The Rockies Outfield Defensive Debacle - Running the Numbers

Going into the 2009 season, one of the biggest concerns for this team is looking to be the atrocious outfield defense.

I made a fanpost a couple of weeks ago about a computer model that took game-by-game defensive and offensive run contributions from each player and had the program simply maximize runs scored by the players. Limitations were put on players like Helton due to injuries cutting into his playing time.

I redid the program using Tom Tango's positional adjustments for defensive prowess, in an attempt to predict how each player on the Rockies would do at various positions in the field.

The thought is this: Everyone can play any position. They may not play said positions very well, but they can indeed play them. Given the time it would take to learn a position, Tango suggested that there is an adjustment that would be made to a fielder's performance with the glove in terms of runs over the course of a season. For example, a completely average 2B would probably be a completely average 3B. However, if that 2B moved to SS, he would probably play at about 5 runs/season below average. If he then moved over to RF, he would probably play at 10 runs/season above average.

Here is the spectrum:

Shortstop: +7.5 runs

Second Base: +2.5 runs

Third Base: +2.5 runs

Center Field: +2.5 runs

Left Field: -7.5 runs

Right Field: -7.5 runs

First Base: -12.5 runs

 

More explanation after the jump

Think about how the Nats/Cubs moved Soriano from 2B to LF. For his career, he is a -8.8 (runs/150 games) 2B, but he is +1.7 in LF. That means moving him to LF was a 10.5 run improvement in terms of defense. That lines up nicely with the spectrum.

Based on these adjustments, I altered the defense for all Rockies at every position and then had the program maximize the number of runs scored over 162 games. By doing this, the adjustments are saying something along the lines of, "Tulowitzki is worth X number of runs per game offensively, and worth Y number of runs per game at 1B, Z at 2B..." by adding the two numbers for every player at every position.

What makes this interesting is that it does not just say "TULO AT SS AND EVERYONE ELSE ADJUST", it looks to move everyone around as to optimize the benefit of good team defense and minimizes the damage from bad defenders. So, let's say that as a SS, Tulo is worth 10 runs, Spilborghs in CF is worth -1 run, and Barmes at 2B is worth 4 runs. However, using these adjustments, Barmes is worth 1 run at SS, Tulowitzki in CF is worth 15 runs, and Spilborghs in LF is worth 9 runs. By doing this, the team just improved their defense by 11 runs. The program made all these adjustments so as to minimize the number of runs the team will give up in a season.

Here is the best run-preventing defensive alignment that the program was able to come up with. Based on position where they make the majority of their starts:

C Iannetta

1B Hawpe/Smith/Stewart

2B Baker

3B Atkins

SS Barmes/Stewart

LF Spilborghs

CF Tulowitzki

RF Helton/Barmes

Now, take this with a grain of salt. I am not really proposing that the Rockies move Helton to RF and Tulo to CF. However, what really stands out to me is the simple fact that the outfielders are abysmal defenders. Spilborghs is the best of the 3 starters, and he is definitely below average in CF, which would be why the program moved him to LF. Hawpe is a joke in the field, and by virtue of that only ends up starting 75 games, all at 1B. The one problem with this model is that I could not put Seth Smith's numbers into it with any real confidence.

The initial reaction is might be "So why move all of our great IF talent to the OF?" Well, to answer that, do not think of it as IF or OF talent. Think of it as generic fielding talent and, rather than having a few good to great positions, it spreads the talent out a bit (via the positional adjustment spectrum) as to try and cut the losses that the current OF incurs.

The real story here though is how awful the OF defense is. Time and time again, we have discussed how terrible Hawpe is in the field. This program really suggests that his fielding is SO bad, that not even his above-average bat is worth enough keep him in the lineup for 162 games, even if he does improve (which we have all agreed will happen in 2009).

What I imagine happening is that you will see Gonzalez or Podsednik being a consistent defensive replacement for Hawpe in late innings. Spilborghs would move over to RF, and one of the two replacements would take over in CF.

We all knew that the Rockies' key to victory was keeping the ball on the ground. But we thought that was only because of home runs.

------------------

That's the end of the article itself, but I'd like to open this up to discussion now.

I think these are some key talking points:

1. Getting data from Fangraphs, is Brad Hawpe really as bad as UZR would suggest?

2. Are the position adjustments fair? Does CF play equal to 2B and 3B in Coors? What about LF and RF?

3. What players should I add to the simulation? Free agents? Prospects? You tell me.


X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Purple Row

You must be a member of Purple Row to participate.

We have our own Community Guidelines at Purple Row. You should read them.

Join Purple Row

You must be a member of Purple Row to participate.

We have our own Community Guidelines at Purple Row. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker