GDT 9/18/19 Philadelphia Phillies (Eflin) @ Atlanta Braves (Teheran)

[zambonir]

You're only looking at free agents. (And I'm sure you wrote "why would he come here?" about Harper 100 times).

But I agree with SF, it doesn't matter what you call them. But it does matter what you have. Right now it doesn't matter whether we think someone is "Number 2," only if they would be the second-best pitcher in the Phillies rotation today, i.e, better than Eflin and Arrieta and VV and Pivetta. That's a low bar. If Eflin puts together a complete season next year and turns out to be a #3, fantastic, but he's not that yet. And neither is Jake really (unless he has a miracle comeback season at full health).

Any team where Eflin is considered the 5th starter obviously has a better rotation than one where he's the 4th starter. That doesn't mean we're going to add two guys for sure but they are certainly going to add one, and more than half the guys on that list would be fine.

Hamels becomes more intriguing than Bumgarner or Wheeler since you figure he'd be cheaper/shorter-term, but after this season I seriously doubt Klentak is going to repeat his aversion to dead money pitcher contracts.

[Patchmk88]

If I had a dollar for every time Austinfan said “Why would they come here” about a free agent, I’d have a new TV to watch the Phils on.

[andyb]

We probably have only 2 guarantees in the rotation (Nola and Arrieta) and one of them is only there because of his contract though he could easily be a mid-rotation starter if healthy. Of the three younger arms, we too often overreact to the most recent starts compared to the entire body of work. Eflin is ahead of the other two today, but his stuff is pedestrian enough that the order could easily change.

We are going to want an upgrade that is better - either free agent or trade. We'll have to pay money or prospects for the upgrade but it will be the right thing to do. A separate question is whether they go and get a 2nd mid-level rotation pitcher or not. I'd probably hope the younger pitchers can compete for the last 2 spots in the rotation (Eflin, Pivetta, Velasquez, maybe Smyly for 2 spots) which would provide depth and the losers in the competition could have a more defined bullpen role (where Smyly could be good also). A bullpen with more starters in it and more tandem starts might be a good thing for overall bullpen usage. I'd experiment with the LH starters like Smyly and Irvin as once around the lineup 2-3 inning starters that would then let us flip a RHP in the 3rd inning to get platoon advantages.

[zambonir]

Yup. But I think if you're trading for the starter you might include Velasquez or Pivetta in the deal. Two prospects plus one of them rather than two prospects plus two lesser prospects.

Vargas is also looking less and less like he's worth the $6 million guaranteed (and is it $8 milllion against the tax or does the part the Mets pay go to the Mets?).

I think you can get by with VV or Smyly as your #5 (or, sure, tandem starts) until July, rather than getting that second pitcher (who may or may not actually be better than Vargas or Smyly. Vargas was that pitcher when the Mets originally signed him, as was Tyler Chatwood).

[andyb]

Agree that is probably necessary (including excess arms in a multi-prospect trade). I am just not sure how much they add. Would hate to trade Pivetta for $0.25 on the dollar when all he might need is a different pitching coach to get him back to where he was a couple of years ago.

Or if we are serious about pitchers not going 3 times through the order, we really need to develop multi-inning relievers. I don't dismiss that Velasquez and Pivetta could be good there. They just also have value as traditional starters if they can put it together. Whatever we do we need to make decisions and roll the dice on some younger pitchers. We can't just have expensive free agent arms as we have gotten rid of a layer of inexpensive position players already (Crawford and Alfaro). Thus we need a couple of inexpensive starters somewhere or it all gets too expensive.

[Austinfan]

Harper wanted to go to LA, had they seriously pursued him, he's not a Philly. We overpaid and got our man, so that added what, 1 win? 2 wins?
Cole could be bought with a major overpay, but then you have to live with that contract for six years.

Andyb is so last millennium, today the emphasis is no longer on K-rate but exit velocity, and Eflin is one of the top guys in that regard. Soft contact is better than high K-rates, because you can go deeper into games and spare your bullpen.

As I pointed out, there's only a few pitchers who'll be 3+ WAR, so we're probably looking at adding 1-2 WAR starters unless we get lucky and one bounces back. Which is a good reason to re-up Smyly on an incentive laden deal, but not one with serious money. If you can't add quality, go for quantity and hope to get lucky.

[andyb]

Andyb is so last millennium, today the emphasis is no longer on K-rate but exit velocity, and Eflin is one of the top guys in that regard.

But Nola is below average in that category. Time to dump him before the league figures this out I guess.

Sometimes one needs to take a step back and look at all the data. Eflin needs to miss more bats. It is what prevents him from being a top starter. That is completely independent of saying that introducing weak contact is not also a plus. You get so incensed when I state the consensus opinion on Eflin as a back end starter? The question is why?

[allentown]

today the emphasis is no longer on K-rate but exit velocity, and Eflin is one of the top guys in that regard. Soft contact is better than high K-rates, because you can go deeper into games and spare your bullpen.

I agree strongl with this, but see it as being a contradiction to your normal emphasis on FIP (which I see as an artificial construct, which basically assumes that the 3 true outcomes are all that really matter). Producing weak contact for outs is better than Ks, at least to my mind, because it can lead to fewer pitches per out.

To Andy: I think that just goes to show that, although today more things are viewed as important, basically because they are measurable for the first time, there is no magical, secret approach to pitching or hitting success, which works for all players. The team which can tailor its instruction to the strengths of each player is the team which will excel.

[andyb]

To Andy: I think that just goes to show that, although today more things are viewed as important, basically because they are measurable for the first time, there is no magical, secret approach to pitching or hitting success, which works for all players. The team which can tailor its instruction to the strengths of each player is the team which will excel.

Yep. There are many ways to succeed. Velocity, changing speeds, movement, weak contact. Some approaches will work with some players. Others will not. Successful teams like the Dodgers and Astros are probably better at figuring out what works for each individual. Though having Verlander and Cole at the top of a rotation (along with Grienke who does things completely differently) does not hurt.

[Austinfan]

I think FIP applies to 90% of pitchers.

The outliers will be low exit velocity guys with concurrent lower BABIP than would be expected, similar to hitters with high exit velocities and/or above average speed who post higher BABIP than you'd expect.

[Philsmania]

I think FIP applies to 90% of pitchers.

The outliers will be low exit velocity guys with concurrent lower BABIP than would be expected, similar to hitters with high exit velocities and/or above average speed who post higher BABIP than you'd expect.

I think FIP is a load of bunk. It only takes into account TTTO and there is more to the game than that. Balls batted into play are a part of the game, always have been and, despite what a few eggheads say, always will be. And, you know what? Good pitching can influence how balls are batted into play. Good pitchers can induce weak or otherwise poor contact into outs. So there, by all means, balls batted into play and not hit of the park matter. Hence, I believe that FIP is a bovine excrement statistic.

[julio]

I think FIP is a load of bunk. It only takes into account TTTO and there is more to the game than that. Balls batted into play are a part of the game, always have been and, despite what a few eggheads say, always will be. And, you know what? Good pitching can influence how balls are batted into play. Good pitchers can induce weak or otherwise poor contact into outs. So there, by all means, balls batted into play and not hit of the park matter. Hence, I believe that FIP is a bovine excrement statistic.

A load of bunk?

Batting Average ignores bases on balls, HBP. SLG assumes that a home run is twice as valuable as a double. Fielding percentage ignores stupid mental errors - any error so egregious that the fielder doesn't get to the batted ball.

ALL baseball statistics have limitations. If you treat ANY statistic as a holy grail, as a be-all, end-all, you will eventually look foolish. But, does that mean that we should throw them all out, and abandon any systematic attempt to actually understand what we're seeing? Well, for some, it appears that it does.

The rest of us, however, will appreciate that a whole toolbox of limited tools can give us a whole range of partial understanding of the game... as long as we don't fall in love with one measure, and convince ourselves that it tells us everything. In this regard, FIP is no different than a lot of other tools we use. To use any particular statistical tool effectively, though, one must understand it - and it's certainly true that some fall short here, and treat some baseball stats as a kind of "magic" - which, I guess, leads others to reject the stat completely, because it has limitations.

[Philsmania]

Yes, all statistics have their weaknesses. But, I believe that FIP shows a glaringly incomplete picture.

[julio]

Hey, you updated your avatar! I like!

[zambonir]

Good pitching can influence how balls are batted into play. Good pitchers can induce weak or otherwise poor contact into outs. So there, by all means, balls batted into play and not hit of the park matter.

I think we want this to be true, and it is true anecdotally, but when you add up every pitch thrown over the course of the season, is isn't. Or at the very least - FIP measures what it measures, and it measures things way better than, ERA, because even if all the above is true you still have luck and poor defense (and inherited runs).

Now, whether FIP is as projectable as the Phillies wanted it to be is another question. Clearly it was not, but why? I'm assuming it's because our FIP "all-stars" started giving up more home runs and striking out fewer guys, more than it being BABIP variance (whether you think they can control that or not).

[julio]

Now, whether FIP is as projectable as the Phillies wanted it to be is another question. Clearly it was not, but why? I'm assuming it's because our FIP "all-stars" started giving up more home runs and striking out fewer guys, more than it being BABIP variance (whether you think they can control that or not).

The projectability of FIP should be testable. If it's really a characteristic of the pitcher, as opposed to environmental (e.g., team) factors, it should be reasonably consistent season to season for individual pitchers. This could be tested statistically.

We have to be careful, though. There's normal variance in pretty much any such stat. Individual BA fluctuates year to year (I seem to recall the standard deviation is something like .040 - the difference between a .260 BA and a .300 BA), and it's reasonable to assume that FIP would be subject to similar variation - so if you're trying to project a pitcher's FIP in the upcoming season, you have to look not only at what he did last year (and prior years), you have to take into account what a "typical" year to year variation might look like.

And of course, if you're a GM, you also have to try to address the part of team outcomes that isn't a function of FIP - you have to make sure that your club's fielding is making things better, not worse!

[mdougherty12]

Individual BA fluctuates year to year (I seem to recall the standard deviation is something like .040 - the difference between a .260 BA and a .300 BA),

Spot on, it's .039 in 2019, this Fangraphs article from today happens to have a table with mean and standard deviation for a few stats for 2019:

Which Hitter is Most 2019?

[zambonir]

Spot on, it's .039 in 2019, this Fangraphs article from today happens to have a table with mean and standard deviation for a few stats for 2019:

And that being the case, even when you are trying to argue that 50-60% of a certain team's line-up has all experience a drop in BA, that's not really a significant number (especially when many of the drops aren't even that small).'

What's the standard deviation for OPS+ though. I assume one reason for advanced stats is there's less noise.