chart tomorrow.

Also, I'm planning on doing positional breakdowns starting tomorrow, so we can start to look at some chaining issues, see how injury might affect the team, and take a look at positions where the team might have some unanswered questions.

Labels: Baseball, Linear Weights, Projections, WAR

Harry Pavlidis over at Cubs f/x has a good overview of how capable Cintron is(n't).

Let's take a look at this based on our Cubs WAR chart:

	PA	wOBA	Defense	WAR
Ryan Theriot	542	.312	-.10	.69
Ronny Cedeno	275	.331	-.10	.79
Alex Cintron	275	.234	-.19	-1.30

What does this mean?

1. Hendry somehow managed to find a way to sign a shortstop worse than what we have already.
2. Interestingly enough, Theriot is somehow less valuable than Ronny Cedeno despite getting almost twice as many PAs on my depth chart.
3. Wow, Alex Cintron is bad.
4. Wow.

So... suffice it to say that Alex Cintron is not an infield predator. This is just a minor league, though, and Hendry says the right things:

"Obviously he had a few snags last year coming off (elbow) surgery," general manager Jim Hendry said. "But our people have been watching him in Puerto Rico. I think he'd done some working out with Carmelo Martinez earlier in the year, so we’ve had our eye on him. We felt like it made sense, especially with (Ronny) Cedeno getting some outfield time in the spring. So we felt like we needed someone else in camp that could play shortstop other than (Ryan) Theriot and Ronny."

It's okay to sign a replacement player at replacement salary if that's all you use him as. If Cintron ends up in AAA Iowa I'm perfectly happy with this. If not... well, kiss two wins goodbye.

Oh, and for those curious:

	PA	wOBA	Defense	WAR
Mark DeRosa	556	.342	-.57	1.32
Brian Roberts	630	.349	.19	2.55

Replacing Cedeno with Cintron already pretty much eats away the benefits of a Roberts trade. (I don't know that taking a few hundred at-bats away from Mike Fontenot justifies the deal either, but I still have a little more work to do with the chaining here.)

Labels: Baseball, Brian Roberts, Chicago Cubs, Linear Weights, Mark DeRosa, Projections, Ronny Cedeno, Ryan Theriot, WAR

WAR - what is it good for?

Well, in this case WAR references Wins Above Replacement. Fans, readers and detractors will note the similarity with Wins Above Replacement Player, or WARP. The concept is the same, the internals are different. The full techincal explanation is available at the link; I'll skim over the relevant highlights below.

The short version: you calculate WAR by summing up a player's contributions on the field - pitching for pitchers, offense and defense for position players - converting that to wins, and then comparing those contributions to hypothetical "replacement players," generally defined as the sort of freely-available talent one can acquire for close to the league minimum salary.

[Note: I owe a debt of gratitude to Tangotiger, whose WAR methodology I'm using and who gave some excellent - and necessary - early feedback, and David Cameron, whose post at USS Mariner inspired this.]

First, let's take a look at the hitters:

Plate appearances shouldn't require much explanation. In this case, I apportioned out most of the playing time for the Cubs on offense; playing time not accounted for here is assumed to be replacement level.

wOBA is Weighted On-Base Average, essentially a rate stat version of Linear Weights. Functionally it works just like Equivalent Average, except on the OBP scale instead of the batting average scale; .340 is (generally) league average. These wOBA figures are based on the CHONE projections, available from Sean Smith's website. wOBA, being a linear weights equation, can be easily converted to runs, and thus wins, above average.

Also available there are defensive projections, which form the basis for my defensive figures. (I made manual adjustments for Lee, Blanco, Fukudome, Pie and Soto.) Smith (and any plus/minus system) rates defenders in terms of plays or runs (in this case, runs) saved compared to average at the position. To convert runs to wins, simply divide by 10.5.

Okay, now here's the secret sauce that lets us compare these players across positions. In the National League, the average player is two wins above replacement, so everyone above gets credit for those two wins.

Players also get credited (or debited) wins for what position they field:

+1.0 C
+0.5 SS/CF
+0.0 2B/3B
-0.5 LF/RF/PH
-1.0 1B
-1.5 DH

This is because a league-average hitter who plays catcher is more valuable than a league average hitter who can only play first base or designated hitter.

Combine those elements, and you get a player's WAR. $WAR is simply the cost value of those wins on the free agent market (the formula assumes that a team pays $4.4 million per win on the free agent market, which has held up pretty well for the 2008 offseason). Actual$ is, well, the player's 2008 contract. Difference is how much surplus value that player is providing; if you've ever sat there and wondered how much Henry Blanco is overpaid, well, now you know.

One thing that pops out at me is how the severe backloading of the contracts means that the salary for players like Ramirez and Soriano are actually very reasonable right now. There's a whole lot of risk wrapped up into the back ends of those contracts, though.

Now, for the pitchers:

Let's start off discussing the innings pitched. It's easier to pitch as a reliever than it is as a starter; usually a pitcher who does both will have a lower ERA as a reliever. So we need to use different replacement baselines for starting and relieving; I've designated two pitchers as "swingmen," Jason Marquis and Sean Marshall. I'm sure some of you are chuckling at that - I happen to be pessimistic about what Marquis can achieve this season and think Lou will pull him from the rotation when things get too bad.

Innings pitched adds up to 1457; there's roughly 1458 innings in a season, which changes because of extra-inning games and home halves of the ninth inning skipped when the home team leads.

ERA is based off several projections, with CHONE getting more weight than the others, mostly because it was handier; beyond that some of it was simply my personal preference. These figures have been revised upwards to make them "defense neutral," since we already gave the position players credit for defense above. (Technical explanation: I calculated FIP ERA for the CHONE projections and used that as a baseline.)

WAR is a function of runs allowed compared to replacement, converted to wins using the Pythagenpat method, a refinement of the Pythagorean theorum.

Just take a gander at the difference between what the Cubs are paying Hill and what he would be worth on the free agent market. (And that's ignoring the fact that teams generally will overpay for pitching wins.) That's why, in case you were wondering, the people advocating for a Hill-for-Bedard trade were being, at best, shortsighted. Bedard may well be a better pitcher, but because of his many remaining cheap years, Hill is insanely more valuable.

And just look at how overpaid our bullpen is; Howry is our only free agent reliever expected to provide value up to his contract.

Now, what I have failed to do is assign a "premium" to the closer for leverage; this is because nobody knows who the hell our closer is. I did play around with it for a bit, and: hey, we lose a whole win by making Dempster the closer as opposed to a middle reliever. That's a pretty drastic difference for a relief pitcher.

Add up our position player WAR and our pitching WAR, and we get 42 wins above replacement; a team composed of entirely replacement level players would be expected to win about 50 ballgames. So, given the assumptions I used, you could say the Cubs look like about a 92 win team. That's based on some optimistic assumptions involving the starting rotation (namely that Marquis gets dumped from it before too long, and Dempster is never used in it). I plan on redoing this little exercise again several times during spring training, as the final roster spots shore up and we get a better idea of how the rotation and bullpen shake up.

The full spreadsheet used, in all its glory, is available for your perusal on EditGrid. You can also download it to your computer as an Excel file and play around with it. Don't like my assumptions? Change them! Have fun with it. Let me know what results you get.

Labels: Baseball, Chicago Cubs, Linear Weights, Projections, WAR

It's probably the most interesting question to the statistically-inclined baseball fan - how well will certain players perform next season? (Followed closely by the almost-as-vexing question, how well did they perform last year?)

To that end, a lot of various modeling systems have been developed to project player performance. Some of the good, widely available ones are:

• Dan Szymborski's ZiPS
• Sean Smith's CHONE
• Nate Silver's PECOTA
• SG's CAIRO
• Marcels (Tango refuses to take credit for them)

PECOTA is the best available; ZiPS is the best system you don't have to pay for. CAIRO has the best toys. Marcels is probably good enough. [If you want to really get down to brass tacks on just how good they are, Tango, Silver and Smith have a great discussion on those issues.]

During this post, I'll be referring to none of them in particular. All of them work on the same basic principles, and its those principles that I'll be discussing. (There are other projection systems that don't work on these principles - or at least, don't talk about what principles they're based on. Let's put it this way - if they're marketed to a fantasy baseball audience, there are some... problems. We won't discuss those systems.)

There's not going to be enough detail here for you to go out and implement your own projection system, and I'm not going to spend a lot of time on the differences between systems. This is just supposed to be a cursory, general overview.

The most simple method possible

Assuming you have no time to look up projections or generate your own, what's the best way to project a ballplayer's performance? Generally speaking, last year's numbers are the starting point you want to look at.

Does that work? Sure, I guess. Performance for baseball players is pretty consistent year to year during their peak years, and if you skip batting average and go to stats like OBP and SLG the year-to-year correlation is pretty good. (.36 for OBP and OPS, .38 for SLG, if you were wondering).

So, basically, what I'm telling you is that you can expect that good baseball players will outperform bad baseball players, most of the time. In other words, nothing you already didn't know.

You want more accuracy than that? Well, there's a few things we can do.

Increase the sample size

You often hear baseball stats geeks yammering on about "sample size." What that means is - the more information you have, the more likely it is to be accurate.

This is one of those things you already know and understand, whether or not you realize it yet. The odds of a coin toss being heads are right about 50-50 (there's actually about a 1% bonus for whatever side is upright prior to the coin toss in actual practice), but in a series of 10 coin tosses, almost anything CAN happen. In order to make sure you have an even distribution of heads and tails, you need to run thousands of coin tosses.

Baseball is a lot more complicated than a simple coin toss, so even in a full season (where a player could accumulate anywhere from 600-700 at-bats) there is a lot of random "noise" to account for. The movie Bull Durham explains it pretty well:

"Twenty-five hits a year in 500 at-bats is fifty points. Okay? There's six months in a season, that's about 25 weeks. You get one extra flare a week—just one, a gork, a ground ball with eyes, a dying quail—just one more dying quail a week and you're in Yankee Stadium!"

So preferably we'd use more than one season of data in our projections.

We have to be careful, though: a baseball player is not an unchanging thing, and if we use too much data - or use it improperly - we can draw the wrong conclusions. For example, here's Barry Bonds, career 1986-2006 (translated into a seasonal basis), compared to how he did in 2007:

G	AB	R	H	2B	3B	HR	RBI	SB	CS	BB	SO	BA	OBP	SLG
157	523	118	156	32	4	40	106	28	8	133	82	.299	.471	.608
126	340	75	94	14	0	28	66	5	0	132	54	.276	.480	.565

Some of it looks pretty close, and so we might not notice some of the incongruities right away. The first thing you have to notice is speed; in his career he was pretty speedy, hence the triples and the stolen bases. Old Barry does not hit triples, score runs or steal bases as well as he used to. Another thing Old Barry doesn't do: play as often. Or play baseball with other guys that can hit - that drives down his RBIs.

Barry Bonds is the extreme case, and I know that - I chose him because I figured there would be some things that would pop out of his stat line when looked at. But this applies to every baseball player - players do not stay the same. Maybe they give up switch hitting, maybe they get a new hitting coach and discover religion, maybe they suffer an injury that limits them in ways long after they've come off the DL. The most recent data is always the most relevant, and so that data is simply more important than the other data.

So you use a weighted average, which means that you multiply the stats you're using by a constant depending on which season they're from. If you're using three years of stats, you could use a 5/4/3 weighted average, where you multiply all of a player's 2007 stats by 5, his 2006 stats by 4, and his 2005 stats by 3. Then you divide by 12.

But using Barry brings up another interesting point, so we'll go there next.

Account for age

We know that young players are likely to get better, and old players are likely to get worse. But what does that mean, exactly? Let's look at a graph.

That, friends, is the average aging curve of a baseball player. I use wOBA because of my well-documented love affair with all things linear weights... and because Tango already did the work for me.

There's the upswing of the chart, a flattish-looking bit from roughly age 24 to 28, and then begins the downward slope. Projection systems account for this by adjusting projections based upon the age of the player.

Regress to the mean

The curve above is a representation of a very large sample size, and as such looks very smooth. A graph of a single ballplayer's career looks a lot less smooth - players have career years, players have off years, players miss time due to injury. When Luis Gonzales puts up a 57-homer season almost entirely out of the blue, we need to adjust for that.

That's accomplished by regression toward the mean. Now, you'll often hear people when the speak of baseball players talk about players regressing as a negative; it's important to remember that players who perform below the mean regress upwards; it's a double-edged sword.

When regressing to the mean, you have to answer two questions:

1. What mean do you regress to?
2. How much do you regress?

For a simplistic projection system, you'd use the league average as your mean. How much you regress is dependant on playing time - the more playing time a player has, the less you regress.

What the simple projection ignores

The simple projection, incidentally, is Marcels. If you want to look into the guts and play around, there it is - every equation you need is spelled out for you

Everything we've talked about so far assumes that all ballplayers develop pretty much the same way. We know that's not exactly true, and so we can get better accuracy by taking into account more specific information about ballplayers

We also haven't looked at how players with a large amount of minor-league playing time - and little or no major league experience - are handled.

And we haven't touched upon how you can use batted-ball information to improve your projections.

Oh, and we've left out pitchers. That may be important. We'll get to all of that next time.

Labels: Baseball, Linear Weights, Projections

Two people (which is a not-insubstantial proportion of my readership at this point) are curious about Ryan Theriot's defense, and how he stacks up compared to the league average. One specifically wants to know how he stacks up against the Cubs 2007 MVP-in-Spirit, Mark DeRosa. (Derrek Lee being the MVP-in-Fact.)

Well, I'm happy to oblige. Welcome to Thunderdome. Two scrappy, white middle infielders enter; one scrappy, white middle infielder leaves.

Mark DeRosa was +2.5 Batting Runs compared to the league average in 2007; Ryan Theriot was -22.8 Batting Runs. (It turns out that SS, 2B and CF hit for roughly the league average. This doesn't matter so much just comparing the two, but it's useful to note.)

But what about next season? I would expect DeRosa's offense to fall of a bit, and Theriot's to improve by a bit. CHONE sees DeRosa as being right around league average, and Theriot to be -12 compared to average. ZiPS sees DeRosa about 2-3 runs below average, Theriot roughly -15. So 12 runs different on offense, or about a win.

The short answer on defense: Sean Smith makes his defensive projections available; they're not the best projections on defense but they're the best readily available. (MGL's UZR defensive projections are probably the best I'm aware of, but UZR is published sporadically at best.) The projection sees Theriot defensively at -1 at shortstop, and DeRosa at -6.

So between the two, I'd say that DeRosa is the better choice at shortstop between the two. Both of them would be below-average shortstops, but I think we had all figured that out already.

This is just a quick, cursory look at the question; I plan on doing some posts in a while that go more in-depth into offensive and defensive metrics in general, and may touch on the specific merits (or lack thereof) of Ryan Theriot more when I do.

For those curious - CHONE has Cedeno at -6 on offense, and ZiPS has him about even with Theriot. The defensive projections see Cedeno as even with Theriot.

Edit: If you believe ZiPS, then your best infield (with current personnel) is probably: Lee, Fontenot, DeRosa, Ramirez. (Fontenot is -11 at shortstop, if you're curious.) If you believe CHONE, then it's Lee, DeRosa, Cedeno, Ramirez.

Labels: Baseball, Chicago Cubs, Defense, Linear Weights, Mark DeRosa, Ryan Theriot, Steel-Cage Match

Ricky O'Donnell from the Sun-Times is breaking down the Cubs' young players. So, of course, we get this:

SS Ryan Theriot

Last season: .266/.326/.346, 3 HR

Outlook: What Theriot lacks in size and power he makes up
for in defense and speed. His 28 steals last season went a long way to helping the Cubs produce runs, and he’ll once again be asked to provide the spark he did last season.

What the Cubs are hoping for: To find a regular spot near
the top of the lineup and continue making clutch hits while providing stellar defense. A little higher on-base and slugging percentages would be nice as well.

Let's break that down.

.266/.326/.346 - Bad. Even for a shortstop, pretty mediocre. His on-base percentage only looks good in the context of his other skills.

"What Theriot lacks in size and power he makes up for in defense and speed." - Essentially wrong. What he lacks in power he didn't make up for, which is how you produce -22.8 Batting Runs.

"His 28 steals last season went a long way to helping the Cubs produce runs." - More wrong. Using .06 as the linear weights value of a stolen base, you get 8.4 runs out of 28 stolen bases. A long way! It looks even smaller if you account for the 2.4 runs his caught stealings cost the team.

"A little higher on-base and slugging percentages would be nice as well." - It would help if he learned how to play baseball.

Oh, and he's going to be 28 this season. He's the same age as Cesar Izturis was, and older than such young guns as Rob Bowen and Matt Murton.

Labels: Baseball, Chicago Cubs, Linear Weights, Ryan Theriot