Philliesflow.com

Remember this? Earlier this month I pointed out that the 2010 Phillies called on their relievers to throw fewer innings than any NL team had over the past five seasons. That’s not the only remarkable achievement of the pitching staff, however.

In 2010, Phillies pitchers combined to walk 416 batters on the season. That was the not only the fewest number of walks issued by a National League team in 2010, but the fewest number of walks issued by an NL team since 1995.

The table below shows the NL team that walked the fewest batters in each of the past 16 seasons:

In 1995, the Expos also issued 416 walks, finishing second in the league in the category.

In 2010, it wasn’t close. The Phillies walked 416 batters for the year. The team that issued the second-fewest number of walks overall was the Cardinals — they walked 477.

Charlie Manuel continues to talk as if John Mayberry has a real chance to get significant time in right field this season. Ross Gload notably does not appear on his list of potential right fielders.

Last week’s post pointed out that both the bullpen and the starters for the 2010 Phils posted a winning percentage that was .109 higher than the average winning percentage for a National League pen and rotation. The thing that seems odd about that, of course, is that the starters were great in 2010 and the bullpen was mediocre. Shouldn’t the winning percentage for the starters have been a lot better than the winning percentage for the pen, given that they shared the same excellent offense?

Whether we think the winning percentage for the relievers should have been worse or not, it wasn’t. The pen had a record of 27-17, a .614 winning percentage. The Phils had a better winning percentage in games where the pen got the decision than they did overall (97-65 overall, for a .599 winning percentage) or in the games where their starter got the decision (.593).

Given how small the number of decisions the bullpen got in 2010, a fluke seems like the best guess as to why the winning percentage was so good. The .614 winning percentage for the bullpen was third-best in the NL last year. By runs allowed per nine innings, the Phils were the eighth-best team at preventing runs. Was the difference in the ranking of the win percentage and the ranks allowed per nine innings unusual? Not really.

The table below shows the rank for runs allowed per nine innings pitched for the 16 NL teams this year, the team’s rank in that category, the bullpen winning percentage rank and the difference between the two:

So the Phils had the eighth-best bullpen in the league by runs allowed per nine innings pitched, but the third best winning percentage. That’s a difference of five. There were three NL teams that had a bigger difference between the rankings of their runs allowed per nine innings and winning percentages. The Nats and Cardinals both had good pens with bad winning percentages. The Pirates had a miserable pen at preventing runs, but saw their relievers combine to post a 23-21 record (it wasn’t really enough to make up for the hide-your-eyes 34-84 put up by their starting pitching).

Here’s the table for the starters:

By runs allowed per nine innings pitched, the Phillies had the second-best rotation in 2010. The 70-48 mark put up by their starters gave them a league-best .593 winning percentage for the rotation. The teams that showed the biggest differences between the ranks of their winning percentage and runs allowed per nine innings pitched were the Giants, Reds and Rockies. The Giants had fantastic pitching, the best in the league, and an unimpressive offense that finished ninth in the NL in runs scored. The Reds and the Rockies are the other two teams that had the biggest differences between the rank of their runs allowed per nine innings pitched for their starters and the winning percentage for their starters. For each of those teams, the rotation’s winning percentage was much better than their rank for runs allowed per nine innings. The Reds and Colorado finished first and the third in the league in runs scored — Cincinnati led the league in runs scored and the Rockies finished third, just behind the Phils.

Here’s how the winning percentage for starters and relievers compare to the overall winning percentage for the team for each of the NL teams last season:

Using the Phillies as the example, the way to interpret the table above is that the Phils overall played to a .599 winning percentage last year, the winning percentage for their relievers was .614 and the winning percentage for their starters was .593. The bullpen winning percentage was .015 higher than the winning percentage overall for the team and the winning percentage for the starting pitchers was .006 lower than the overall winning percentage for the team.

The table shows that the Pirates were the team whose bullpen winning percentage was better than the overall winning percentage for their team by the largest margin (followed not so closely by the Giants). At the other end of the scale, the Cubs had the bullpen winning percentage that was worse than their overall team winning percentage by the most.

It was the also the Cubs who had the biggest positive gap between the winning percentage for their starters and the team overall. As was mentioned above, the Pirates starting pitchers threw to a miserable record. Pittsburgh had the biggest negative gap between the winning percentage for their starting pitchers and their team overall.

Back to the Phillies, though. The point about the Phillies is that both the difference between their winning percentage for the starting pitchers and the winning percentage for the bullpen are very small. If you take the absolute value of each of the differences, only three teams, the Mets, Brewers and Reds, had a smaller difference between the winning percentage for the bullpen and the winning percentage for the team. Only two teams, the Mets and the Reds, had a smaller difference between the winning percentage for the rotation and the winning percentage for the team. If you combine the absolute values of both, there are three teams with a smaller combined difference (Mets, Reds, Brewers). So while it is a bit surprising that the Phillies got such an impressive winning percentage out of a so-so bullpen, it may be even more remarkable that the both of their winning percentages were so similar to the team’s winning percentage overall.

In this Q&A from the Phillies web site, Todd Zolecki suggests that it might make sense for the Phils to keep Blanton in case Oswalt is not back in 2012. Blanton will make $8.5 million in 2011 and $8.5 million in 2012.

Manuel talks about the possibility that John Mayberry would start the year with the team here. I’m enthusiastic about the runs like a deer part, but less so about the .330 career on-base percentage in the minor leagues. Mayberry turned 27 in December.

Domonic Brown was fourth on the recent list of the top 50 prospects by MLB.com. He also hits left-handed, unlike Ben Francisco, and can play the outfield, unlike Ross Gload. Also runs like a deer and without the .330 career on-based percentage in the minors (.373, for the record). My concern about what the Phillies are doing in right field is that everything seems to be based on the idea of finding a stopgap in the short term to get to Brown shortly after the season starts. If Brown is ready to play regularly soon after the season starts, I don’t think it matters a whole ton what the combination of Francisco, Mayberry and Gload man right until he does. What if he’s not, though? And how much difference is another couple of weeks or months in the minors going to make? A whole season of Francisco, Gload and Mayberry in right would combine to put up some ugly numbers. If Gload has to play defense regularly in right, they would put up some real ugly defensive numbers as well.

Here’s the bullpen record by starting pitcher for the Phils for 2010:

The bullpen had a total of three decisions in the 52 games started by Halladay or Moyer, but 13 in the 18 games started by Blanton.

Halladay pitched well enough to win in the games he didn’t get a decision, too. On July 10, Halladay threw nine shutout innings against the Reds, but the Phillies didn’t score until the eleventh when a double by Ruiz and a walkoff single by Rollins gave them a 1-0 win. Halladay wasn’t as dominant in his other no-decision in 2010. On May 12, he allowed three runs over 6 1/3 innings against Colorado. That game was also decided in extra innings — this time when Miguel Olivo homered off of Chad Durbin in the bottom of the tenth to give the Rockies a 4-3 win (it was Olivo’s fifth hit of the game).

Moyer had a decision in each of his first 18 starts on the season. His 19th start came on July 20 and he went just one scoreless inning before leaving the game with a strained elbow. Carpenter and Baez combined to allow six runs over the next four innings and the Phils lost the game 7-1.

Overall for the season in the NL in 2010, the teams that weren’t the Phillies saw their starter get a decision in 70.5% of their starts. As you know, the Phillies had the best winning percentage in the NL last year. Here’s how the difference in winning percentage breaks down between the starting pitchers and the bullpen:

The Phillies starting pitchers went 70-48 for a .593 winning percentage. The starting pitchers for the other teams in the league that weren’t the Phillies went 829-884, a .484 winning percentage. So the winning percentage for the starters was .109 higher than the winning percentage for the teams in the league other than the Phils. The winning percentage for the bullpen was also .109 higher. The pen went 27-17 for a .614 winning percentage. The bullpens of every team in the league other than the Phils combined to go 362-355, a .505 winning percentage. .614 minus .505 is .109.

Roy Halladay, Roy Oswalt, Cliff Lee and Cole Hamels have combined to make 206 starts for the Phillies over the past five years. Here’s the team record in starts by those pitchers over the past five seasons:

The 206 starts haven’t been distributed at all evenly among the four. Hamels has made 149 of them, Halladay 33 and Oswalt and Lee have each made 12.

The Phils also haven’t had a lot of success with Hamels on the mound in the past couple of years.

In each of the last two years, the Phillies have had a better winning percentage in games that Hamels didn’t start than in games he did. Over the past three years combined, the Phillies are 53-45 (.541) when Hamels starts and 229-159 (.590) when he doesn’t. They also have a better winning percentage in games started by Kendrick (36-27, .571) over the past three years than in the games started by Hamels.

The mediocre record in games started by Hamels has a lot to do with them going 16-16 in the games Hamels started in 2009 when he didn’t pitch very well, but also with the offensive support that Hamels has gotten from the team. In 2008, the Phillies scored 4.61 runs per game in the games he started and 5.02 in the games he didn’t. In 2009 it was 4.69 in his starts and 5.15 in his non-starts. Last year they scored just 3.76 runs per game when he started and 5.02 runs per game when he didn’t.

Here’s what the Phillies have done in games started and not started by Halladay, Oswalt, Hamels and Lee over the last five years.

In 2009, the Phillies had a better record in the games that Hamels and Lee didn’t start (69-49) than the games that they did (24-20). In the other four seasons, the Phils fared better in the games started by Halladay, Oswalt, Lee or Hamels.

If the concerning news is that the Phillies are just 53-45 when Hamels starts in the last three years (85-64 over the last five), the good news is that they are a rather amazing 40-17 (.702) when Halladay, Oswalt or Lee start. Despite the 40-17, though, the Phils are just 74-48 (.607) when Halladay, Oswalt, Lee or Hamels start over the last two seasons, thanks to the 34-31 put up by Hamels.

So how many starts are Halladay, Oswalt, Lee and Hamels going to combine to make in 2011? A lot, we all hope, but over the past five seasons there have only been two years in which all four of those guys each made 30 starts, 2008 and 2009.

Halladay and Oswalt have made at least 30 starts in each of the five seasons.

Hamels has made as many starts (98) as Halladay over the past three years and more than Oswalt (Oswalt has made 94 starts over the past three seasons), but made under 30 starts in each of his first two years in the league (2006 and 2007). Since his age 24 season in 2008, Hamels had made at least 32 starts every year.

Lee has also made 30 starts in just three of the last five years. In 2007, he strained his groin during Spring Training and didn’t pitch till May. At the end of July he was sent to the minors with a 6.38 ERA for the season. When he returned to the Indians in September, he pitched out of the bullpen. Last year he made just 28 starts as an abdominal strain kept him from making his first start of the year until April 30.

Here’s the number of starts each of the four pitchers has made in the past five seasons:

So, over the past five years, the group has made an average of 124 starts a season. 2007 was the year in which they made the fewest starts (107) and 2009 the year of the five in which they made the most (138).

Over the last three years, the group has averaged 131 starts.

All four members of the group have started at least 28 games in each of the past three seasons. Lee started 28 games in 2010 after missing part of the early season with his abdominal muscle. In 2009, Oswalt was struggling with problems with his lower back at the end of the season for an Astros team that wound up going 74-88 and finished 17 games out. After throwing to a 5.81 ERA over his last five starts, Oswalt didn’t make a start after September 15 and wound up with just 30 for the year. The last time that Oswalt made less than 30 starts in a season was 2003.

The Phillies traded Sergio Escalona to the Astros for 23-year-old second baseman Albert Cartwright. The lefty Escalona made 14 appearances for the Phillies in relief in 2009, throwing to a 4.61 ERA and a 1.24 ratio. He was designated for assignment on Saturday to make room for Romero on the 40-man roster. Cartwright saw his first action at Double-A in2010, hitting 229/289/271 over 154 plate appearances.

Quick — who allowed fewer runs per start in 2010, Cole Hamels or Roy Halladay?

I’m guessing Halladay is your answer. Or at least it would have been mine before I looked it up. But it’s a trick question. Both Halladay and Hamels made 33 starts for the Phils last season and each of them allowed 74 runs, which is 2.24 runs per game.

They didn’t have the same ERA, of course. It wasn’t close. Hamels 3.06 and Halladay 2.44. Hamels had a higher percentage of the runs he allowed go as earned runs (if all of the runs that both pitchers allowed were earned, Hamels would have thrown to a 3.19 ERA and Halladay a 2.65 ERA) but that’s not the biggest factor in accounting for the difference. As you surely know, the reason that Halladay’s ERA was so much better and Halladay’s year was so much better was that Halladay pitched way more innings. Halladay threw 250 2/3 innings in his 33 starts compared to 208 2/3 innings for Hamels in his 33 starts.

The 208 2/3 innings that Hamels threw is still a lot — only 13 pitches in the NL threw more than that in 2010. But it’s not 250 2/3. Nobody in the NL threw more than that in 2010. Nobody was close. Chris Carpenter finished second in the NL in innings pitched behind Halladay with 235.

There were 39 NL pitchers that made at least 30 starts in 2010. Of those 30, four of them (Halladay, Hamels, Oswalt and Kendrick) pitched at least part of the year for the Phillies. Of those four, Halladay and Oswalt threw an unusual high number of innings per start, Kendrick threw an unusually low number of innings per start and Hamels was in the middle.

Again, among the NL pitchers making at least 30 starts, Halladay and Oswalt had unusually high numbers in terms of innings per start, Hamels was in the middle of the pack and Kendrick was near the bottom.

In 2011, the Phils will presumably go into the season with Halladay, Oswalt, Hamels and Lee as their first four starters. How many innings might that save their bullpen is that group stays in the rotation for a full season?

Last year the Phillies threw a total of 1,456 1/3 innings. The starters combined to throw 1,035 1/3 innings and the pen threw 421 innings. The starters on the 2010 Phillies that weren’t Halladay, Oswalt or Hamels threw an average of 5.88 innings per game.

The table below looks at how many innings the rotation and pen might throw in 2011 if

1. The Phillies overall (starters and relievers combined) threw the same 1,456 1/3 innings in 2011 as they did in 2010.

2. Halladay, Hamels and Oswalt all threw the same number of innings per start in 2011 as they did in 2010.
3. Lee threw the same 6.44 innings per start in 2011 as he has over his career (206 of his 218 career starts have come in the AL — in his 12 career NL starts he has thrown 6.64 innings per start).
4. The Phillies starting pitchers who aren’t Halladay, Hamels, Oswalt and Lee will throw the same 5.88 innings per start in 2011 that they did in 2010.

So, for example, the table above suggests that if each of Halladay, Hamels, Oswalt and Lee made 32 starts in 2011 at the rates defined above, they would combine to throw 863 innings over those 128 starts. The other 34 starts would be made by other pitchers, who would throw 200 innings. That would give the rotation 1,063 total innings pitched and would leave about 393 1/3 innings to be pitched by the bullpen if the staff overall was going to throw the same 1,456 1/3 innings in 2011 they threw in 2010.

Last year, the Phillies starters overall threw 6.39 innings per start. That means Hamels was below the team average at 6.32 and Lee, had he actually made starts for the Phils in 2010 and thrown 6.44 innings per start, would have been very close to the team average. While Lee may have thrown 6.44 innings per start over his career, his numbers over the past three years are way up from that. Over the past three seasons, Lee has made 93 starts in which he has thrown about 7.18 innings per start. In 2010, he made 13 starts for Seattle and threw 103 2/3 innings on a team that would wind up going 61-101 . That’s a ridiculous 7.97 innings per start. It sounds like the kind of thing that might not even be that good for you.

The difference between 7.18 innings per start and 6.44 innings per start is a lot of innings (a little more than 22 over 30 starts). Even the difference between Lee’s 6.64 innings per start (what he threw in his 12 starts with the Phillies in ’09) and his career mark of 6.44 innings per start adds up. If you replace the 6.44 innings per start for Lee with 6.64 , the TOTAL IP BY PEN numbers at the bottom of the table would switch from 417.5, 400.3, 393.4, 383 to 412.5, 394.3, 387 and 376.

Also important to consider is that while it’s true that the non-Halladay/Hamels/Oswalt starters for the Phils in 2010 combined to throw 5.88 innings per start in 2010, they could easily throw fewer than that in 2011. Blanton was the guy outside of the big three who made the most starts and he went pretty deep into games, throwing 174 2/3 innings over 28 starts or 6.24 innings per game. That inflates the number for the group. By comparison, the group of Kendrick, Happ, Figueroa and Worley combined to make 37 starts in which they threw an average of 5.62 innings per game.

The deal with Dennys Reyes fell through and this article says that JC Romero is still hoping to come back to the Phillies.