With all the stats in baseball, I've never seen one like this. Maybe it exists, but I've never seen it and I don't know what it's called. And if it doesn't exist, someone should name it and start tracking it. I want to see a starter's ERA vis-a-vis his opponents. It would be sort of like a +/- in hockey. For example, Pitcher A allows an average of .3 runs per start more than the guys he faces. (That's probably Jose Quintana.)

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

You must have way too much time on your hands. Or else terrible sleep patterns. Get some rest.

Doesnt exist...it is Unpossible!

_________________
"That's what the internet is for. Slandering others anonymously." Banky
“Been that way since one monkey looked at the sun and told the other monkey ‘He said for you to give me your fuckin’ share.’”

An interesting idea but what ERA do you use?

Opening day both guys have an ERA of 0.00 so there is no difference.

Start two after their first start suppose Sale went 6 gave up 1 earned his opponent just say Kershaw went 5 and gave up 5
Sale has an ERA of 1.50 and Kershaw is at 9.00 .

Mid season Sale and his 3.04 gets matched against a just rehabbed Verlander who is at 0.00

so for these three starts would be

game 1 = 0
game 2 = -7.50
game 3 = + 3.04

Would his season +/- be -4.46 ?
Do you use the average of -1.487 ?

What would that number show you?

_________________

To make it really simple, if Jose Quintana went nine innings against Bruce Chen and lost 1-0, Quintana would be +1 (or -1 if you wanted to do it that way) in RERA (Relative Earned Run Average). Obviously, the math gets more complicated when guys are pitching different amounts of innings. And I don't think that relievers should count in the equation. The game begins with the two starters and they have more effect on the proceedings than all the other players.

It's really measuring almost the same thing as a W/L record but in a more elegant way. The arbitrariness of the five inning requirement is gone. I think people would be a lot more willing to accept the information as pertinent if it were reduced to a number as above. You have an entire generation of fans who have been taught that a W/L record is meaningless. But I would bet a shitload of cash that if Tom Tango developed this statistic, dan bernstein would be lapping it up although it's really just another form of W/L record.

In any case, I think that if you saw a pitcher with multiple seasons of negative numbers in this statistic, it would be a lot less likely that people would make excuses for him, talking about his "bad luck" and "lack of 'run support'".

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

JORR, you're saying for example the next time felix hernandez toes the rubber (3.38 ERA) if he's up against a guy with a 5.00 ERA you want him to be -1.62 or +1.62 as the overall stat? i think the stat would only apply to the pitcher who's leading in ERA at the time and since -1.62 is the same as +1.62 you'd have to sort that out....

but what about bullpens then? i guess since you're calling this an analogue to W/L it only matters for the SP but if you have a 6 inning pitcher versus a guy who regularly goes 7-8+ you might wanna factor bullpens in tho.

i think people naturally do this when they see a guy with a 2.something ERA up against a guy with a 4.something or 5.something ERA when they think "wow that pitcher with the lower ERA should win this" but i can see where your stat would be a nice frame of reference.... unless i'm getting it completely wrong.

_________________

It would be the same thing.

He could be a very good pitcher on a team with a very bad offense.

He could give up one over 9, and his team cannot score any against the opposing pitcher.

Giving up one over 9 is excellent. He would still have a losing statistic.

_________________
Rick Hahn is the best GM in baseball.

Who is the pitcher who spent a career on a team(s) with very bad offenses.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

I'm only looking at it head to head. It's a comparison of the starter's record vs. the other starters he faced in the games in which he faced them. Of course, there is always the factor of the different offenses, but as I've stated time and time again, the majority of big league offenses average within a single run of each other. If you have a large enough sample, the difference in offenses should smooth out.

Some umpires have far lower "ERA"s than others. To me that's a far more pertinent factor in run scoring in a specific game than offenses with a wide array of moving parts. The pitchers and the plate ump are going to be the most important factors in how many runs are scored.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

In an odd kind of way, you are arguing for sabremetrics. You are looking for a different way of advocating a methodology of measuring starting pitcher performance. I think this idea falls in the same category that non-advanced metrics guys regurgitate frequently: there is no stat, traditional or "advanced" that encapsulates the totality of a given player's value (such as, oh idk, WAR). All stats should be used as tools, and just one part of a totality in evaluating what happened (or, may be likely to happen going forward). Such a stat you are proposing could paint a pitcher in a meaningfully different way; but is yet only one brushstroke in the micro and macro

_________________
1926-1931-1934-1942-1944-1946-1964-1967-1982-2006-2011

This leads us right back into a core argument that a pitcher is or is not pitching against the other pitcher.
I am of the belief that the starting pitchers are not pitching against each other, rather, the offense of the other team. You are of the belief that the pitchers are pitching against each other.
There is an arguable merit to both sides of the coin.

_________________
"Play until it hurts, then play until it hurts to not play."
http://soundcloud.com/darkside124 HOF 2013, MM Champion 2014

That's fair. And I know that the current conventional wisdom is that a pitcher's job is "to prevent run scoring" - or even "to prevent baserunners". And I absolutely disagree. I think a pitcher's job is simply to allow less runs than the opposing pitcher(s). That's a reasonable enough goal. And a smaller goal is not to allow the opponent to score first. Teams that score first are winners. We talk about the difference in offenses in terms of runs per game, and it's usually less than one, but the difference is even more marginal over a specific four inning stretch and highly dependent upon the pitcher that offense is facing. A big part of Samardzija's failure is that the opponent scores first so often. This puts his team in a position to lose. If you were looking at a team's real odds to win a game, they drop precipitously upon the opponent scoring first.

Anyway, I appreciate the polite response. The idea isn't to bash the other guy over the head. It's to join together to seek real illumination. And in that vein I had another idea for my statistic. In my original Quintana/Chen example I had Quintana at -1 for the game. But let's say we were calculating the number after the 162 game season and we saw that the Royals (or whatever team Chen was on) averaged .5 runs per game more than Quintana's team. We could then adjust Quintana's number to -0.5. I call that RERA+ (Adjusted Relative Earned Run Average).

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

JORR, don't you think there is a lot of luck in baseball? I do. Therefore, the only stats that matter are ones that have a pretty good sample size. In that case, the basic stats we use are pretty good.

_________________
Reality is your friend, not your enemy. -- Seacrest

Yeah. There's luck in everything. I don't think we need a complicated algorithm to let us know strikeouts are good and walks are bad from a pitching standpoint.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

Somewhat related, and I think we've discussed before, I'd like to see a modified ERA, as originally proposed by Mike Murphy (or he's the first person I heard suggest it). If a pitcher gave up a double, was taken out, and the relief pitcher allowed the runner to score, each pitcher would be charged with 0.5 run. Similarly, if a pitcher gave a triple, and the next pitcher allowed the run to score, it would be divided up 0.75 to the first pitcher and 0.25 to the second pitcher.

_________________

I agree it would be more accurate to charge each participating pitcher with a portion of the run.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

WAR addresses this

_________________

A pitcher cannot give up a fraction of a run the same way a team cannot score a fraction of a run.

_________________
Rick Hahn is the best GM in baseball.

*SIGH* A full run has been scored. What's being assigned would not be a fraction of a run, but a fraction of the responsibility for said run.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

Full runs are also scored in every MLB game every played as far as I know, and yet still we can come up with average runs scored for a team over the baseball season, and sometimes that number is a fraction.

Amazing how that works.

_________________
Rick Hahn is the best GM in baseball.

No shit. I'm happy you understand what an average is. But what you continually fail to acknowledge is that when the difference in average runs scored by two separate offenses is a fraction of a run, the difference between those teams is so thin as to be utterly insignificant and, in fact, impossible to measure within the space of a single real life game.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

Last year, Colorado led the National League with 755 runs scored. 4.66 runs a game.

Last year, the Cubs scored 3.79 runs a game. Or 0.87 runs per game less.

The Colorado offense was highly praised. The Cubs offense was ridiculed all season.

Why?

Did the Cubs pretty much have the best offense in the National League and had a couple of bad breaks in a couple games?

_________________
Rick Hahn is the best GM in baseball.

Wasn't hot enough?

_________________
2018
#ExtendLafleur
10 More Wins

Right off the bat, there is an obvious inherent run-scoring advantage in playing 81 games in Colorado. That aside, unless you know a way to score 9/10 of a run in a game, these offenses are virtually equivalent in any single contest. The difference between them only shows up over the course of many games. And I sure don't expect that .87 of a run to be the excuse why a guy someone is claiming to be a great but "unlucky" pitcher lost a game that was won by a guy many would call an ordinary pitcher.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

You are the worst sports poster on the board. You're essentially saying all 30 MLB teams have complete parity, since pretty much every team within each league has RS and RA within 162 of each other.

2014 Regular Season

Colorado and Arizona were the only RA outliers in the NL, and Colorado and Dodgers were the only RS outliers. And you've already given the reason why Colorado would be an exception for both.

So the worst team in the NL is as good as the best team in the NL in both offense and pitching/defense?

You have flawless logic. Impeccable. It is unfathomable how only you see this and everyone disagrees with you. You're just worlds ahead of the rest of baseball.

_________________
Rick Hahn is the best GM in baseball.

I'm saying that in the space of a single game most MLB offenses are similar and I'm saying it because it's a fact. I'm sure I'm not the only one that sees it. And if I am, I'm unconcerned. For years everyone thought the scrifice bunt was a great strategy. Today some still do.

Stop trying to reflexively argue with me and think logically for a minute. If an offense averages .87 runs per game more than another offense, what does that actually mean beyond theory? A team can't score .87 runs in a single game. So sometimes that team may lose to the opponent by a run and sometimes they may beat them by two runs. The two run win isn't worth anymore than the one run loss. In any case, it's fair to expect that in most cases the wins of the high-scoring offensive team aren't occurring vs. great but unlucky pitchers, but rather against shitty pitchers, while most of the losses of the high-scoring team to a lower-scoring one are likely due to that offense facing superior starting pitchers.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

It means that a team scored 141 more runs than the other team did, and those runs occurred over the course of 162 games. And that offense is more effective than the one that scored 141 runs less over those same 162 games.

Simple and logical. This isn't theory at all, but fact. Teams that score more runs and give up less runs win more than lose. And while most teams score within 162 runs of each other, and give up runs within 162 of each other, offenses and pitching vary WIDELY between teams each and every season.

If you own a hundred thousand shares of Ford and it drops a fraction of a dollar between today and tomorrow, how are you feeling?

Fractions matter. Don't be a dolt.

_________________
Rick Hahn is the best GM in baseball.

Are you familiar with a team called the Oakland Athletics?

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don

We're talking about 162 game seasons. Not 60 game seasons.

Also, there will always be anomalies.

_________________
Rick Hahn is the best GM in baseball.

Of course. There's obvious correlation between run differential and record. And yet, correlation isn't the god of all things.

_________________
Anybody here seen my old friend Bobby?
Can you tell me where he's gone?
I thought I saw him walkin' up to The Hill
With Elon, Tulsi, and Don