Outs per Runs Produced.

[willshad]

The Garvey/Cey thread has me thinking; how many extra outs is it worth sacrificing to get an extra run? I was thinking that , if an average team scores 4.5 runs a game, and makes 27 outs per game, then they average about 6 outs per run. But, then I looked at one extreme example...Jimmy Rollins in 2009. he made 526 outs (second most ever), and produced 156 runs. Would that mean he actually had better than average production? That comes out to about 3.5 outs per run produced. I know for a fact that Rollins flat out sucked that year, that's why I find it odd. I'm pretty sure that something is wrong with my logic or numbers, but I can't figure out what it could be. I know the average player cannot possibly average 6 outs per run produced, yet when you divide 27 by 4.5 you get 6!

[filihok]

The Garvey/Cey thread has me thinking; how many extra outs is it worth sacrificing to get an extra run?

That's dependent upon the score of the game. And the answer is nearly always "it's not a good idea to sacrifice an out to get a run".

I was thinking that , if an average team scores 4.5 runs a game, and makes 27 outs per game, then they average about 6 outs per run. But, then I looked at one extreme example...Jimmy Rollins in 2009. he made 526 outs (second most ever), and produced 156 runs. Would that mean he actually had better than average production? That comes out to about 3.5 outs per run produced. I know for a fact that Rollins flat out sucked that year, that's why I find it odd. I'm pretty sure that something is wrong with my logic or numbers, but I can't figure out what it could be. I know the average player cannot possibly average 6 outs per run produced, yet when you divide 27 by 4.5 you get 6!

sigh

1 team run scored usually produces 1 run scored and 1 RBI.

[leewileyfan]

It's not just the single player's outs that produce 4.5 runs. It's the entire team that produces [generates] the total out investment in the creation of team runs per game.

Ideally modeled, it's 27 outs that it takes to create 4.5 runs. [4.5/27 = .167]. However, that's only the outs portion. If we had only outs, run production would be: 0.

We have to introduce the whole set of game incidents, H, BB, E, HBP, Sac, etc. into the mix to get individual player run creation equations.

Yes, outs can generate runs in certain circumstances; but they are not the norm.

A player who creates 88 runs in 600 PA and with a .280 BA can be looked at several ways:

- runs created/PA;

-runs created/AB;

-runs created in any particular base-out situation ...

... but runs created BY outs made needs tweezing one's way through play-by-play to sort out "outs" that made/allowed runs to score.

Your example of Jimmy Rollins [2009] is correct. He had RC in 725 PA = .1200 RC/PA. League average was 11,481/99,541 = .1153; so in 725 PA, Rollins was just + 3.38 runs above League average.

It was all those outs that kept him so low.

[willshad]

It's not just the single player's outs that produce 4.5 runs. It's the entire team that produces [generates] the total out investment in the creation of team runs per game.

Ideally modeled, it's 27 outs that it takes to create 4.5 runs. [4.5/27 = .167]. However, that's only the outs portion. If we had only outs, run production would be: 0.

We have to introduce the whole set of game incidents, H, BB, E, HBP, Sac, etc. into the mix to get individual player run creation equations.

Yes, outs can generate runs in certain circumstances; but they are not the norm.

A player who creates 88 runs in 600 PA and with a .280 BA can be looked at several ways:

- runs created/PA;

-runs created/AB;

-runs created in any particular base-out situation ...

... but runs created BY outs made needs tweezing one's way through play-by-play to sort out "outs" that made/allowed runs to score.

I am not really looking at it as 'runs produced BY outs', but rather, how MANY outs by an individual player makes during the course of a season it would be worth sacrificing in order to get a single run home. Obviously, it is worth one out to get a run home, in any circumstance, since runs are much more scarce. Yet, how come when a player produces more runs, and yet also makes more outs than another player, then it is seen as bad?

[willshad]

That's dependent upon the score of the game. And the answer is nearly always "it's not a good idea to sacrifice an out to get a run".



sigh

1 team run scored usually produces 1 run scored and 1 RBI.

So, are you saying i can basically cut my number in half?It's worth 3 outs to get a run?

[filihok]

So, are you saying i can basically cut my number in half?It's worth 3 outs to get a run?

Am I saying that?
What do you think?

What makes sense?

You also took the home runs out of Rollins' numbers...

[Second Base Coach]

We are coming back (slowly) to Base-Out Percentage. Tom Boswell looks better and better all the time.

[filihok]

http://www.tangotiger.net/wiki/index.php?title=Base-Out_Percentage

BOP = (TB + W + HB + SB + SH + SF)/(AB - H + CS + SH + SF + GIDP)

[Jackaroo Dave]

p
You also took the home runs out of Rollins' numbers...

That's Runs Produced: R + RBI - HR. Because a homer is counted twice: as a run and an RBI. (I had a conversation about this with toomanyhats on one of the Garvey threads the other day, about why you shouldn't do it, because EVERY play with an RBI double counts.)

By the way, filihok, Tom Tiger has voiced a preference for this over R + RBI. I never followed it up. If you are interested . . .

[drstrangelove]

The Garvey/Cey thread has me thinking; how many extra outs is it worth sacrificing to get an extra run? I was thinking that , if an average team scores 4.5 runs a game, and makes 27 outs per game, then they average about 6 outs per run. But, then I looked at one extreme example...Jimmy Rollins in 2009. he made 526 outs (second most ever), and produced 156 runs. Would that mean he actually had better than average production? That comes out to about 3.5 outs per run produced. I know for a fact that Rollins flat out sucked that year, that's why I find it odd. I'm pretty sure that something is wrong with my logic or numbers, but I can't figure out what it could be. I know the average player cannot possibly average 6 outs per run produced, yet when you divide 27 by 4.5 you get 6!

Runs produced +R + RBI -HR (which is where the 156 runs comes from) is not a very good stat since a large chunk of it depends on opportunity (to have runners on ahead of you or have good hitters behind you.) It does not reflect your real contribution. One 'similar sounding' and better (and more meaningful) stat is RC (runs created.) In 2009 he had 87 RC with 526 outs. His team made 4357 outs and had an RC of 890. (The team actually scored only 820, meaning that they were less efficient, likely in base running, than average. E.g., their XBT was 39%, compared to Colorado 47%, STL 44%, LAD 44%, etc.)

Rollins was 526/87=6.05
Team was 4357/890=4.90

Meaning he was below average for his team, and pretty much below average for the league. This dove tails with his oWAR (was 1.6) and his OPS+ (was 87.)

As you can see, Philly had an RC of 4.9 outs per run, so that makes sense as they were a good offensive team (tops in the league in runs scored in 2009.)

RC isn't adjusted for league or park effects, so it can't exactly be used 'raw', but it is a good starting point. There are certainly better tools than RC, but that is the one I had more handy and which is easier to show.

[JDanger]

Runs Produced: R + RBI - HR.

So we've taken two pointless stats and combined them together to create something even more pointless?

[willshad]

So we've taken two pointless stats and combined them together to create something even more pointless?

Why are they pointless? because you don't like them? I happen to LOVE the stat Runs produced per plate appearance. It gives the player credit for his walks, but does not over credit them, like most stats do. I also like to deal with REAL runs , and not hypothetical runs, such as the ones with 'runs created'.

[willshad]

Runs produced +R + RBI -HR (which is where the 156 runs comes from) is not a very good stat since a large chunk of it depends on opportunity (to have runners on ahead of you or have good hitters behind you.) It does not reflect your real contribution. One 'similar sounding' and better (and more meaningful) stat is RC (runs created.) In 2009 he had 87 RC with 526 outs. His team made 4357 outs and had an RC of 890. (The team actually scored only 820, meaning that they were less efficient, likely in base running, than average. E.g., their XBT was 39%, compared to Colorado 47%, STL 44%, LAD 44%, etc.)

Rollins was 526/87=6.05
Team was 4357/890=4.90

Meaning he was below average for his team, and pretty much below average for the league. This dove tails with his oWAR (was 1.6) and his OPS+ (was 87.)

As you can see, Philly had an RC of 4.9 outs per run, so that makes sense as they were a good offensive team (tops in the league in runs scored in 2009.)

RC isn't adjusted for league or park effects, so it can't exactly be used 'raw', but it is a good starting point. There are certainly better tools than RC, but that is the one I had more handy and which is easier to show.

What I am having trouble understanding is, since obviously runs are more scarce than outs (a good team will produce 5 outs per run, as you have showed), then why is it ALWAYS seen as a bad thing when a guy produces more runs than another guy, and yet makes more outs. For instance, if Jim Rice is responsible for 40 more runs than Roy White, then would it not make sense that he would have to make about 200 more outs than White in order for him to be less productive?

[drstrangelove]

What I am having trouble understanding is, since obviously runs are more scarce than outs (a good team will produce 5 outs per run, as you have showed), then why is it ALWAYS seen as a bad thing when a guy produces more runs than another guy, and yet makes more outs. For instance, if Jim Rice is responsible for 40 more runs than Roy White, then would it not make sense that he would have to make about 200 more outs than White in order for him to be less productive?

Well, I don't think it's "ALWAYS seen as a bad thing when a guy produces more runs than another guy, and yet makes more outs", so it can't be always. I know that a lot of people are better at this than me, and I'm sure that some people say that, but I'm also sure that many other people would know that it's wrong.

Also, you can't simply look at RC without adjusting for context. This is common when you see people saying that someone playing in the 1990's in Colorado is far superior to someone playing in the 60's in the Astrodome. RC without adjusting for contect could be a problem.

But again, I haven't seen anyone say something like "using 20% more outs to make 30% more RC is bad." Ignoring context for the moment, it certainly is the opposite of bad. Your point is well taken: it makes no sense to say that and people who blindly say that are not understanding what it means (or what they are saying.)

[filihok]

Why are they pointless? because you don't like them? I happen to LOVE the stat Runs produced per plate appearance. It gives the player credit for his walks, but does not over credit them, like most stats do. I also like to deal with REAL runs , and not hypothetical runs, such as the ones with 'runs created'.

Well...the problem with 'real runs' is it only deals with the final play of a run scoring sequence, when the run scores. It completely discounts how the player reached base, how he got to 2nd base and how he got to 3rd base. And it doesn't really consider how the player got to home plate. An RBI single, and RBI ground out are both equal according to Runs Produced, but are certainly not equal in real life.