Last Sunday, I wrote about the flawed All-Star selection process. Before getting into today’s topic, I want to respond to a few comments from last week. First, I used Joe Crede as the example because Alex Rodriguez should have been a unanimous selection at 3B. For Crede to be elected because some players couldn’t see that Rodriguez is the best AL 3B is ridiculous. I’ll admit that Jason Varitek would have been a better example than Crede, but Rodriguez was a better example than Joe Mauer.

The second issue is with the start of All-Star balloting. I have no problems with voting starting at the beginning of May. The point brought up against it was that Carlos Quentin wasn’t on the ballot because he hadn’t stepped in as starter yet. Actually, that’s false. Quentin has been starting since April 3rd. Instead of listing Quentin on the ballot, Jerry Owens was listed, and he hasn’t even had a single plate appearance in 2008. This isn’t really an issue of when the voting starts; it’s an issue of “how did MLB end up listing Jerry Owens on the ballot without a single plate appearance?”

The last issue is that several of the comments were complaining about the fan vote with many of you suggesting that the players and managers have a better idea of who should be on the All-Star team. That is the main topic of this post.

Instead of looking directly at the players the fans voted in, I decided to look at the rosters as a whole and find the players that don’t belong objectively. When people choose All-Stars, there are two separate paths that are usually used. The first is the most common, and that is to select the All-Stars based completely on the first half of the season. The second is the one I generally use, and that is to pick the players that are most likely to put together the best season, which takes into account both their season-to-date stats and expectations on whether or not they’re likely to keep it up. For example, Ryan Ludwick and Ryan Braun are performing similarly, but Braun would get my vote well before Ludwick would. Back to the objective process, I wanted to incorporate both of those selection methods, so using the players’ WARP totals from 2005 to 2008 (minor league numbers included), I computed the average 2008 WARP and the average 2005-to-2008 WARP for the All-Stars. I have listed all of the players that have WARP totals worse than one standard deviation below average in either category.

Poor All-Star Selections by 2008 WARP: Joe Crede, Derek Jeter, David Ortiz, Carlos Quentin, Alfonso Soriano, Jason Varitek, Billy Wagner

Poor All-Star Selections by 2005-to-2008 WARP: Justin Duchscherer, Josh Hamilton, Tim Lincecum, Nate McLouth, Dioner Navarro, Joakim Soria, Edinson Volquez, Kerry Wood

Poor All-Selections by Both Methods: Cristian Guzman, Carlos Marmol, George Sherrill, Brian Wilson

Let’s look at how those players made the All-Star team, starting with those selected by the fans.

Poor Fan Selections: Josh Hamilton, Derek Jeter, David Ortiz, Alfonso Soriano (4 out of their 17 selections, or 23.5%)

Poor Player Selections: Joe Crede, Tim Lincecum, Nate McLouth, Carlos Quentin, Joakim Soria, Jason Varitek, Edinson Volquez, Brian Wilson, Kerry Wood (9 out of 33, or 27.3%)

Poor Manager Selections: Justin Duchscherer, Cristian Guzman, Carlos Marmol, Dioner Navarro, George Sherrill, Billy Wagner (6 out of 12, or 50%)

The data suggests that the managers are the ones screwing up. If we want to look at just the players that don’t are worse than one standard deviation below average in both categories, that’s one player selection and three manager selections. No matter how I look at the data, it doesn’t appear to me that it’s the fans screwing up.

Looking at it more subjectively, Varitek, Crede, Guzman, and Sherrill are the players that don’t belong on the All-Star team. As I’ve already stated, I think Varitek and Crede were selected because Joe Mauer and Alex Rodriguez were close to unanimous selections at their positions, which resulted in Varitek and Crede only needing a handful of player votes to make the team. This is a problem that needs fixed. Guzman is the Nationals’ rep, and looking at their roster, there’s not really anyone else to pick. I have no problem with requiring one player from each team even when it means that Cristian Guzman is a 2008 All-Star. Likewise, Sherrill is the Orioles’ rep, but Brian Roberts would have been a much better selection. If Roberts were selected instead of Sherrill, the 12-pitcher restriction would have knocked Dioner Navarro off the team. Roberts and CC Sabathia (an AL player at the time) are better choices than Sherrill and Navarro. In total, I find that to be two bad choices by the player vote process (not the players), one bad choice by necessity, and one bad choice by Terry Francona.

If two bad choices were in fact by the player vote process, how do we go about fixing it? As I suggested last Sunday, I think it can be fixed with one easy change: make the players vote for two players at every position. That should remove any chance of a player making the All-Star team because someone else (Joe Mauer or Alex Rodriguez) was a near-unanimous selection.

Oh, and we should have Senator Mitchell look into how Jerry Owens got listed on the All-Star ballot.