A look at some numbers

[shoeshineboy]

Most of us value OBP has a key stat indicating offensive production along with its combination with SLG (OPS). These are typically the best things to look at to show whether a player is getting something done that should ultimately produce runs. Players can't take care of what other guys in the lineup are/aren't doing. So if you see a guy like a Dukes with an ability to get on base, or a guy like Flores who seems to have quality at bats with an ability to hit the gaps with RISP, you want them positioned in the order properly. Thankfully, both Flores and Dukes have been moved to positions in the order that capitalize on their stats. Afterall, OBP and OPS are great predictors of run production, but you don't see the results of that unless other guys in the lineup produce or you give the guy a chance to play in the spot where he works best.

Many prefer to look at the end results of these indicators first. But runs scores and RBIs are not always in the control of the player. So I wouldn't rank a player with a much higher OPS behind another player simply because they didn't score or knock in as many runs. Nevertheless, it is nice to look at these numbers to see how they compare, at least on the same team.

The typical formula used to gauge run production is Runs Scores + Runs Batted In - Home Runs. This gives an overall run-production value for players. Looking at the Nats on the season, I ran the numbers on this. To level out the playing field since guys have been in and out, I then divided by the number of At Bats to get the runs produced per AB for each player.

The numbers that stand out are that Dukes and Flores have clearly been productive, especially over the last month. Willie Harris has actually been surprisingly productive. Of course you have to credit that to his role as a pinch runner, but that means he is being used properly. Pena of course, is pathetic. Ultimately, I look at this stuff to show not only who is actually being productive, but how well the manager is actually utilizing players, and whether certain players should simply be removed from the mix. It looks like the patience should be wearing this on several players. Pena sticks out like a sore thumb.

On the season here are the rankings:

Player                  Runs produced per at bat
J Flores                0.283
N Johnson   0.275
W Harris                0.256
W Nieves                0.254
R Belliard                0.242
E Dukes                0.238
R Zimmerman   0.209
C Guzman   0.206
A Boone                0.194
P Lo Duca   0.193
L Milledge   0.191
R Mackowiak   0.189
A Kearns                0.180
F Lopez                0.173
R Langerhans   0.167
K Casto                0.158
D Young                0.147
W Pena                0.096
J Estrada                0.077

Over the last 30 days

Player             Runs produced per AB
E Dukes           0.283
J Flores           0.282
R Zimmerman   0.273 (Only included 3 games)
R Belliard           0.219
W Harris           0.212
L Milledge        0.206
W Nieves           0.200
K Casto           0.200
A Boone           0.177
R Langerhans   0.167
K Casto           0.152
C Guzman        0.148
P Lo Duca        0.143
F Lopez           0.129
D Young           0.120
R Mackowiak    0.083
W Pena           0.077

[spidernat]

Harris' .256 would probably drop significantly if he had more at bats (same for Nieves).

[shoeshineboy]

Harris' .256 would probably drop significantly if he had more at bats (same for Nieves).

Likely so. Harris's numbers are skewed given the number of times he has likely scored as a pinch runner as well. It is rather compelling to see the consistency that Flores has maintained in both the overall and the last 30 days. I think this demonstrates that consistent at bats are really the key to production. That has been the problem that has appeared to plague this team, and the numbers would back it up.

[blue911]

Likely so. Harris's numbers are skewed given the number of times he has likely scored as a pinch runner as well.

But isn't being fast an asset? I think Bill James's Runs Created and RC27 are better indicators than runs produced.

[JMUalumni]

Very interesting analysis, but instead of dividing by at-bats do you think that it would produce a more accurate number if you used Plate Appearances, because ? (just a thought)

What do you call this stat? RPAB (Runs Produced per At-bat)?

[shoeshineboy]

But isn't being fast an asset? I think Bill James's Runs Created and RC27 are better indicators than runs produced.

Yes, speed is an asset (also just base-running ability in general), and that is why he is a valued player even though his batting numbers may not make people think so. This is the type of thing that gives me even more appreciation for a guy like Harris and what he can do as a role player.

I'll have to look at the James formula and see how it compares. I am not a big fan of the R+RBI-HR as a general way to judge players, but many people look at that as a bottomline stat since run production is the name of the game.

[blue911]

Yes, speed is an asset (also just base-running ability in general), and that is why he is a valued player even though his batting numbers may not make people think so. This is the type of thing that gives me even more appreciation for a guy like Harris and what he can do as a role player.

I'll have to look at the James formula and see how it compares. I am not a big fan of the R+RBI-HR as a general way to judge players, but many people look at that as a bottomline stat since run production is the name of the game.

I not too sure of the formula. Baseball-reference lists it with the player stats,so it's easy to find.

[shoeshineboy]

Very interesting analysis, but instead of dividing by at-bats do you think that it would produce a more accurate number if you used Plate Appearances, because ? (just a thought)

What do you call this stat? RPAB (Runs Produced per At-bat)?

I'll re-run it with plate appearances. I was just using at bat to even things out between players. I had also done it by game originally. I think the suggestion of using the James formula may be an interesting comparison.

[spidernat]

I think this demonstrates that consistent at bats are really the key to production. That has been the problem that has appeared to plague this team, and the numbers would back it up.

Too many outfielders splitting AB's.

[blue911]

Too many outfielders splitting with crappy AB's.

fixed

[PatsNats28]

Great writeup, very interesting

[PANatsFan]

Most of us value OBP has a key stat indicating offensive production along with its combination with SLG (OPS). These are typically the best things to look at to show whether a player is getting something done that should ultimately produce runs. Players can't take care of what other guys in the lineup are/aren't doing. So if you see a guy like a Dukes with an ability to get on base, or a guy like Flores who seems to have quality at bats with an ability to hit the gaps with RISP, you want them positioned in the order properly. Thankfully, both Flores and Dukes have been moved to positions in the order that capitalize on their stats. Afterall, OBP and OPS are great predictors of run production, but you don't see the results of that unless other guys in the lineup produce or you give the guy a chance to play in the spot where he works best.

Many prefer to look at the end results of these indicators first. But runs scores and RBIs are not always in the control of the player. So I wouldn't rank a player with a much higher OPS behind another player simply because they didn't score or knock in as many runs. Nevertheless, it is nice to look at these numbers to see how they compare, at least on the same team.

The typical formula used to gauge run production is Runs Scores + Runs Batted In - Home Runs. This gives an overall run-production value for players. Looking at the Nats on the season, I ran the numbers on this. To level out the playing field since guys have been in and out, I then divided by the number of At Bats to get the runs produced per AB for each player.

The numbers that stand out are that Dukes and Flores have clearly been productive, especially over the last month. Willie Harris has actually been surprisingly productive. Of course you have to credit that to his role as a pinch runner, but that means he is being used properly. Pena of course, is pathetic. Ultimately, I look at this stuff to show not only who is actually being productive, but how well the manager is actually utilizing players, and whether certain players should simply be removed from the mix. It looks like the patience should be wearing this on several players. Pena sticks out like a sore thumb.

On the season here are the rankings:

Player                  Runs produced per at bat
J Flores                0.283
N Johnson   0.275
W Harris                0.256
W Nieves                0.254
R Belliard                0.242
E Dukes                0.238
R Zimmerman   0.209
C Guzman   0.206
A Boone                0.194
P Lo Duca   0.193
L Milledge   0.191
R Mackowiak   0.189
A Kearns                0.180
F Lopez                0.173
R Langerhans   0.167
K Casto                0.158
D Young                0.147
W Pena                0.096
J Estrada                0.077

Over the last 30 days

Player             Runs produced per AB
E Dukes           0.283
J Flores           0.282
R Zimmerman   0.273 (Only included 3 games)
R Belliard           0.219
W Harris           0.212
L Milledge        0.206
W Nieves           0.200
K Casto           0.200
A Boone           0.177
R Langerhans   0.167
K Casto           0.152
C Guzman        0.148
P Lo Duca        0.143
F Lopez           0.129
D Young           0.120
R Mackowiak    0.083
W Pena           0.077

I wonder where our pitchers fall? Maybe above Mackowiak

[houston-nat]

I wonder where our pitchers fall? Maybe above Mackowiak

It should be remembered that Joel Hanrahan was a .277 hitter in AAA baseball, and his first major league at-bat was a triple.

[JMUalumni]

Shoeshine, you said you were going to look at Bill James' RC and RC27 for comparison.  I had a little extra time today and was really interested in what you were trying to do.  So I looked into and made a little chart for comparison.  When comparing the two, what you have came up with seems to be fairly consistent with the RC27 formula, but the two cases that stood out to me were Willie Harris and Will Nieves.  The formula you use seems to give both of those too much credit and that seems to be where the RC27 compensates the most for technical issues such as pinch running (Harris).  But the formula you are using seems to be a simpler and almost as accurate way of figuring out production.

Player          RC     RC27   RPAB
Jesus Flores   23   7.0   0.283
Nick Johnson   23   7.0   0.275
Elijah Dukes   20   5.5   0.238
Cristian Guzman   46   5.4   0.206
Ronnie Belliard   15   5.1   0.242
Dmitri Young   14   5.0   0.147
Aaron Boone   20   4.8   0.194
Will Nieves            9   4.3   0.254
Ryan Zimmerman25   4.2   0.209
Lastings Milledge   31   3.6   0.191
Felipe Lopez   26   3.6   0.173
Kory Casto           4      3.5   0.158
Ryan Langerhans   3   3.2   0.167
Willie Harris   8   2.9   0.256
Austin Kearns   12   2.5   0.180
Paul Lo Duca   4   2.1   0.193
Wily Mo Pena   10   2.1   0.096
Johnny Estrada   2   1.7   0.077

*I have a decent chart, but I cannot figure how to load it.  It is in PDF or Excel, but it says I can only load jpeg or gif.

[shoeshineboy]

Shoeshine, you said you were going to look at Bill James' RC and RC27 for comparison.  I had a little extra time today and was really interested in what you were trying to do.  So I looked into and made a little chart for comparison.  When comparing the two, what you have came up with seems to be fairly consistent with the RC27 formula, but the two cases that stood out to me were Willie Harris and Will Nieves.  The formula you use seems to give both of those too much credit and that seems to be where the RC27 compensates the most for technical issues such as pinch running (Harris).  But the formula you are using seems to be a simpler and almost as accurate way of figuring out production.

Cool, thanks for posting. I looked into it briefly before heading out earlier prior to the game. What I found was the RC/27 formula which was meant to help calculate the production of an entire lineup. I assume the RC27 is similar in that the incremental value for each player is measured. What I saw included stats for things like SB but also included CS, etc. It would make sense that this would help even out some of the spikes. It would also be interesting to see the Baseball Prospectus winshare and VORP rankings at this point as well.

[blue911]

RC27 is Runs Created per 27 at bats. Now if we could talk Tom into a VORP chart.

[JMUalumni]

Figured out how to do the chart finally, here it is just so it is easier to read:

Just as a comparison too, I looked at two of the better players in the league to see what their RC27 was:

Chipper Jones- 12.9
Lance Berkman - 12.6

[Minty Fresh]

What comes to mind when reading all of these charts is:

"WHY THE freak HAVEN'T WE LET GO OF LO DUCA AND ESTRADA?!?!?!??!?!?!?!?!?"

[JMUalumni]

Here is an update of the charts from earlier in the season, but this time I have also added in VORP.  For those who do not know what VORP is, here is a definition straight off of BP:

Value Over Replacement Player. The number of runs contributed beyond what a replacement-level player at the same position would contribute if given the same percentage of team plate appearances. VORP scores do not consider the quality of a player's defense.

I also included batting runs, which seems to be a better indicator of how many runs a player has ended up adding or costing the team throughout the season.  It is similar to Runs Created, except is adjusted to the different parks and leagues for each game AND the negative values allow us to see how many runs our players have cost us (striking out in key situations, Double Plays, caught stealing, etc.).  You will notice we barely have any players with a positive net value for Batting Runs, which basically coupled with pitching data indicates that the majority of our losses have been due to lack of production and things associated with scoring runs.

The biggest improvements since I last posted the chart have been in Belliard and Harris.  Dukes was able to almost triple the amount of runs he created during his June hot streak and Flores took the biggest drop because of the slump he has been going through.  Looking at the current VORP numbers, it is interesting to note how valuable Guzman actually is to the team with a rating of 24.8.  No other player, besides Dukes, even has a rating over 10 (around 15 I would consider key component of the team).  Players with a 10 VORP rating are somewhat important to the team, and as you can see we don't have many.  The average team has between 3-5 guys hovering at or above 15 VORP.  The most interesting of all the numbers is Austin Kearns, his numbers demonstrate how unimportant he actually is to the batting lineup (although defense is not considered), maybe this will turn around in the second half of the season.

As for Pena, it is pathetic that he kept getting a chance to play.  His VORP and Batting Runs indicate what we already know, he has been doing horrible this season.  I am sure that his VORPORP (value over replacement player over replacement player) would even be low, meaning that everyone on our bench would do a better job of producing runs, or at least not costing us so many.  These numbers, for Pena, have been stagnant for a few weeks now and it is just inexcusable that they allowed him to keep playing, much less start.

[JMUalumni]

I figured I might as well take a look at the production/value numbers in light of these recent releases.  Most of us here do not need the numbers to confirm what we already knew, that Lo Duca and Lopez (and Estrada) were 3 of the 5 least valuable and least productive players on the team.  Using the Batting Runs metric, we can see that Lopez probably cost us around 16-17 net runs this season from his striking out/double plays with RISP.  As far as his production over a period 27 PA, one of the lowest on the team and the absolute lowest of any regular on the team.  In the case of Felipe Lopez it actually should be an "addition by subtraction" (specifically when noting Lopez had the third most PAs on the team), unless Bonifacio performs even as poorly as Lopez did, which would seem impossible at the moment.  Lo Duca proved that contact isn't worth much unless you have some type of speed or ability to direct where you hit the ball.  His OPS staggered in the sub .600 area for most of the season (finishing at .598). 

Of the bottom five performers on the team metrically, only one of them remains on the 25-man roster, that man is our "everyday right fielder" Austin Kearns.  Mostly, Kearns has failed to capitalize on key opportunities and has wasted away most of the chances he has gotten to score or move runners.  He has a dismal .623 OPS (even worse that he is an OFer) and has costed us about 12 net runs over the season.  Unless the burners suddenly kick into booster mode, Kearns is going to go down as a huge blunder.  Some other observations:

Milledge is absolutely killing us.  As a center fielder, he should be one of our most productive bats, but he has actually probably hurt us the most, next to only Kearns.  Even for a young kid, his numbers are troubling (.651 OPS, 3.2 RC27, -13 BtRuns).  Why have a hitting coach that has never coached before try to teach guys that have barely played in the Major Leagues is beyond me.  At least get an experienced guy to help kids like Milledge find his stride.  His RC27 numbers are bit skewed because of the bubble of production he had earlier in the season, but his BtRuns numbers show that he has pissed away way to many opportunities.  Since this season is already in the gutter, fine, let him finish the season.  If he shows no improvement and starts much the same next season, then it might be time to let him ride the pine for awhile, or even send him down to AAA to "find his stroke."

The good news is that we will have one of our statistically most productive guys back in the lineup with Elijah Dukes' return.  Flores has steadily improved his numbers over the last two weeks, finally breaking out of that unbearable slump he fell into.  Belliard has great numbers.... for a utility player and should be good coming out of that role for the rest of the season.  Zimmerman has had a below average season, but since defensive numbers aren't calculated into value stats, he is at least a little more valuable than the numbers would say because of the amount of outs he probably contributes.

Man, it will feel good to finally see Lopez and Lo Duca's names finally removed from this chart.....   

[BBQ]

Interesting. No surprise to see Willy Mo sitting at the bottom.

[JMUalumni]

Well, the month of August has come and gone.  Looking at the metrics, a few things become obvious.  Part of the winning streak can be attributed to the subtraction of the most harmful players on the squad.  Lo Duca, Kearns, Lopez, and Pena were amongst the worst producers on the team AND were in the starting lineup almost everyday.  Lo Duca and Lopez were cut, Pena/Kearns are now on the DL.

Another reason for the sudden increase in winning is the increased production of players like Milledge and Zimmerman.  Both were grossly underproducing for most of the season, but have seen spikes in their numbers during the last month.  Milledge may have made the most miraculous move (.651 OPS --> .734 OPS/ 3.2 RC27 --> 4.3 RC27) and I am left wondering exactly how his numbers would look if he were able to play like he has been all year.  Zimmerman has also turned on the burners and has been constantly getting hits, his numbers are now creeping closer to his career averages, with the possibility that he may surpass some metrics in September.  Belliard's lights-out August has also allowed for him to drastically improve his numbers.

Looking at the VORP numbers, whereas we only had three players above a 10.0 VORP rating (what you would expect from a solid member of a winning team), we now have 5 players.  Three of these players (Dukes, Guzman, Milledge) are all part of this team's future plans.  The other two (Harris, Belliard) may be sticking around for next season as well.  Ryan Langerhans has also proven that he can be a solid replacement player and, much to my surprise, a very good option off the bench.  Guzman's numbers are at a season high and it appears that hitting from the #2 spot may be the best option for him (gets on base a lot, has a chance to knock in some runs).

August was a cruel month for Jesus Flores, as he has seen a dramatic drop in all of his metrics.  He was one of the most valuable players for this team until this month, now is barely a top 10 player (performance-wise) on this team.  Flores witnessed his OPS drop nearly 100 points during August and will need to break out his slump if he wants to end the season with some respectable numbers.

[blue911]

Chico wrote that WMP has been moved to the 60-day. Maybe Montz will get a call up.