Wednesday, August 31, 2011

Top 25 Highest Paid Players and WAR

I found this list of MLB's 25 highest paid players in 2011* and decided to take a look at their WAR values for this season to determine which players are providing the best value for their salary so far, and which ones are just dead weight. I averaged out each players' WAR value from FanGraphs and Baseball Reference, just to get a better overall look. Now obviously some of the players on this list have been injured this year, and I haven't taken this into account, but that's part of the gamble when you're signing a player to a huge contract.

Rank Player Team 2011 Salary B-R WAR FG WAR AVG WAR $/WAR
8 Roy Halladay Phillies $20,000,000 6.4 7.2 6.8 $2,941,176
21 Matt Holliday Cardinals $16,317,774 4.4 5.1 4.75 $3,435,321
19 Josh Beckett Red Sox $17,000,000 5.6 3.6 4.6 $3,695,652
8 Miguel Cabrera Tigers $20,000,000 5.1 4.8 4.95 $4,040,404
3 CC Sabathia Yankees $24,285,714 5.5 6.2 5.85 $4,151,404
11 Carlos Beltran Mets $19,325,436 3.1 3.6 3.35 $5,768,787
22 Michael Young Rangers $16,174,974 2.2 3.4 2.8 $5,776,776
12 Carlos Lee Astros $19,000,000 3.4 2.6 3 $6,333,333
7 Todd Helton Rockies $20,275,000 3 3 3 $6,758,333
4 Mark Teixeira Yankees $23,125,000 2.4 4 3.2 $7,226,563
23 Jake Peavy White Sox $16,000,000 0.9 2.6 1.75 $9,142,857
1 Alex Rodriguez Yankees $32,000,000 3 3.9 3.45 $9,275,362
15 Torii Hunter Angels $18,500,000 1.5 2.1 1.8 $10,277,778
23 Roy Oswalt Phillies $16,000,000 1.2 1.7 1.45 $11,034,483
8 Ryan Howard Phillies $20,000,000 2.2 1.3 1.75 $11,428,571
5 Joe Mauer Twins $23,000,000 1.3 1.3 1.3 $17,692,308
12 Alfonso Soriano Cubs $19,000,000 1 1.1 1.05 $18,095,238
14 Carlos Zambrano Cubs $18,875,000 0.8 1 0.9 $20,972,222
20 A.J. Burnett Yankees $16,500,000 0 1 0.5 $33,000,000
17 Jason Bay Mets $18,125,000 0.4 0.1 0.25 $72,500,000
6 Johan Santana Mets $21,644,707 0 0 0 NA
25 John Lackey Red Sox $15,950,000 -1 1 0 NA
18 Ichiro Suzuki Mariners $18,000,000 -0.6 -0.1 -0.35 ($51,428,571)
15 Barry Zito Giants $18,500,000 -0.4 -0.2 -0.3 ($61,666,667)
2 Vernon Wells Angels $26,187,500 -0.5 -0.1 -0.3 ($87,291,667)

So there you have it, of the top 25 highest payed players in the majors, Roy Halladay is providing the best value for his team in 2011 - providing 6.8 wins above replacement on a salary of $20 million. That works out to be just under $3 million/WAR. Coming in second, somewhat surprisingly, is Matt Holliday. It's hard to say how his contract will pan out by the end of 2016 when he's 36, but the Cards have got their money's worth in the first two years.

Completely unsurprisingly, the player providing the worst value for his salary is none other than Vernon Wells. He's getting the second most money in baseball this year, and he's cost his team 0.3 of a win. Do I need to mention again how amazing that trade was for the Blue Jays?

For teams as a whole, here's how the ones who have more than one player on the list average out:

Phillies - $56 million for 3 players - Halladay, Oswalt and Howard - for 10 total WAR - $5.6 million/WAR

Red Sox - $33 million for 2 players - Beckett and Lackey - for 4.6 total WAR (all from Beckett) - $7.2 million/WAR

Yankees - $95 million for 4 players - A-Rod, Sabathia, Burnett and Teixeira - for 13 total WAR - $7.4 million/WAR

Mets - $59 million for 2 players - Beltran (I'm including all of Beltran's WAR for the Mets) and Santana (who's missed the whole year) - for a total of 3.6 WAR - $16.4 million/WAR

Cubs - $38 million for 2 players - Zambrano and Soriano - for 1.95 total WAR - $19.4 million/WAR

Angels - $45 million for 2 players - Wells and Hunter - for a total of 1.5 WAR - $29.8 million/WAR

Is it coincidence that the 3 highest ranked teams will make the playoffs this year, while the latter 3 will miss them? (I realize the Angels still have a chance to make it. I do not know how they are able to accomplish this while committing so much money to two horrible outfielders, but somehow they do.)


*http://content.usatoday.com/sportsdata/baseball/mlb/salaries/player/top-25

Stadium Park Factors

I've been thinking a lot about stadiums lately, and specifically how it seems like a lot of the newer stadiums are being built to favor pitchers. This strikes me as a strange phenomenon, because it seems to me that if one were to want more fans to come to a ballgame, it's a lot easier to entice them to come if the score has a better chance of being 6-4 than 2-0. Obviously teams can't completely control how their new park is going to play until it's built, but it seems self explanatory that a field with New Yankee Stadium dimensions is going to be more of a hitters park than one with the dimensions of Citi Field. I decided to take a look at the Park Factors of all 30 stadiums and compare that to how recently the stadium opened. If I was correct in my assumption, the more recent stadiums would mostly favor pitchers, in comparison to the older hitter-friendly confines of Fenway and Wrigley.

The park factors that I used are the multi-year park factors from Baseball-Reference. I decided to use just the hitters rating, although for most parks the pitchers rating is exactly the same.

The first thing I did was split the parks into 3 groups of 10 based on their opening date. This came up with 3 groups:

Group 1 - Fenway Park - Sun Life Stadium - 1912-1993 - average park factor of 100.2
Group 2 - Ballpark at Arlington - AT&T Park - 1994-2000 - average park factor of 100.3
Group 3 - Miller Park - Target Field - 2001-2010 - average park factor of 100

At first glance, not much of a difference between any of the 3 groupings. Each of the last two groupings has at least one very hitter friendly park though, with Coors Field at 118 in Group 2, and Yankee Stadium at 109 in Group 3. This method of grouping them didn't really come up with any conclusive results, so I decided to look at the parks a little differently.

I next split the stadiums into hitter friendly and pitcher friendly (with 100 being included in hitter friendly.) This came out with 17 pitcher friendly parks, and 13 hitter friendly parks. Of those 13 hitter friendly parks, the breakdown is as follows:

Group 1 - 6 teams
Group 2 - 3 teams
Group 3 - 4 teams

Pitcher friendly:

Group 1 - 4 teams
Group 2 - 7 teams
Group 3 - 6 teams

This shows that of the 10 most recently built parks, 6 are classified as pitcher friendly, so that's more than half, but still not as many as the 7 pitcher friendly parks that were built between 1994 and 2000. Still not really conclusive.

The next thing I did was split up the stadiums into only 2 groups, the 16 oldest, and the 14 newest. The reason I cut them into two uneven groups is because it seems that's what Bud Selig would want.  Not really, it's actually because the cut-off falls between Tropicana and Chase Field, which both opened in 1998. I put both of these into the older group, because it seemed like the thing to do.

In this case, the older group has an average park rating of 101.31. This is in comparison to 98.86 for the newer group. Finally a somewhat significant result. Of these newest 14, only 4 are considered hitter's parks (Yankees, Phillies, Reds, Brewers), while the other 10 favor the pitchers. This is in contrast to 9 hitter's parks and 7 pitcher's parks in the older group. So in this regard, it does appear that teams building new stadiums are leaning towards pitcher friendly parks. I don't think that this is what I would do if I was making the decision, but it seems to be trending in that direction. I'd like to take a look at attendance figures for hitter and pitcher friendly parks, but then you have a ton of other factors to consider, such as team performance and location. Perhaps someone with considerably more intelligence or free time will someday take a look at this. (Or perhaps they already have.)

Tuesday, August 30, 2011

David Price

David Price had, for all intents and purposes, a great year in 2010. He went 19-6 with a 2.72 ERA, good enough for a second place finish in Cy Young voting. At first glance, his 2011 season to date looks to be not as impressive. 12-11 with a 3.40 ERA this year, David Price's season hasn't gone quite as well as many hoped. When you dig a bit deeper into his stats though, this season takes on a different light. To start off, Price outperformed his FIP last year by a significant margin - 3.42 to 2.72, good for a .70 advantage. To compare, this year his FIP has actually decreased to 3.19, while his ERA increased to 3.40. This shows that Price perhaps benefited from better defense behind him or better luck in 2010 than he has this year.

A similar story occurs when looking at Price's peripheral stats. The most significant change is his decrease in BB/9 from 3.41 in 2010 to 2.22 in 2011. Price's H/9 has increased this year from 7.3 to 7.6, but his BABIP has also increased from .270 to .282 showing that it probably hasn't been solely his fault. Combined, Price's WHIP has decreased from 1.193 to 1.096, or about 1 baserunner/9 innings. The only stat that has really declined this year over last year for Price is his HR rate. He has allowed 3 more HR this year than he did in all of 2010, and his HR/FB rate jumped from 6.5 last year to 9.4 this year. This is likely due to some bad luck, or else his 2010 rate was unsustainably low.

A significant contributor to Price's poorer W/L record this year has been the decrease in run support by the Rays. Last year, Price received 5.3 runs of support/start, while this year that number has dropped to 4.01.

Price's WAR figures also back up the notion that Price is having a very comparable, or perhaps even better year in 2011 than in 2010. FanGraphs lists his WAR at 4.3 for 2011, the exact same figure as in 2010, although in 3 less starts and 18 less innings pitched. All told, Price should probably have a better W/L record this year than he does, but if you look past the basic stats, Price's year has still been very good.

In Defence of the Wild Card

It seems as though a lot has been written this year about Bud Selig and his decision to add an additional wild card team in each league. Many have decried that this will only further lessen the tension of pennant races. Most of these people seem to be neglecting to consider the whole picture though, by bringing up past pennant races in which the race came down to the final day of the season, such as 1993 when the Braves and Giants fought till the very end for the NL West crown. Adding an extra wild card team in each league may in some cases reduce September drama, such as last year in the NL West when the Padres and Giants both would have made the playoffs, but in other cases it would add additional drama. Just look at this year for example.

In the American League, the Division winners are basically all decided. The Angels could still catch Texas from 3.5 games back, but that's the only possible race. If an additional wild card team was added, suddenly you have the Tampa Bay Rays and LA Angels separated by only 1.5 games going into September. Adding an additional wild card team would in this case tighten up the playoff race going down the stretch.

Similarly in the NL, the closest race is in the West, where San Francisco is 5 games back of Arizona, and looks all but out of it. Add another wild card team, and suddenly San Francisco and St. Louis are only separated by .5 games and primed for an exciting September. Now obviously it's not going to work out this way every year. Some years the race for the #4 spot (current wildcard) in each League will be close, and it would make for a more exciting September if there were only one wild card spot. Other years, like 2011, the race for the #5 spot (additional wild card) will be closer and an additional playoff team will allow for a more thrilling final month. Picking and choosing certain years in the past where there was an exciting pennant race is a misleading way to convince people one way or another on expanded playoffs.

A lot of folks use 1993 as an example of how the wild card has ruined pennant races. Then there are cases like last year in the NL West that come along, and make you wonder what these people are complaining about. The same thing will happen with adding an additional wild card team. Just because there's one more team being added, doesn't mean that there aren't going to be exciting races. You can't just look at each year individually when thinking about playoff races. Some years there are going to be great races and some years there aren't. Adding another wild card team will do nothing to change this fact.

The Philllies' Big Three by Game Score

First off, I realize that this makes two straight posts that have a Phillies theme to them. This was not intentional. Please do not think that I have suddenly become a huge Phillies fan, as I have not.

By all accounts, the Phillies' big three are all having excellent seasons. Roy Halladay leads the group with a 15-5 record, while sporting a 2.56 ERA. Cliff Lee sits at 14-7 with a 2.71 ERA, while Cole Hamels rounds out the 3 with a 13-7 record and a 2.58 ERA. The advanced metrics back up those stats, with Roy Halladay first in WAR among pitchers at Fangraphs with 6.9 (6.0 at B-R), Cole Hamels at 6th with 5.3 (5.3 at B-R), and Cliff Lee at 8th with 5.2 (5.2 at B-R). I wanted to take another route at comparing the three though, so I decided to have a closer look at their individual pitching game scores (as per baseball-reference).

Conveniently for the sake of comparison, all 3 pitchers have made 26 starts this year as of this posting. I'm comparing their game scores to see if any of them stick out as being more consistent as well as each pitcher's high and low scores. Here is a summary of the three:



avg high low st dev sub 50 50-59 60-69 70-79 80+
Halladay 62.31 83 31 13.01 5 4 8 7 2
Lee 62.54 92 18 19.55 6 4 5 5 6
Hamels 62.73 81 18 16.34 4 3 8 10 1

The first thing that should be mentioned is how close their average game scores are. I suspected that Halladay might have the highest average, but he in fact has the lowest average, although only by .4/start. Hamels actually comes out on top in terms of highest average score at 62.73, while Lee falls almost exactly between the two.

Lee has the highest game score of the three this year with a 92, a complete-game 3 hit shutout that came April 14 against the Nationals. He struck out 12 while walking only 1 in that game. Somewhat curiously, this came in the start directly following Lee's worst start of the year, which tied with Hamels as the worst game score of any of the three - in which Lee went 3 1/3 while allowing 10 hits and 6 earned runs - good for a game score of 18. Halladay's high was an 83 - 8 2/3 of 1 run ball against the Padres on April 24 with 14 K's. His best game score actually came in the start directly following his worst game score of the year as well, a 31 coming in 6 2/3 against the Brewers in which he gave up 10 hits and 6 runs. Hamels high was an 81, on July 22 against the Padres in which he went 8, while scattering 3 hits and striking out 10. You might have noticed a theme with the previous 2 that their highest game score of the year came directly after their lowest, and Hamels it seems is not one to break tradition, in that his low score of 18 came in the start just before he hit 81. His 18 came against the Mets in which he gave up 8 runs over 4 1/3.

As for consistency, Halladay's game scores had the lowest standard deviation, at 13.01, while Lee had the largest deviation at 19.55. This can be highlighted by Lee having 6 starts with game scores of less than 50, but also 6 game scores of 80 or above - the highest total in both ranges. Lee's top 3 game scores of 92, 86, and 85 are actually higher than Halladay or Hamels hit in any start this year.

Any way you look at the Phillies' big three starters, they appear to be a pretty scary sight for any team come October.

The Phillies vs >.500 Teams

While looking at the MLB standings today, I noticed that the Phillies had only played 47 games against teams with records above .500. This is the lowest amount of any team in baseball, and is exactly half the amount of Minnesota, who have played 94 games against teams with records above .500, the most of any team in baseball. In their 47 games against teams above .500, Philadelphia has gone 26-21, good for a .553 winning percentage. Compared to their record against teams below .500, in which the Phillies have gone 58-25 (.699), this .553 winning percentage is substantially lower. Now, to be fair, most teams in baseball have better records against sub .500 teams. As of games finishing August 29, all of MLB has a .443 winning percentage against >.500 teams, but a .565 winning percentage against <.500 teams. Still, this got me wondering how the MLB standings would look if every team had played 94 games against >.500 teams this year, just like the Minnesota Twins. Obviously this is impossible, but it's not supposed to be a factual study, but rather a fun exercise.

What I did was apply every team's winning percentage against teams with records above .500, and teams with records below .500 to the Twins breakdown of games - 94 games against >.500 teams, and 40 games against <.500 teams. This is what the standings would look like as of today:



AL



NL


EAST W L GB
EAST W L GB
BOS 82 52

PHI 80 54
NYY 80 54 2
ATL 79 55 1
TBR 73 61 9
NYM 64 70 16
TOR 65 69 17
WSN 62 72 18
BAL 53 81 29
FLA 54 80 26









CENTRAL W L GB
CENTRAL W L GB
DET 74 60

MIL 71 63
CHW 66 68 8
CIN 66 68 5
CLE 65 69 9
STL 62 72 9
MIN 56 78 18
PIT 61 73 10
KCR 55 79 19
CHC 56 78 15





HOU 43 91 28
WEST W L GB




TEX 72 62

WEST W L GB
LAA 71 63 1
SFG 70 64
OAK 60 74 12
ARI 67 67 3
SEA 57 77 15
LAD 60 74 10





SDP 57 77 13





COL 57 77 13




As you can see, not wildly different than the actual standings, but there are some changes. The only division lead change came in the NL West, where the Giants jumped ahead of the Diamondbacks. This change had almost nothing to do with the Giants (70-64 compared to actual 71-64), but was based entirely on Arizona, who changed from 76-59 to 67-67. This was due to their .442 winning percentage against teams with winning records, at 23-29 in contrast to their record against losing teams at 53-50 for a .639 winning percentage.

As for the Phillies, they drop 4 wins, going from 84 to 80 and dropping to only 1 game up on the Braves, losing 5.5 games of their actual lead. The AL West also gets a bit closer with the Rangers dropping a few wins and ending up only one game up on the Angels. The NL Central as a whole all got worse, starting with Milwaukee, who lost 10 wins. Luckily for them, the rest of the division fared poorly as well, so they still sit comfortably with a five game lead over the Reds. The Wildcard races stay virtually the same, with the winners both still coming out of the East in both leagues.

The only team with a better winning percentage against winning teams than losing teams so far this year is the Detroit Tigers, at 40-32 vs >.500 (.556) compared to 33-28 vs <.500 (.541). In conclusion, there's probably nothing too groundbreaking to be learned here, I just found it interesting comparing how teams would fare if they all played the exact same number of games against winning and losing teams.





Welcome

Welcome to my new blog, Two-Seam Statistics, a baseball blog focusing mostly on statistics. There will probably be a bit of bias towards the Toronto Blue Jays, as I'm Canadian, or the Cincinnati Reds, because they were my father's favorite team, but there will surely be a lot of non-team-specific postings as well. I haven't actually done much writing in my life, so please forgive the odd grammatical error or run-on sentence. Anyway, please enjoy.


Josh