Showing posts with label Adrian Gonzalez. Show all posts
Showing posts with label Adrian Gonzalez. Show all posts

Monday, July 7, 2014

Waking up to the Dodgers in first place

There's a reason major league baseball seasons are 162 games and not 62 games. Sixty two games are not enough for skill to dominate luck in the results. Just as a poker hand is largely luck, so is a baseball game. But if you play enough of either, it'll be clear who's got more skill.

I took a look at the skill of each player on each roster and concluded the Dodgers ought to run away with the division. The Giants started off winning way more games at a way faster rate than I thought they would, the Dodgers way slower. But lately, talent has been shining through.

The Dodgers have more top end talent and more depth. Depth is important for two reasons. First, you never know who's going to overperform or underperform expectations, but if you have several guys that can carry a team, one or another probably will (Adrian Gonzalez slumps but Yasiel Puig is hot). Second, injuries happen. How the roster can reconfigure around an injury is critical (Carl Crawford is out but they still have four solid outfielders).

Do I still feel like the Dodgers will coast into the playoffs? Well, I thought the Dodgers had the talent to be a 98-win team, and the Giants an 83-win team. According to the individual achievements of the players so far, the Dodgers are on pace to be a 92-win team and the Giants a 81-win team. BUT, because of luck, timing and other things that are not likely to continue, the Dodgers and Giants have won about the same number of games, both on a 90-win pace. I was probably a bit ambitious with my Dodgers predictions, but right on with my Giants ones. The Giants are not very good. To be a good baseball team, you need many good and very good baseball players, and the Giants just don't have that. Given what's already in the books for this season so far, mix that with the talent-levels going forward (of course we have yet to see what the trade deadline brings) and we see the Dodgers finishing with maybe 92 wins, and the Giants 85.

Here's a helfpul talent visualization. The horizontal axis is each player's projected (by me) WAR. The vertical axis is the average of the FanGraphs and baseball-reference WAR achieved to date, scaled to a full season. Click to see it at a larger scale. The blue names are Dodgers hitters, the red names Dodger pitchers. The orange names are Giants hitters, the black names Giants pitchers. You can see who has over and underachieved (underachievement comes in some cases from time missed due to injury, and you wouldn't expect the same people to be injured in the second half of course).  By the way, "Staff" is the WAR due to pitcher hitting (it counts, and both teams are relatively good at it).



What you see in this graph is a lot of blue and red on the upper right. It looks like the teams have a comparable number of mid-range players but remember those are the Giants best players but the Dodgers second-tier players. At the worst end, the Dodgers have guys they can cut like Paul Maholm & Chris Perez. They can at least get replacement-level guys in there to replace them! While the Giants have Matt Cain & Sergio Romo that are not going anywhere. Matt Kemp is also a sore spot for the Dodgers because his defense has been unfathomably bad, but it's reasonable to think he'll improve all the way up to just 'bad' so his valuable offense can continue to play.

So bottom line, I'm feeling pretty good about the projections and about the Dodgers. Most of the misses in individual projection are due to playing time, with a few surprises nobody could see coming. And in the aggregate, the projections are pretty on. The Dodgers meanwhile continue to look substantially stronger than the Giants and that ought to play out over the next 75 games.

Photo credit: File photo

Friday, May 30, 2014

Shuffling deck chairs in the Dodgers' outfield, featuring Matt Kemp

Earlier this year, I questioned if the Dodgers defensive weakness up the middle would work out. I looked at it in terms of offsetting factors that could make it work -- having good corner defense and getting good offense from the defensively questionable middle. Plus, while I accepted the gloves up the middle would be bad, I thought they'd be bad within the range of Major League players.

Best case, Matt Kemp would have speedsters Carl Crawford and Yasiel Puig covering so much ground in the corners, he could cover a minimal center field and the team would be in good shape. And since his bat is much better than the average center fielder, the team, all in all, would have a great outfield situation. Similarly, Juan Uribe could cover some ground and let Hanley Ramirez cheat to the middle a little. And counting the offense and defense, the Dodgers' infield would also be in great shape.

Just on team defense, while the Dodgers have indeed committed a lot of errors, eight of those have come from pitchers (all last year Dodgers' pitchers committed 10). That rate should slow down. But a better metric to look at for team defense is defensive efficiency -- what percentage of balls in play turn into outs. The Dodgers are about middle of the pack, maybe a little lower (but definitely not in the basement) on either a straight or park-adjusted basis. So, it's not a disaster situation overall, but there is room for improvement.

Kemp and Ramirez's defense has been much worse than feared. Their offense has been good, not great, so we are not willing to overlook as much. At the same time, the Dodgers corner defenders are not necessarily lighting up like they're supposed to either. Crawford and Adrian Gonzalez are a couple notches below where they used to be. Puig and Uribe are doing their part, and Andre Ethier has been fine. Second base of course has turned out to be mostly Dee Gordon, who is more or less getting the job done out there.

So the question comes down to Kemp and Ramirez. Neither one has probably been playing at full strength. Both have been terrible in the field. Their UZR/150 ratings (a measure of defensive play converted to runs saved above/below average for that position in the major leagues, scaled to a full year) are in the -25 to -30 range.  Negative 10 is about as bad as you can hope to be and hold a job. Out of 161 Major League regulars (non-catchers), all rankings normalizes for positions played, Hanley & Matt are very near the bottom in both major aspects of defense. You can't be bad at both and stay where you are. Gordon's error rate is just as bad, but he does compensate with above-average range.



Range Rank
Error Rank
UZR/150 rank
Ramirez
154
153
154
Kemp
159
148
156


The season's about a third over, and the Dodgers are not running away with the division as expected. In fact, they're 6 1/2 games behind the Giants. The hitting is good, the starting pitching is good (especially now with the rotation healthy), the bullpen is a problem and poor defense is exacerbating the bullpen problem.

So, what to do? Let's leave the shortstop question aside for a later day. Don Mattingly has done one bold thing, which is stop sending Kemp into center field every day. The intention is he'll play mostly left field now and a little bit of center. Ethier will play center field most days, while Puig is in right field every day.

What can we expect from Matt Kemp, left fielder? Well, left field is easier for sure. I thought I'd look at Major League players who played a good amount of both left- and center field last year, and see how their performance was at each spot. There are only three such players -- Alejandro De Aza, Gregor Blanco and Mike Trout.

We see an average increase in UZR/150 of about 5 when we look at their left field ratings vs. their center field ratings. The error rate is going to be about the same, but limited range is less of a problem in left. Now, if Kemp only improves from a -30 CF to a -25 LF, I'll be surprised. Then again, his ankle is clearly not 100 percent, so it might be awhile before we see him running 'em down again. And playing left field, if you're used to center- or right field, is not entirely easy. Just look at Ethier. While solid in right- and center field, he's actually struggled a bit in limited time in left.

Kemp's attitude is not great right now. He apparently "took a few days" to get used to the idea of not being the center fielder. And then he shagged balls in left field for half an hour, once. With Crawford injured, the outfield rotation is pretty set now: from right-to-left field, it'll be Puig/Ethier /Kemp vs. righties and Puig/Kemp/Van Slyke vs. lefties (even if Ethier is starting on Friday against the Pirates).  Puig, an above-average right fielder, could be turned into a good center fielder, but in that would take time.

It's not clear that even once that learning takes place, Puig in center field and Ethier in right field is a better situation than Puig in right and Ethier in center. So leave that be.

Jamie Romak, for now, is the fifth outfielder, with Chone Figgins and/or Gordon as emergency options after that. Like always, the four outfielder situation seems to have resolved itself without a trade, but there's definitely going to be some trade talks with teams as we get closer to July 31 (i.e. the Red Sox, Rangers, Yankees, etc.). Because Crawford is coming back at some point and Joc Pederson continues to knock on the door.

It may not be fair, but how Kemp and Ethier play over the next month or so will impact which one gets traded. If Crawford was almost entirely untradeable while healthy, he is entirely untradeable on the disabled list.

Photo credit: SD Dirk, Flickr

Tuesday, May 6, 2014

Dodgers still project to win NL West comfortably, despite slow start

Preseason, I felt confident the Dodgers would run away with the National League West. Partially because they have so much talent, and partially because nobody else does. But the Giants and Rockies are off to great starts, the Dodgers are showing some holes and the Dodgers are in (gasp!) third place as of the morning of May 5. Do we expect a real race from either of these teams?

Well, first suppose the preseason projections stand unchanged, and I still expect each team to win the percentage of their remaining roughly 130 games that I originally expected them to win of 162. Add that to their current win total and where would things sit?


Preseason
To-date
To-date + preseason
LA
98
18-15
96
SF
82
21-11
87
COL
78
20-14
82

The Dodgers have such a talent advantage (ignoring any information learned this season about all the respective players), they would still run away with it. But of course, everything is not going exactly as expected -- we should incorporate the information from this season into the forward projections. Some guys are doing much better than expected (Dee Gordon, Charlie Blackmon, Brandon Crawford) while others are doing much worse (Carl Crawford, Pablo Sandoval, Carlos Gonzalez). Some players are injured and not expected back until a certain time. Playing time is being allocated differently than expected, etc. So, I want to update the projections for these three teams, and see what to expect.


Dodgers
Giants
Rockies
Adjusted Up
Dee Gordon 
(offense & defense),
(Offense & Defense),
Charlie Blackmon
Adjusted Down
(offense & defense),
(offense & defense),
Carl Crawford,
Pablo Sandoval,

So if I knew then what I "know" now about all the players' performance, I would take two games each from the Dodgers and Giants while giving two to the Rockies. Bottom line, counting the performance to date and the updated projections, I still see the Dodgers comfortably advantaged although with only half the lead I once thought they'd have.

Going Forward
To-date
To-date + Forward
LA
95
18-15
93
SF
80
21-11
85
COL
80
20-14
83

Why have the first 33 games or so not played out entirely like expected? Basically, the Dodgers are doing all the right things, just not at all the right times. Some research by Dave Cameron revealed the Dodgers could have reasonably been expected to score 15 more runs and allow seven fewer had the number of each event (home run, walk, stolen base, great catch, etc.) been the same but the order been shuffled. This 22-run deficit due to sequencing is the largest in the majors. The Giants, conversely, have nine more runs than their context-free play would suggest. The Rockies are just getting a bit lucky with balls falling in for them. Baseball's a funny game in short time; there's a reason the season is so long.

Photo credit: File photo