Re: Harden close to signing with Rangers

[Transmogrified Tiger]

Harden could have dictated where he went a bit, but I have a hard time believing he would have barred the Cubs from trading him entirely.

[Guest]

Harden could have dictated where he went a bit, but I have a hard time believing he would have barred the Cubs from trading him entirely.

Especially if we told him he'd be a reliever if he stayed here. :D

 

(yes, I'm still hung up on this idea)

[Schwarber Fan]

MH says its official:

 

http://www.miamiherald.com/sports/baseball/wires/story/1376009.html

[Rob]

I notice that nobody is debating that when Harden is pitching, he's pretty much just as good as any other pitcher out there. (Which is good, as simply looking at xFIP the last few years would eviscerate most arguments to the contrary). The problem most seem to have is that he doesn't pitch all that many innings.

 

In a bit of math that I'm sure everybody here can understand, let's walk through this.

 

How much money do the best pitchers in baseball make? $20 mil or so? Harden pitches what, 50% as many innings as a conservative estimate? Well then, Harden's probably worth about half as much... or somewhere in the range of $10 mil.

 

There are mitigating factors of course, his bat, the chance he finally has that one healthy season, the chance he throws six innings and breaks down, and all that jazz about shuffling the rotation and bullpen to account for the innings he doesn't pitch.

 

But if anybody is coming up with a drastically different figure than that $10 mil, they're probably doing something wrong... my money would be on being unable to see past ERA and a flukey home run / fly ball rate.

Yeah that math doesn't work. A guy half as good isn't worth half as much.

 

If you wanted to take that sort of approach, you'd compute what 50% of Harden's innings plus 50% replacement level innings would give you in terms of production, then try and find a pricetag for that, then deduct the second guy's salary from that number, and then make some downward adjustment for the fact that you need to use 2 roster spots for one role.

 

(Even that's a simplification, because the 50% of innings Harden doesn't give you is not spread to one player, but to many, some starters covering missed starts, and some relievers covering early exits.)

 

Except that Harden isn't half as good. He's just as good in half the time. Value per IP is a pretty linear function unless we want to take leverage into account (which is silly for our purposes)

 

You're trying to figure out what all we can get from that spot in the rotation, and it's confusing your evaluation.

[haubrich91]

Nice move by the Rangers. I hope Rich stays healthy and has a good year.

[davearm2]

I notice that nobody is debating that when Harden is pitching, he's pretty much just as good as any other pitcher out there. (Which is good, as simply looking at xFIP the last few years would eviscerate most arguments to the contrary). The problem most seem to have is that he doesn't pitch all that many innings.

 

In a bit of math that I'm sure everybody here can understand, let's walk through this.

 

How much money do the best pitchers in baseball make? $20 mil or so? Harden pitches what, 50% as many innings as a conservative estimate? Well then, Harden's probably worth about half as much... or somewhere in the range of $10 mil.

 

There are mitigating factors of course, his bat, the chance he finally has that one healthy season, the chance he throws six innings and breaks down, and all that jazz about shuffling the rotation and bullpen to account for the innings he doesn't pitch.

 

But if anybody is coming up with a drastically different figure than that $10 mil, they're probably doing something wrong... my money would be on being unable to see past ERA and a flukey home run / fly ball rate.

Yeah that math doesn't work. A guy half as good isn't worth half as much.

 

If you wanted to take that sort of approach, you'd compute what 50% of Harden's innings plus 50% replacement level innings would give you in terms of production, then try and find a pricetag for that, then deduct the second guy's salary from that number, and then make some downward adjustment for the fact that you need to use 2 roster spots for one role.

 

(Even that's a simplification, because the 50% of innings Harden doesn't give you is not spread to one player, but to many, some starters covering missed starts, and some relievers covering early exits.)

 

Except that Harden isn't half as good. He's just as good in half the time. Value per IP is a pretty linear function unless we want to take leverage into account (which is silly for our purposes)

 

You're trying to figure out what all we can get from that spot in the rotation, and it's confusing your evaluation.

This is not nearly so simple as you're making it out to be.

 

For that $20M, an absolute stud starting pitcher might give you a 3.00 ERA over 34 starts and 238 IP (or 7 IP per start on average).

 

Now let's consider what half of that same production might look like, or 119 IP.

 

At one extreme you could have a guy that makes 34 starts and averages 3.5 IP per start.

 

At the other extreme you could have 17 starts @ 7 IP per start.

 

Now I think (hope) we can agree that those two hypothetical pitchers are not of equal value. I'm not sure you'd want the first guy at any price -- that's way too much wear and tear on the bullpen. The second guy could be valuable, but it would depend greatly on the sort of production you could expect from the guy making the other 17 starts -- and how much you'd have to pay him. If Harden gives you the 3.00 ERA, but the AAA fill-in pitches to a 6.00 ERA, then the appropriate value to assign to Harden would be that of a 4.50 ERA "workhorse" starter. Something in the Jason Marquis ballpark, in other words. And I seem to recall everyone hating the $7M per year Hendry gave Marquis.

 

The true Rich Harden would be approximately 1/3 the first guy, and 2/3 the second guy -- not pitching deep into games is part of his problem, but missing starts due to injury is the bigger problem. That tells me that $7M is probably on the generous side, since the second guy is more valuable than the first.

[imb]

Well, at least we go to Texas this year. Maybe I can see Harden in person.

 

I'll actually feel more comfortable buying a Harden shirtzee if he's not a Cub. Maybe this is what made Hendry want to no re-sign him.

[A New Era]

I liked Harden don't get me wrong. And I always felt pretty comfortable when he was out there. But he could never go much more than 5 innings and that was always uncomfortable considering trying to rest our bullpen, etc. I don't put too too much anger toward Hendry on this one.

[17 Seconds]

I liked Harden don't get me wrong. And I always felt pretty comfortable when he was out there. But he could never go much more than 5 innings and that was always uncomfortable considering trying to rest our bullpen, etc. I don't put too too much anger toward Hendry on this one.

 

that only costs the bullpen like one inning every 5 games

[Shilzzz]

I liked Harden don't get me wrong. And I always felt pretty comfortable when he was out there. But he could never go much more than 5 innings and that was always uncomfortable considering trying to rest our bullpen, etc. I don't put too too much anger toward Hendry on this one.

 

this is the misconception. he certainly doesn't throw complete games ever, but the following were the innings pitched in each Harden start in 09:

6

3

6

6

3

2

7

6

6

6

6

5

5.1

7

2

5

6

7

6

5

6

7

7

6

5

5

4

3

 

and, here is a statement from the article on cubs.com when Harden was dealt to the Cubs, which is all i could quickly find about his innings per start in 2008 (this is concerning his time in Oakland):

Harden spent one month this season on the disabled list (April 10-May 11) with a strained right shoulder and has averaged six innings per start since returning, four times pitching into the seventh inning or beyond. He has turned in a quality start in seven of his last 10 outings, posting a 2.02 ERA (14 ER/62.1 IP) and a .190 batting average against in that span starting May 17. Overall, he has allowed two earned runs or less in 10 of his 13 starts.

 

So, yeah, he has issues with going deep into games, but with the kind of work he provides, if the Cubs had a more resolved bullpen that can shut a team down in the last three innings, there is less to worry about than people say.

[erik316wttn]

Is there any word on if his familiarity with the AL West parks and players had anything to do with his decision to sign with Texas?

[imb]

Is there any word on if his familiarity with the AL West parks and players had anything to do with his decision to sign with Texas?

 

i'm sure it played somewhat less of a role than his familiarity with how much dough he was getting paid

[Eric Clipperton]

davearm2:

You say if Harden has a 3.00 ERA and Samardzjia has a 6.00, this makes Harden end up being worth a 4.50 ERA guy in a full season. The problem with this is that 6.00 is well below replacement level pitching, therefore it would have negative value. This negative value is then being subtracted out from Harden's true performance value to arrive at your $7MM figure. You can't just assume that the fill-in is horrible, and it makes no sense to me why the poor choice of fill-in starter should penalize Harden in the analysis. You can see how bumping up the bad half of the platoon pitcher to a 5.00 ERA level now makes "Harden+caddy" a half point ERA better over the full season, to a suddenly nice 4.00 ERA pitcher. That is to say more Ryan Dempster than Marquis, and value in this altered version would be well above 10MM.

This is all to say that valuing a player based on the potential (poor) contributions of another injury fill-in is generally way off as a model in aggregate, and while Harden is gone now, this saga has brought up some interesting points to consider in the process.

 

On a related note, I am all for taking a chance on Ben Sheets- perhaps he wants to stick it to Milwaukee?

[Rob]

This is not nearly so simple as you're making it out to be.

 

For that $20M, an absolute stud starting pitcher might give you a 3.00 ERA over 34 starts and 238 IP (or 7 IP per start on average).

 

Now let's consider what half of that same production might look like, or 119 IP.

 

At one extreme you could have a guy that makes 34 starts and averages 3.5 IP per start.

 

At the other extreme you could have 17 starts @ 7 IP per start.

 

Now I think (hope) we can agree that those two hypothetical pitchers are not of equal value. I'm not sure you'd want the first guy at any price -- that's way too much wear and tear on the bullpen. The second guy could be valuable, but it would depend greatly on the sort of production you could expect from the guy making the other 17 starts -- and how much you'd have to pay him. If Harden gives you the 3.00 ERA, but the AAA fill-in pitches to a 6.00 ERA, then the appropriate value to assign to Harden would be that of a 4.50 ERA "workhorse" starter. Something in the Jason Marquis ballpark, in other words. And I seem to recall everyone hating the $7M per year Hendry gave Marquis.

 

The true Rich Harden would be approximately 1/3 the first guy, and 2/3 the second guy -- not pitching deep into games is part of his problem, but missing starts due to injury is the bigger problem. That tells me that $7M is probably on the generous side, since the second guy is more valuable than the first.

 

I'll say it again, you're confusing yourself.

 

Do you factor in how many innings our starters are pitching and how effective they are when deciding how much a reliever is worth? No. They're worth what they're worth... then you just decide whether or not it's a fit for the organizational philosophy and the funds you have available.

 

Harden is worth the contract he's getting. He may not fit what you want out of a pitcher, and you'd be well within your rights to pursue other options. But the innings he pitches at his level of effectiveness makes him worth the contract, whether or not his issues mesh with your philosophy or not.

[A New Era]

If we had the money I'd be all for keeping Harden. But damn Hendry for being stupid and shelling out bad contract after bad contract. Now we have to go after a guy like Cameron with that money because of another one of Hendry's many many mistakes.

[davearm2]

davearm2:

You say if Harden has a 3.00 ERA and Samardzjia has a 6.00, this makes Harden end up being worth a 4.50 ERA guy in a full season. The problem with this is that 6.00 is well below replacement level pitching, therefore it would have negative value. This negative value is then being subtracted out from Harden's true performance value to arrive at your $7MM figure. You can't just assume that the fill-in is horrible, and it makes no sense to me why the poor choice of fill-in starter should penalize Harden in the analysis. You can see how bumping up the bad half of the platoon pitcher to a 5.00 ERA level now makes "Harden+caddy" a half point ERA better over the full season, to a suddenly nice 4.00 ERA pitcher. That is to say more Ryan Dempster than Marquis, and value in this altered version would be well above 10MM.

This is all to say that valuing a player based on the potential (poor) contributions of another injury fill-in is generally way off as a model in aggregate, and while Harden is gone now, this saga has brought up some interesting points to consider in the process.

 

On a related note, I am all for taking a chance on Ben Sheets- perhaps he wants to stick it to Milwaukee?

How do you like this approach:

 

Replacement level ERA = 5.00 (?)

In 119 innings, that translates to 66 earned runs.

 

Harden assumed ERA = 3.00

In 119 innings, that translates to 40 earned runs.

 

Difference = 26 runs, or roughly 2.6 Wins.

 

How much is a 2.6 WARP starting pitcher worth on the open market? I dunno, I'm honestly asking.

 

For reference:

Dempster has averaged 5.6 WARP since the Cubs converted him back to a starter.

Lilly has averaged 4.2 WARP in his three seasons with the Cubs.

Zambrano has averaged 5.1 WARP since 2002.

Wells was good for 4.7 WARP in his first year.

 

Since becoming a starter in STL, Marquis has averaged 1.9 WARP in 6 seasons.

 

So although that production from Harden is a bit above Marquis levels, he'd still be the Cubs' least valuable starter.

 

Incidentally, applying this sort of analysis to the Randy Wolf deal makes it look pretty awful for the Brewers: Wolf was good for 4.6 WARP in 2009, but in the 6 years before that he couldn't crack 2 (1.6, 1.5, 1.4, 0.5, 1.8, 1.3). Yuck.