Sunday, November 13, 2011

The Dickey Report



Why write an analysis of R.A. Dickey?
What’s so interesting about a 36 year old #3 starter on a 4 th place team in his 9 th season, seven of which were abysmal?  Part of what makes a statistical analysis of Dickey interesting is that for seven years he followed a very predictable pattern and then, with little notice, totally breaks the pattern and becomes a new pitcher. In his first full season he posted an ERA over 5, a WHIP over 1.4 and gave up over 10 hits per nine innings. For the next few years he continued along this path: He never broke a 5 ERA until 2009  (even then it was only 4.62) his WHIP hovered between 1.5-2 and the most innings he ever managed to pitch was 116. His lowlight was 2006 when he pitched 3.1 major league innings in which he gave up 8 hits 6 of which were home runs. His 2006 line will forever show an ERA of 18.9 and a WHIP of 2.7. I can’t imagine anyone who didn’t write off R.A. Dickey after the 3 inning 2006 disaster.
But then, it clicks. The forkball he was throwing all his life turned into knuckleball, which he honed over the course of a decent season with the Twins. He gets invited to Mets spring training in 2010 but doesn’t make the team outright. He gets called up on May 19 to pitch six innings, gives up 2 earned runs and strikes out two. At this point he breaks away from what any statistical model would predict and goes on to have a classic comeback season. He posts a 2.84 ERA (10 th best in all baseball) a WHIP of 1.19 and pitches 174.1 innings. His 2010 season breaks the mold of his whole career.  But flukes happen and critics assume he was a flash in the pan. However he defies the odds again and settles in for another successful season as the Mets third starter. He pitches 208 innings and posts a 1.23 WHIP and a 3.28 ERA. Baseball-reference.com gave him a WAR of 4.9, 6 th best in the National League.
Purpose
The goal of my report was to establish where in every MLB team R.A. Dickey would fall in the rotation. Clearly I am an R.A. Dickey fan and since his rise have been making claims about how good he is and how much better he is then other pitchers. Through this I hope to provide a statistical basis for how good or bad R.A. Dickey is and where he falls in comparison to your favorite teams starters.
Methodology
In order to establish where in each teams rotation Dickey falls I used a simple system of ranking statistics. The goal was to judge Pitchers by pure pitching ability, so I exclude many traditional stats such as wins and ERA because they are generally affected by non-pitching factors. The stats I used were: Innings Pitched, WHIP, FIP (fielding independent pitching), Hits per 9, Walks per 9 and Strikeouts per 9. I then listed the stats for each player on a given team and ranked each players place amongst each stat. I then tallied each player’s number and assigned them a rotation spot based on the lowest score being the ACE and the highest score being the 5 th or 6 th starter.
For example, lets look see how Dickey fares in the Yankees lineup:
1.34
9.33
2.72
5.89
4.12
146.00
1.23
8.71
2.19
5.78
3.77
208.00
1.22
8.72
2.31
8.72
2.88
237.10
1.33
8.87
3.1
5.33
4.01
165.1
1.29
9.42
2.2
7.39
3.83
164.00
1.44
8.98
3.9
8.18
4.77
190.00

Above is a table, which shows R.A. Dickey and the New York Yankees starting staff.
1

3
1
1
1
8

5
5
4
4
5
6
29

4
3
5
6
4
4
26

3
6
2
3
3
5
22

6
4
6
2
6
3
27

2
2
1
5
2
2
14
2nd

Above is the table, which shows how the rankings are determined. For instance in Row 1 CC Sabathia has the lowest WHIP and so he received a 1, Dickey has the third lowest so he received a 3. This is done for each category then the numbers are summed and the lowest score received the first spot. In this case, Dickey had the second lowest score and is therefore considered the 2 nd best Pitcher for the Yankees.
Dickey and the Mets
My first goal was to establish where Dickey should fall in the Mets rotation. He currently pitches 3 rd and I believed he should be the Mets ace, the stats proved me correct. We will later see that Dickey should be the staff ace on several teams but the Mets are the team where he makes the clearest case. He was by any measure the best pitcher on the team in 2011 (he was also statistically the best pitcher on the team in 2010, a team which included Johan Santana and a good Mike Pelfrey, however his 2010 numbers are much closer to his teammates unlike 2011 where we will see he is much more productive than anyone else).
First lets exam R.A. in terms of traditional stats, he posted an ERA of 3.28 more than full run lower than his next closest teammate Jon Niese who had a 4.40. He had an ERA + of 113, the next closest Mets were Niese and Dillon Gee, tied for 84. He had only 8 wins but his other numbers clearly show this number was indicative of poor Mets relief and offense and not on Dickeys pitching ability. Now lets look at the system we are using in the report.
157.10
10.18
2.52
7.89
1.41
3.36
208.20
8.71
2.33
5.78
1.23
3.77
186.00
9.58
2.56
8.13
1.35
4.04
193.20
10.22
3.02
4.88
1.47
4.47
160.20
8.40
3.98
6.39
1.38
4.65

As you can see R.A. excels in almost all categories. The only category he does relatively poor in is K/9. This is to be expected to some degree as a knuckleballer will generally see less strikeouts and more outs from the fact that the knuckleball is hard to hit well and often results in poorly hit, easily fielded balls.
Below are the rankings:
1
2
1
4
1
2.
11
Ace
5
4
2
2
4
1
18

4
1
5
3
3
5
21

3
3
3
1
2
3
15

2
5
4
5
5
4
25


You can see here Dickey leads in three categories, comes in 2 nd in another and 4 th in one.  This gives him a sum of 11. Compared to other teams this is a pretty clear cut first spot case. A perfect score would be a 6, which only Clayton Kershaw and Justin Verlander achieved.
The current Met’s #1 starter is Mike Pelfrey, according to my rankings, he should be the 5th starter. Dickey surpasses Pelfrey in every category; even Dickeys weakest category across the board, K/9 is still better than Pelfs.  Dickey was much more effective in playing deep into games, Pelfrey started one more game the Dickey did but pitched 16 less innings.  Even after Pelfrey’s quality 2010 season he was still a lesser player than Dickey, if the Mets continue to play Pelfrey ahead of Dickey will show a complete lack of understanding of there pitchers abilities.
According to virtually any statistical analysis R.A. Dickey should be the Mets opening day starter next year.
Dickey around the league
After looking at the Mets the goal was to see where on each team Dickey would fall. The most common spot he fell in was # 2 , the overall results were:
1 st
6
2 nd
10
3 rd
6
4 th
7
5 th
1

The Breakdown of the teams were as follows:
1st
2nd
3rd
4th
5th
Mets
Yankees
Marlins
Rangers
Phillies
Nationals
Cardinals
A’s
Braves

Orioles
Astros
Dodgers
Brewers

Pirates
Bluejays
Padres
Diamondbacks

Rockies
Cubs
Rays
Giants

Royals
Indians
Reds
Mariners


Red Sox


Tigers


Twins


White Sox


It should be noted that some of these numbers are skewed by some aces being hurt for a large part of the season. For instance Josh Johnson is clearly the Marlins first starter but didn’t pitch enough in 2011 to be used in the comparison, therefore to be a true predictor for where in the 2012 starting rotations Dickey should fall he should be pushed to 4 th for the Marlins. The same goes for the Cardinals and Nationals, where the injured Adam Wainwright and Stephen Strasburg will be those team’s first starters.
As you can see Dickey statistically should be the first the first starter on six teams. However these are six poor performing clubs with a common theme of less than stellar starting pitching.
Even more interesting is that Dickey would be the 2 nd pitcher on ten teams including three play off teams and the World Series champion Cardinals. Also included are the Red Sox who were one of the best teams in baseball despite a September collapse. These numbers make it even more outrageous that 16 teams should feature Dickey as a #1 or 2 yet the Mets pitch him third.
The only team that Dickey was the 5 th starter on was the Phillies. This was his poorest performance on any team. Here he is tied for 5 th with Roy Oswalt. However the Phillies have by far the best rotation in baseball so it is to be expected that Dickey would be low on totem.
Conclusion
My original guess was that Dickey would be the starting pitcher on about 12 teams. By my analysis I was clearly wrong. This probably has more to do with unrealistic hopes than his actual performance. For a thirty six year old knuckle baller to be the ace on six teams and the second starter on ten teams is pretty impressive. For that matter for a thirty six year old knuckleballer to make the rotation of every team is a pretty good accomplishment. Ultimately I believe that with two full seasons of stats to examine we can comfortable say that Dickey is not a flash in the pan and will continue to be a useful and effective pitcher for several years to come.
More than the numbers though is what Dickey brings to the team. He was the guy in the clubhouse who tells the truth, who said that maybe the team needed to admit they were not unlucky but not very good. He is the guy who runs around the clubhouse wearing a ridiculous Jose Reyes t-shirt. He is rides a bicycle to spring training and in the winter is climbing Mount Kilimanjaro. He talks about Star Wars and named one of his bats Ocrist the Goblin Cleaver. He has a book coming out in the spring. He isn’t just a good pitcher, he’s fun to watch and fun to listen to. In a time when their isn’t much fun or success in the Mets, Dickey brings both.

12 comments:

  1. As a lifelong hater of the Mets, I unfortunately must admit that there are many valid points made in this report. Fortunately, it is in reference to one of the most likable members of the team, so I am okay with accepting the information laid out above. Have you considered doing the same report to a pitcher who is held in very high regard? Perhaps a Roy Halladay. That way you can see if the method holds up when used for someone who is by all accounts one of the top 3 starters in the game.

    Reply
  2. I have a feeling The Halladay report might end with him being the Ace on 30 teams but I will do it and find out.

    Reply
  3. One problem with your study is that in general AL pitchers stats are not as good as NL pitchers due to the DH. Looking at NL playoff teams, the only team Dickey would have been better than a 2nd starter was the Cardinals, and only because Wainwright was out for the entire year.

    Even though I love Dickey, he's still a 3rd starter on a quality NL team. That his stats show that he would be the ace of the Mets is more of an indictment on the Mets and their starters than how great Dickey pitches.

    Bottom line is that until the Mets get better starting pitching to go along with Dickey they won't be a NL playoff team.

    Reply
  4. I like this and I don't like it. I'm a huge Dickey fan as well and think he is one of, if the not, THE most underrated pitcher in all of baseball. Showing some statistical evidence of this is great. BUT, you have clearly not weighed pitching effectiveness in a fair manner. For example, FIP ITSELF is a measure of a few of the other categories you are weighing; so if you aer using FIP as one of the categories it has to be weighed more heavily than some of the others. K/9 is FAR MORE important than H/9 when judging a pitcher's overall effectiveness. Also, IP doesn't matter for pitching effectiveness at all. It certainly helps the team out and should be considered as a category when determining an ACE, but it shouldn't have equal weight with K/9, FIP, BB/9, WHIP, etc. And don't forget FIP itself is a measure of a pitcher's overall effectiveness and can be used without any other categories. PLUS, if you mean to use stats, you have to include some other important stats like HR/FLYball ratio, FIP from the stretch and WAR, etc.

    Reply
  5. @Prodigal

    Thanks for reading and for the feedback. All the points you make are fair. I considered a ton of different stats when writing the report but ultimately chose the ones I did because they were simple and reflected pure pitching ability. I do agree, that a weighting system would be ideal but again left it out for simplicity sake as I think an argument could be made for each of the categories weighing higher than the others.

    I do disagree that IP is not a measure of effectiveness. It shows both an ability to deal with fatigue, both mental and physical. And shows an ability to stay consistent over a longer period of time. You can have the greatest stats in the world but if you can never make it past the 4th inning you wont be a good major league starter (just ask Mike Pelfrey)

    Reply
  6. Good article. You left out the Angels. I assume Mr. Dickey would be number 4 or 5 on the Angels.

    Reply
  7. @Tom- Thanks for the comment, I did do an analysis of the Angels but must have left them off the chart. He was fourth in their rotation, 5th now with Wilson in.

    Reply
  8. So, we've lost our #2 (who was a complete dice roll incentive signing) and the opening day starter performed like a #5. Sounds about right. Ugh.

    Reply
  9. It's going to be a long year. In Dickeys defense he is only a #5 compared to the really, really good Angels rotation, I prefer to think of him as a Royals # 1.

    Reply
  10. The Orioles, Pirates, Rockies, and Royals are maybe the only other 4 teams I would not be happy to trade franchises with. Makes sense that our ace would be their ace.

    And consider that $24 million(?) went to a starter who logged zero innings.

    The only way I can look at this and not be negative is to think "It can't get worse, can it?". But this year may prove that statement wrong when your system proves Miguel Batista our 2012 ace.

    Reply
  11. The Pirates had a decent year going at one point, to be honest if given the option I might roll the dice and take any franchises lineup.

    As Mets fans we know it can always get worse. Getting rid of the last vestiges of Minaya contracts will be a nice start.

    Reply
  12. Ron, the royals have a stocked farm system, I would trade franchises with them...

    Reply