Theo Epstein Sploosh Thread

[Quality Start Machine]

PITTSBURGH PIRATE Liriano should get some consideration.  When he hasn't been hurt the last two seasons, he's been very good to dominant.  He's proven that he can win against the division.

You'd.

Just kidding. He'd be a nice pickup.

I see him as a real nice #3, behind Ace-To-Be-Determined and Arietta.

[Oleg]

I think FIP is useful for finding undervalued guys, but it doesn't seem to do justice to elite pitchers. I'm not smart enough to know why that is. Probably something about guys who are good at getting weak contact? I think Matt Cain was the first guy who consistently did this, and it was always interesting to try and figure out why (though no one did).

Agreed, it's definitely not cut-and-dry around the tails of the distribution -- the really elite guys and the borderline replacement players. Once exit velocity/angle data becomes available at every stadium (and hopefully freely available to the sabermetric community), it should allow for the development of better ERA estimators that don't rely on a simple categorical LD/GB/FB variable for batted ball type.

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Also, about the only team I'd consider dealing with for pitching is The Mets.  Not sure how injury history plays into any of this but if they were willing to trade Harvey or Wheeler or anyone else like that, I'd consider liking a Baez or Castro or Russell trade.  But, that would need to happen next off-season, I think.

I'm also not sure they can't win without a true number 1.  If they go out and get McCarthy or Tomlin to go with Arrieta and Hendricks, they'd still outscore everyone.  Maybe even a trade for Minor?  Beachy and Medlen coming back from TJD?  Go somewhat cheap on Ervin Santana?

I'd love to see Lester or Scherzer but I'm not convinced they're necessities for the next great Cubs team.

[Eli]

I'm also not sure they can't win without a true number 1.  If they go out and get McCarthy or Tomlin to go with Arrieta and Hendricks, they'd still outscore everyone.  Maybe even a trade for Minor?  Beachy and Medlen coming back from TJD?  Go somewhat cheap on Ervin Santana?

There's only one guy on that entire list that I feel certain can provide 200 innings of above-league-average pitching next year. I guess Hendricks has a good shot at it, but everyone else is just a big question mark. It's fine to take fliers on guys like that, but they shouldn't be counted on. I not really sure why everyone is so eager to settle for mediocre pitching.

[InternetApex]

I'm also not sure they can't win without a true number 1.  If they go out and get McCarthy or Tomlin to go with Arrieta and Hendricks, they'd still outscore everyone.  Maybe even a trade for Minor?  Beachy and Medlen coming back from TJD?  Go somewhat cheap on Ervin Santana?

There's only one guy on that entire list that I feel certain can provide 200 innings of above-league-average pitching next year. I guess Hendricks has a good shot at it, but everyone else is just a big question mark. It's fine to take fliers on guys like that, but they shouldn't be counted on. I not really sure why everyone is so eager to settle for mediocre pitching.

Nobody is settling for mediocre. People are just choosing to bet that the more above average arms you can stockpile, the more likely you are to have an above average staff. And clearing out salary for a bonafied No. 1 talent doesn't guarantee the health of that one individual will be worth the investment. Would you rather have Oakland and Detroit starting staffs or Baltimore's? Clearly you'd take the Dodgers starters over the Royals, right? I'm not so sure.

Kershaw's alarmingly poor performance in the playoffs for his career have to be due in large part to the superhuman workloads he's compiled each season. I think he'll get blown up again in Game 4 and that will be all she wrote.

If you can get an above average staff in need of one bonafied ace, you can usually rent one for the stretch run. It happens every single year. The deadline has been a buyer's market recently. But then getting to October guarantees you nothing but a fighting chance. So bet heavily on your every day players and have DEEP pitching staff that might have something in the tank come postseason. That seems to be working out at least as well as what the Tigers and Dodgers did.

[Eli]

People are just choosing to bet that the more above average arms you can stockpile, the more likely you are to have an above average staff.

In theory. But how do you know which above-average arms end up in the rotation? They already have Arrieta, Hendricks, Wood, Wada, EJax, Turner, Straily, Doubront in the starter picture. That's seven interesting guys (plus EJax) who already need to shake out somewhere, which seems like plenty of depth to sift through.

[InternetApex]

People are just choosing to bet that the more above average arms you can stockpile, the more likely you are to have an above average staff.

In theory. But how do you know which above-average arms end up in the rotation? They already have Arrieta, Hendricks, Wood, Wada, EJax, Turner, Straily, Doubront in the starter picture. That's seven interesting guys (plus EJax) who already need to shake out somewhere, which seems like plenty of depth to sift through.

I honestly don't know. Fortunately for all of us it isn't up to me. Because Juan Cruz and Kevin Hart can tell you how bad I would be at that job. At least that's what I take their restraining orders against me to imply.

[J. Walter Weatherman]

bonafied

I think it's spelled "bonerfied".

[ChuckD]

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Sort of. I mean, xFIP is sort of what you're talking about there.

Instead of using the pitcher's actual HR total (like FIP), xFIP takes the pitcher's FB rate and uses the league average HR/FB rate to get an expected HR total that should account for some luck.

What I'm more talking about is using the actual trajectory data for batted balls. There are companies who use doppler radar to record a batted balls' exit speed off the bat, launch angle, bearing, and some other stuff that can't be charted by a human. My company licenses it from Trackman, but it's the same sort of system/data that (I expect, at least) will be part of MLBAM's HitFX.

Using that sort of data should give a better value of a batted ball beyond simply LD vs GB vs FB. For instance, if you've got two FBs, those would be considered "equal" under xFIP. But if you can differentiate those two FBs to say that one was a ball with a 25 degree launch angle and an exit speed of 100mph, that will have a higher average run value than say a ball hit with a 45 degree launch angle and an exit speed of 80mph. The former would be a home run in most parks, whereas the latter is going to be a lazy fly ball (unless it's directly down the RF line at Fenway). Being able to classify those FBs and apply different weights accordingly should improve the accuracy of things like xFIP or xBABIP.

[World's #1 Astros Fan]

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Sort of. I mean, xFIP is sort of what you're talking about there.

Instead of using the pitcher's actual HR total (like FIP), xFIP takes the pitcher's FB rate and uses the league average HR/FB rate to get an expected HR total that should account for some luck.

What I'm more talking about is using the actual trajectory data for batted balls. There are companies who use doppler radar to record a batted balls' exit speed off the bat, launch angle, bearing, and some other stuff that can't be charted by a human. My company licenses it from Trackman, but it's the same sort of system/data that (I expect, at least) will be part of MLBAM's HitFX.

Using that sort of data should give a better value of a batted ball beyond simply LD vs GB vs FB. For instance, if you've got two FBs, those would be considered "equal" under xFIP. But if you can differentiate those two FBs to say that one was a ball with a 25 degree launch angle and an exit speed of 100mph, that will have a higher average run value than say a ball hit with a 45 degree launch angle and an exit speed of 80mph. The former would be a home run in most parks, whereas the latter is going to be a lazy fly ball (unless it's directly down the RF line at Fenway). Being able to classify those FBs and apply different weights accordingly should improve the accuracy of things like xFIP or xBABIP.

Man, you're like the Freddy Mercury of statfaggery.

[Slaky]

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Sort of. I mean, xFIP is sort of what you're talking about there.

Instead of using the pitcher's actual HR total (like FIP), xFIP takes the pitcher's FB rate and uses the league average HR/FB rate to get an expected HR total that should account for some luck.

What I'm more talking about is using the actual trajectory data for batted balls. There are companies who use doppler radar to record a batted balls' exit speed off the bat, launch angle, bearing, and some other stuff that can't be charted by a human. My company licenses it from Trackman, but it's the same sort of system/data that (I expect, at least) will be part of MLBAM's HitFX.

Using that sort of data should give a better value of a batted ball beyond simply LD vs GB vs FB. For instance, if you've got two FBs, those would be considered "equal" under xFIP. But if you can differentiate those two FBs to say that one was a ball with a 25 degree launch angle and an exit speed of 100mph, that will have a higher average run value than say a ball hit with a 45 degree launch angle and an exit speed of 80mph. The former would be a home run in most parks, whereas the latter is going to be a lazy fly ball (unless it's directly down the RF line at Fenway). Being able to classify those FBs and apply different weights accordingly should improve the accuracy of things like xFIP or xBABIP.

Man, you're like the Freddy Mercury of statfaggery.

I'm pretty sure it's his job irl.

[Oleg]

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Sort of. I mean, xFIP is sort of what you're talking about there.

Instead of using the pitcher's actual HR total (like FIP), xFIP takes the pitcher's FB rate and uses the league average HR/FB rate to get an expected HR total that should account for some luck.

What I'm more talking about is using the actual trajectory data for batted balls. There are companies who use doppler radar to record a batted balls' exit speed off the bat, launch angle, bearing, and some other stuff that can't be charted by a human. My company licenses it from Trackman, but it's the same sort of system/data that (I expect, at least) will be part of MLBAM's HitFX.

Using that sort of data should give a better value of a batted ball beyond simply LD vs GB vs FB. For instance, if you've got two FBs, those would be considered "equal" under xFIP. But if you can differentiate those two FBs to say that one was a ball with a 25 degree launch angle and an exit speed of 100mph, that will have a higher average run value than say a ball hit with a 45 degree launch angle and an exit speed of 80mph. The former would be a home run in most parks, whereas the latter is going to be a lazy fly ball (unless it's directly down the RF line at Fenway). Being able to classify those FBs and apply different weights accordingly should improve the accuracy of things like xFIP or xBABIP.

Man, you're like the Freddy Mercury of statfaggery.

I'm pretty sure it's his job irl.

Whatever it is, I'm totally gay for it.

[World's #1 Astros Fan]

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Sort of. I mean, xFIP is sort of what you're talking about there.

Instead of using the pitcher's actual HR total (like FIP), xFIP takes the pitcher's FB rate and uses the league average HR/FB rate to get an expected HR total that should account for some luck.

What I'm more talking about is using the actual trajectory data for batted balls. There are companies who use doppler radar to record a batted balls' exit speed off the bat, launch angle, bearing, and some other stuff that can't be charted by a human. My company licenses it from Trackman, but it's the same sort of system/data that (I expect, at least) will be part of MLBAM's HitFX.

Using that sort of data should give a better value of a batted ball beyond simply LD vs GB vs FB. For instance, if you've got two FBs, those would be considered "equal" under xFIP. But if you can differentiate those two FBs to say that one was a ball with a 25 degree launch angle and an exit speed of 100mph, that will have a higher average run value than say a ball hit with a 45 degree launch angle and an exit speed of 80mph. The former would be a home run in most parks, whereas the latter is going to be a lazy fly ball (unless it's directly down the RF line at Fenway). Being able to classify those FBs and apply different weights accordingly should improve the accuracy of things like xFIP or xBABIP.

Man, you're like the Freddy Mercury of statfaggery.

I'm pretty sure it's his job irl.

I know that it is.  In fact he even refers to "my company" in the above paragraph.

I was just making a lame attempt at humor, see...

[Slaky]

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Sort of. I mean, xFIP is sort of what you're talking about there.

Instead of using the pitcher's actual HR total (like FIP), xFIP takes the pitcher's FB rate and uses the league average HR/FB rate to get an expected HR total that should account for some luck.

What I'm more talking about is using the actual trajectory data for batted balls. There are companies who use doppler radar to record a batted balls' exit speed off the bat, launch angle, bearing, and some other stuff that can't be charted by a human. My company licenses it from Trackman, but it's the same sort of system/data that (I expect, at least) will be part of MLBAM's HitFX.

Using that sort of data should give a better value of a batted ball beyond simply LD vs GB vs FB. For instance, if you've got two FBs, those would be considered "equal" under xFIP. But if you can differentiate those two FBs to say that one was a ball with a 25 degree launch angle and an exit speed of 100mph, that will have a higher average run value than say a ball hit with a 45 degree launch angle and an exit speed of 80mph. The former would be a home run in most parks, whereas the latter is going to be a lazy fly ball (unless it's directly down the RF line at Fenway). Being able to classify those FBs and apply different weights accordingly should improve the accuracy of things like xFIP or xBABIP.

Man, you're like the Freddy Mercury of statfaggery.

I'm pretty sure it's his job irl.

I know that it is.  In fact he even refers to "my company" in the above paragraph.

I was just making a lame attempt at humor, see...

Success!

[World's #1 Astros Fan]

FIP is nice but hardly perfect.  Combine FIP with FB/HR ratio and line drive % and we may be on to something.  I think this is the sort of thing y'all may be alluding to?

Sort of. I mean, xFIP is sort of what you're talking about there.

Instead of using the pitcher's actual HR total (like FIP), xFIP takes the pitcher's FB rate and uses the league average HR/FB rate to get an expected HR total that should account for some luck.

What I'm more talking about is using the actual trajectory data for batted balls. There are companies who use doppler radar to record a batted balls' exit speed off the bat, launch angle, bearing, and some other stuff that can't be charted by a human. My company licenses it from Trackman, but it's the same sort of system/data that (I expect, at least) will be part of MLBAM's HitFX.

Using that sort of data should give a better value of a batted ball beyond simply LD vs GB vs FB. For instance, if you've got two FBs, those would be considered "equal" under xFIP. But if you can differentiate those two FBs to say that one was a ball with a 25 degree launch angle and an exit speed of 100mph, that will have a higher average run value than say a ball hit with a 45 degree launch angle and an exit speed of 80mph. The former would be a home run in most parks, whereas the latter is going to be a lazy fly ball (unless it's directly down the RF line at Fenway). Being able to classify those FBs and apply different weights accordingly should improve the accuracy of things like xFIP or xBABIP.

Man, you're like the Freddy Mercury of statfaggery.

I'm pretty sure it's his job irl.

I know that it is.  In fact he even refers to "my company" in the above paragraph.

I was just making a lame attempt at humor, see...

Success!

*furiously claps hands*

[Eli]

George Ofman on Twitter:

Are the Dodgers shopping Puig? You bet and the Cubs are of of a number of teams interested.

No idea what that would even take (probably one of Bryant or Russell, plus others), but I'm envisioning some sort of Cuban neo-bash brothers sitcom starring Puig and Soler where they torment Cardinals fans for the next decade.