Stats


It seems that this blog has turned into a blog devoted solely to analyzing Aramis Ramirez’s newfound plate discipline.  Whatever; let’s look at the numbers:

P/PA career: 3.70

P/PA 2008 to date: 4.12

SwgA% (percentage of pitches swung at) career: 50%

SwgA% 2008 to date: 45%

Strk% (percentage of pitches that are strikes) career: 64%

Strk% 2008 to date: 59%

What do these numbers mean?  Well, not only has Aramis started taking more pitches than in the past, but pretty much all the “additional” pitches he’s taken have been balls!  You would expect that an increase in plate discipline, all else being equal, would lead to an increase in the number of called strikes—as well as the number of balls—but this has not been the case.

So what’s going on?  A few possibilities:

1) Nothing.  For some odd reason, ARam has faced a cross-section of pitchers who have thrown more easily identifiable balls than he has seen in the past.  This is pretty unlikely.

2) Overall increase in discipline.  He’s seeing the ball better and not letting himself swing at pitches out of the zone.

3) Better recognition of a specific pitch.  There’s some pitch—a splitter, curveball, something—that has enticed him to swing in the past that he is now able to lay off.  This pitch almost always ends up a ball.

I have a hypothesis along the lines of option 3, but I need more time to test it. 

Three somewhat unrelated things today:

1) I like Rich Hill, and not just because he’s a (relatively) young Cubs pitcher with a lot of upside. I like that he introduced the Cubs before a game on national TV last year as Kip from Napoleon Dynamite; I like that Ozzie Guillen, a piece of shit if ever there was one, called him a piece of shit. I like that he managed to survive the wrath of Cubs fans who wanted him traded ASAP in 2006 (honestly, though, he probably didn’t notice) and came back to put up a very good season last year. And I hate the way Lou and the Cubs’ staff have handled Hill this year, showing no confidence in him and seemingly setting him up to fail. All that being said, he’s not right, and the move to Iowa is probably the right one, provided it is temporary and the Cubs bring him up and reinstate him in the rotation as soon as he finds his command. His mechanics look off this year (at least they did in the couple starts I saw him make), and obviously he can’t find the plate. Better for him to start every fifth day in Iowa and find his command there without taxing the bullpen and hurting the big club. But please, Rich, figure it out quickly.

2) Whenever the Cubs load the bases with nobody out, as they did last night in the 7th inning, my significant other and I look at each other and say, “uh oh.” Why? Well, because it seems to us that the Cubs tend to score either 0 or 1 runs in these situations, and almost never any more. I decided to look into this figure out if our perceptions were right or if we were selectively remembering the disappointing rallies. Turns out we were pretty much right. This year, the Cubs have loaded the bases with nobody out 9 times, and the frequency of runs scored looks like this:

0 runs: 1

1 run: 5

2 runs: 1

5 runs: 2

That’s an average of just under 2 runs per inning. This seems a little low, but what’s “normal”? I don’t have any league-wide data for the past couple years, but Tom Tango has data from 1999-2002. For those years, teams failed to score more than one run after loading the bases with no outs just 38.3% of the time; the Cubs have failed to score more than one run after loading the bases with no outs two thirds of the time this year. It should of course be noted that this is an extremely small sample size: even small samples think this sample is tiny. We can’t conclude that the Cubs suck in these situations based on 9 events. Luckily, there is a slightly larger (but still small) sample that we can look at: the 2007 Cubs.

The 2007 Cubs were even worse than the 2008 Cubs at scoring with the bases loaded and nobody out, failing to score more than one run on 12 out of 19 occasions and coming up empty 6 times. I’ve included a little graphic below comparing the 2007/2008 Cubs’ numbers with the 1999-2002 league-wide numbers from Tom Tango’s site. Remember that we are dealing with a small sample size with the Cubs numbers and that the league numbers are from a more offense-heavy period; still, the Cubs suck.

The average number of runs scored after loading the bases with no outs for the league from 1999-2002 was 2.417; the Cubs have averaged 1.464 runs in these situations over the last two years, almost a full run lower! It’s striking how bad the Cubs are at scoring runs after loading the bases with no outs, and a look back at the games in which they’ve struggled to do so reveals a number of close losses.

This is probably just a small-sample-size fluke, but here are a couple of other explanations, some plausible and some far-fetched:

a) the Cubs feel the weight of their sad history the most in high-pressure situations, causing them to choke

b) the Cubs have been unlucky on batted balls in play in these situations (similar to small-sample-size fluke, I suppose)

c) the Cubs have been unlucky to have their weaker hitters come up in these situations.

I’m done doing research on this topic, so I have no interest in trying to find evidence to support/refute b or c above.

3) Fun fact: the Cubs have lost their last three series two games to one, but they have outscored their opponents in all three series. Pythagoras says: the Cubs will win the division.

That was the first “kick in the balls” loss I’ve seen this year, ladies and gentlemen. That hurt. Many more to come, I’m sure; let’s just hope the Cubs can deliver some painful blows of their own to opposing teams.

Are the Cardinals for real?

NO. NO. NO. IF YOU ARE WORRYING ABOUT THE CARDINALS, STOP. THEY ARE A FLUKE.

How can I be so sure?

1) Schedule

The Cards have been playing craptacular teams all year. They have only played 5 games against teams at or above .500; in contrast, the Cubs have played 11.

2) Fluky Pitching

By “fluky,” I don’t mean “low BAbip.” The Cards’ pitching staff has a BAbip against of .275, only a few points lower than the Cubs’ .278. What I mean by fluky is “not giving up nearly as many home runs or walks as you would expect given the personnel.” The Cardinals rank first in the NL in walks allowed and third in homers allowed. That’s nice, but given the personnel on the pitching staff, is it sustainable? No. Wainwright is legit, and Piñeiro may in fact be a Dave Duncan reclamation project success story. But Wellemeyer? Looper? No no no no no. They will regress. It will hurt.

This team, especially the offense, will certainly be better than anybody expected. It appears that Ankiel was not a fluke, that Duncan is a useful platoon dude, and of course Pujols is a beast. But no matter how good Dave Duncan is, the pitchers on that staff—especially the starters—are going to start giving up more walks and homers, and their inability to strike people out at a good clip (ranked 12th in the NL as of today) will not help matters. The Cardinals might finish at .500, but they aren’t really in the same class as the Cubs and Brewers. Now for the matchups:

Tonight: Hill vs. Wainwright

The Cubs own Wainwright. OWN him. The Cards haven’t really seen much of Hill.

Saturday Afternoon: Lilly vs. Lohse

Unfortunately, this game is on Fox. Knowing how much Tim and Joe love the Cardinals, we will probably “get” to hear them on this game. Bluh. Anyway, Lilly has actually been pretty damn good against the Cards, while Lohse has sucked against the Cubs (although he did beat Lilly last year in a pitchers’ duel at Wrigley on April 15th).

Sunday Night: Marquis vs. Wellemeyer

Marquis has been so-so against his former team; most of the current Cubs haven’t seen former Cub Wellemeyer.

Look, I understand that baseball is a game of large fluctuations and surprises. The worst team can beat the best team any day of the week; I get it. All that being said, I’ll be a little disappointed if the Cubs can’t take two of three in St. Louis: two of the three pitching matchups clearly favor the Cubs (Wainwright might be a better pitcher than Hill overall, but the Cubs own him), and the third is probably a toss-up. So let’s head on down I-55 to the Gateway City and make some of the “greatest fans in baseball” cry.

EDIT: Yesterday’s game appears to have been a Pyrrhic victory of sorts for the Brew Crew.  Hate to see a good young pitcher sidelined by injury; we saw that far too many times with the Cubs, and it’s simply heartbreaking.  I hope he makes a full recovery.

I heard Pat and Ron mention today that through completion of yesterday’s (4/18) games, the Cubs had seen the most pitches per plate appearance (P/PA) of any team in the majors this year. I frankly couldn’t believe this, so I looked it up: it’s true! The Cubs have seen 3.95 P/PA, the most in the majors. This is wonderful news, though I doubt that the Cubs will maintain their lead in this category all year: part of the rise can be attributed to a few players (Ramirez, most notably) taking more pitches than they have in the past, and they’re probably bound to return to their old ways. Still, some of the rise is real, attributable to guys like Fukudome and Soto taking a lot more pitches than their predecessors at their respective positions. Fukudome sees 4.60 P/PA, which is simply astounding.

Not surprisingly, the Cubs have been taking a lot more walks this year than in years past. The offense as currently constructed impresses me with its versatility: there are three legitimate 30+ homer guys, three (or four) fast guys, and two super-patient guys (Lee and Fukudome). This is team that gets on base, hits for power, and can steal some bases and play “small ball” if absolutely necessary. If Ted Lilly can find his freakin’ 2007 form, the Brewers will have a tough time keeping up (especially if Ben Sheets is really hurt).

A quick back-of-the-envelope assessment of the 2008 Cubs:

Hitting:

Should be better.  Slight age-related attrition for Lee, Soriano (maybe), but significant improvements at C and (probably) at RF.  No one had a “career year” last year, so there’s no reason to expect anyone to fall back a lot from their performance last year.

Pitching:

The Cubs certainly got a little lucky in the BAbip department last year (especially Lilly), and I don’t expect them to put up the league’s 2nd best ERA this year like they did last year.  Still, they have a solid core of starters and a large pool of mediocre guys to plug into the 4 and 5 slots, a bullpen with some excellent power arms, and a few AAA-level guys to plug holes if needed.

What is the Baseline?

Were the Cubs “for real” last year?  Certainly: their Pythagorean record was 87-75, and their “third-order” W-L record according to Baseball Prospectus was 84-78.  So, to use a phrase coined to describe another Chicago sports franchise, “they are who we thought they were.”  Thus, in assessing what impact the additions/improvements to the team will have, we can go ahead and start by using 85 wins as our baseline.

This year:

My method will be very rough, as befits a “quick-and-dirty” back-of-the-envelope style calculation: I will ignore the differences between last year’s WARPs and this year’s mean projected WARPs for all positions except catcher, center field, and right field, then add the differences for those positions to 85.  It’s so simple and dumb it might work.  Here are the relevant numbers:

C: last year’s total WARP1=3.3; this year’s projection=4.8 (assuming backups are at replacement level)

CF: last year’s total WARP1≈4.0 (this is approximate); this year’s projection=3.7 (assuming Pie starts and backups are at replacement level)

RF: last year’s total WARP1≈4.9 (this is approximate); this year’s projection=4.4 (assuming backups are at replacement level)

The net change in WARP is then +.7.  But wait!  There’s a bad assumption here, namely that the backups will be at replacement level.  Last year, the “backups” actually contributed a large chunk to the WARP position totals at both CF and RF (DeRosa, Murton, Pagan, and Ward to RF, Pie and Pagan to CF), and that should be the case this year as well.  PECOTA predicts Fukudome will only play in 110 games, so let’s give the remaining 52 to Murton and use 1.7 of his projected 2.3 WARP for RF.  In CF, PECOTA projects Pie to play in 102 games, so let’s give 30 of the remaining games to Fuld and the other 30 to some replacement level guy.  This gives us a total WARP1 of 6.1 for RF and 4.3 for CF, giving us a net gain in WARP of 3.0 over last year.  So I therefore fearlessly predict that the Cubs will win 88 games!

This exercise has been extremely rough, but I think the end result—namely, that the Cubs should win a handful of games more than last year—sounds about right in that it jives with both my intuition and the the results of various simulations and projection systems.  This is still not a great team, but it is a good one that plays in a bad division, so I expect them to be up in the high 80s/low 90s in wins this year.

Here’s my 25-man roster:

C: Soto, Blanco

IF: Lee, DeRosa, Theriot, Ramirez, Fontenot, Cedeño

OF: Soriano, Pie, Fukudome, Murton

Fat PH: Ward

SP: Zambrano, Lilly, Hill, Marquis, Leiber

RP: Wood, Marmol, Howry, Eyre, Wuertz, Marshall, Dempster

OK, let me explain the bullpen: the way I see it, the Cubs have three power arms (Marmol, Howry, Wood) that should be used in high-leverage situations and two “pretty good” guys (Wuertz, Eyre) who can be used in medium-leverage situations.  But when Marquis and Leiber are two of your starters, you’re going to have a significant amount of low-leverage—that is, mop-up—situations, and there’s no need to give those innings to one of your better relievers.  Hence, Marshall is your “long man A” and Dempster is your “long man B” who can also be used in higher-leverage situations if needed.  I like Gallagher, but I think he needs a little more seasoning.  He and Kevin Hart and Carmen Pignatiello should start the year in AAA, ready to be called up when someone gets hurt.

Cole Hamels and Rich Hill are very similar: both are lanky lefties (Hamels is 6 foot 3, 175; Hill is 6 foot 5, 205) in their first full season in the majors who pitch for decent teams. Hamels, though, has earned a lot more attention and accolades, including a spot on the 2007 All-Star team. Part of this surely has to do with age: Hamels is just 23, and thus qualifies as a phenom, while Hill is already 27. Even correcting for that, though, the general perception seems to be that Hamels is simply better, or at least has been this year. Is that true? Let’s look at their away numbers for this year (I’ll ignore home stats as a crude correction for ballpark effects):

Hamels: 92 IP, 3.42 ERA, 2.05 BB/9, 9.29 K/9, 1.17 HR/9

Hill: 82 1/3 IP, 3.83 ERA, 2.84 BB/9, 9.18 K/9, 1.53 HR/9

It seems that Hamels is better, yes, but not by much. The main difference is that Hill is more susceptible to the home run ball.

I’d like to do something now, something very dangerous and somewhat questionable. I want to subtract Hill’s May 22nd start at San Diego from his line and then do the comparison to Hamels again. Now, I know that this looks like a form of the old “if you ignore his bad starts, he’s really good!” selection bias game, but I don’t think it is. You see, I really believe that Greg Maddux, Hill’s mentor with the Cubs, picked up on something in Hill’s delivery that was tipping his pitches and relayed it to the Padres. Hill has faced the Padres twice this year, and he’s given up seven homers in nine innings. Now, if this were the Mets or Brewers, I wouldn’t be surprised. But the Padres? They’ve hit 119 homers all year, so that means that, in nine total innings, they hit 6% of their homers for the year so far off of one pitcher. It’s not an open-and-shut case, but I find the evidence compelling. So what happens if we subtract that start from Hill’s road line?

Hamels: 92 IP, 3.42 ERA, 2.05 BB/9, 9.29 K/9, 1.17 HR/9

Hill: 76 1/3 IP, 3.54 ERA, 2.83 BB/9, 8.96 K/9, 1.18 HR/9

Wow. Pretty much the same guy. Hill walks more people and strikes out fewer, but they’re pretty much the same.

I think Hamels, due to his youth, will probably have a better career than Hill. But right now, I think their differences are pretty minute.  So where’s the love for Hill?

Some people like to endlessly point out when the Cubs get lucky, and there’s nothing wrong with that: it’s good to know when you might be about to fall off a cliff. But looking at team-wide luck-indicator stats like BAbip (batting average on balls in play) sometimes obscures individual performances. I want to look at the three Cubs starters with significant track records (Z, Lilly, and Marquis), compare their important 2007 stats (Ks, BBs, HRs, and BAbip) to their career numbers, and then assess whether they’ve been lucky or good (or both). I’ll start with Z:

2007: K/9=7.68, BB/9=3.90, HR/9=1.07, BAbip=.262

Career: K/9=7.93, BB/9=4.09, HR/9=.74, BAbip=.272

The strikeouts are down a little, though not too much, and the walks are down a little as well, compared to career numbers. Compared to last year, when Carlos struck out 8.83 and walked 4.84 per 9 IP, both numbers are pretty low. The HR/9 number is strikingly higher than his career norm, though last year’s .84 HR/9 was also a little high. The BAbip indicates that Carlos has been a little lucky, but not that much; were he having a “normal” Carlos year, he would have given up about 4 more hits to this point. (Incidentally, I think for pitchers with a significant track record, it’s more useful to compare BAbip to career norms rather than current league numbers. I realize that offensive contexts change and that there is some evidence that pitchers have little control over BAbip, but I also believe that very good pitchers–Carlos, Maddux, some others–do have some control. Carlos has posted BAbips well below league average for the last four years, and I don’t think that’s just luck. When coupled with his K numbers, it points to a pitcher whose movement on the ball makes it hard for hitters to make solid contact, leading to a lot of misses and weakly hit balls in play.)

What to make of this? Well, I think it should be pretty clear that Carlos is (mostly) being Carlos. Since the Barrett punching incident, Carlos has posted the following rates in 10 starts and 69.1 IP:

K/9=9.22, BB/9=3.63, HR/9=.52, BAbip=.193

He should have given up a few more hits in this time (about 12), but the home run, strikeout, and walk rates are great, and sort of indicate that last year’s walk-fest may have been an aberration. So while there will probably be a bit of a regression in BAbip, I expect Carlos to finish the year with great numbers–maybe Cy Young great. Conclusion: mostly good, a little lucky.

Onto Ted Lilly:

2007: K/9=7.33, BB/9=2.17, HR/9=.95, BAbip=.263

Career: K/9=7.64, BB/9=3.49, HR/9=1.33, BAbip=.280

The obvious differences here are the walk and homer numbers. The BAbip indicates a certain degree of luck, though not much,–Lilly “should” have given up something like 6 or 7 more hits all year so far–and the strikeout numbers are right in line with career norms. So what to make of this? Has Lilly, at 31, really learned better control, and thus cut down on his walk and home run numbers? My (perhaps naive) hypothesis is this: in moving from a tough league and a somewhat tough division to a bad league and really bad division, Lilly has been facing lesser competition, and has had to be less “fine” with his pitches, challenging hitters more often. As a result, he hasn’t walked as many people. The drop in home runs is attributable to the weaker competition. Now, is this true? I don’t have the time to definitively answer this question, but I can tell you this: last year, Lilly gave up 28 homers in 181.2 IP–a rate of 1.39 HR/9. He gave up 21 of those to the Angels, Red Sox, White Sox, Tigers, Twins, Yankees, A’s, and Blue Jays–in other words, to the good AL teams. His HR/9 rate against all other opponents was .62. I think Ted Lilly is the answer to the question “what happens when a slightly-better-than-average AL pitcher moves to the NL?” Conclusion: good, and smart for changing leagues.

Now for Marquis:

2007: K/9=5.15, BB/9=3.46, HR/9=1.08, BAbip=.249

Career: K/9=5.42, BB/9=3.48, HR/9=1.26, BAbip=.282

Uh oh. Same strikeout and walk numbers, pretty much, as he has shown in the past. The homer numbers are down–could it be the persistent incoming wind at Wrigley?

HR/9 at home in 2007: .67

HR/9 on road in 2007: 1.64

So, yes. Actually, Marquis has been pretty terrible on the road, walking more than he’s struck out in 49.1 IP. His BAbip also indicates that he’s been very lucky; he should have given up about 12 more hits over the course of the season. I have to think that Jason should be thanking his lucky stars for that Wrigley wind blowing in. Still, I’ll take him as a fifth starter. Conclusion: lucky, lucky, lucky.

So there you have it: our 1-2 veteran punch is “for real,” while our #5 starter is a little flukey. Hill and Marshall don’t have enough innings pitched in the majors to say too much, although Hill’s .251 BAbip seems to be a little lucky. Still, we have a rotation with three pitchers of solidly above-average ability, one with average ability, and one whose ability is not yet clear. I’ll take that.