Oz27

Good stuff and I very strongly agree. While the trend is in favor of this, I think it should be moving more quickly. Third-time-through-the-order-penalty is a very clear thing which we've known about for well over a decade and which has a very substantial impact on a pitcher's ability to get hitters out. On the flip side, relief pitchers are very good at getting batters out in their short appearances. Isn't it obvious then that you want less of the former and more of the latter? Sure, you're not quite as worried about it for the Kershaws of the world but I don't understand why average to below average pitchers are still being pushed so deep into games. Like I said, though, the trend is towards starters throwing fewer innings and that is good. To illustrate that further, here is the number of 180 IP seasons by year -

2016: 46

2015: 56

2014; 66

2013: 64

2012: 62

2011: 73

So as I said, the trend is very much in a downward direction and it's pretty significant. But there are still things about starter usage which I don't understand. For example, 11 starters with an ERA+ below 100 pitched more than 180 innings this year. Why the hell are teams trying to milk a couple of more outs from a tired pitcher who isn't even that good when he's well rested, when they could be handing the game over earlier to relievers far more likely to get those outs than a fatigued, below average starter?

There is a false narrative which existed for awhile here, linking our bullpen struggles in 2016 to overuse. That was just nonsense. We ranked 13th in bullpen innings pitched, with 544. In the days of seven or eight man bullpens, it's preposterous to suggest they cannot handle that work. Hell, they should be able to handle a lot more. Our bullpen was awful, compared to other bullpens, but our rotation was too. At the end of the day, a below average reliever is a better bet to get outs for you than a below average (or worse) starter with a high pitch count going through the order the third time. So even in our bad bullpen situation, I think we should have been quicker to pull the trigger on pitching changes.

If we're actually trying to win in 2017 - and it seems like we're at least making some attempt to do so - then it's crazy for us to not make a big effort to improve the bullpen. Getting down on our knees and praying that Street and Bailey can force out one last productive season each is not a strategy which should be employed by a team aiming to compete but it seems like that is where we are at. Maybe Jansen and Chapman weren't ever realistic (and I wouldn't have wanted Chapman at that price) but it is imperative we improve the bullpen in our position. Our rotation is not likely to be particularly good and it is also unlikely to be durable, so in that position it is common sense to stock up on reliable relief arms and give more innings to the bullpen. It's not too late, I guess, but we need to do something about it if we're at all thinking we can compete in 2017.

32 minutes ago, Erstad Grit said:

My two responses. 

1. Like probably you, I watch a lot of Angel games. I'm pretty sure if CI truly improved from one of the worst to one of the best it would have been more noticeable. I still saw the pitches in the FX box being called balls because he made them look like pitches in the dirt. 

2. This article would carry more persuasion with me had he not "regressed" this last season. The article paints a picture of him figuring it out. If that were the case I'm confused why it would disappear again. It's not like a skill like speed which can just disappear.  

My guess is there is some value in these formulas but CI exposed some limitations. I'm a huge believer of pitch framing. It's a valuable skill that could highly impact a game. I think there have been some good strides in figuring out how to quantify it. 

It sounds like you and Ozzy just conclude that anyone who doesn't buy into it completely is a moron. 

 

No, but I would question the critical thinking processes of a person who decides to dismiss something that has been thoroughly researched and developed by half a dozen of the smartest people involved in this sport because one player's statistics differ to their subjective assessment.

Also, if a shitty player has a one-year offensive breakout and then reverted back to their previous shitty self, would you:

a): believe that player was actually offensively valuable in that one good season, or

b): rubbish offensive statistics because of that one example

Assuming you would take option 'a', as any reasonable person would, why would you treat pitch framing metrics differently?

10 minutes ago, eaterfan said:

It's actually funny because framing used to be the bastion of the old school. Now that pitch FX has made it possible to quantify is suddenly unimportant to the old school crowd.

That is true and yeah it is bizarre. The numbers aren't perfect and will be refined over time but I trust them a lot and consider them very accurate. Yet people are very keen to find "problems" with the measurements, even though in many cases those problems have already been accounted for.

11 minutes ago, Stradling said:

Or that it looks like it counts for only a handful of runs a season.  

I don't understand what you're saying here?

4 hours ago, Stradling said:

Yea, I read that tweet.  He and Victor were tweeting back and forth about it.  I have no idea if it is statistically accurate, I'm sure @Oz27 would know, but it certainly seems like the pitcher that is constantly around the plate gets more calls that the dude that is wild.  

I had a different response written here before but really I should have just said that the Baseball Prospectus metric already controls for any impact the pitcher has. It also does the same for the hitter and umpire. You can read more on that here - http://www.baseballprospectus.com/article.php?articleid=25514

Anyway, people seem pretty keen to play down the importance of framing because it still challenges long-held beliefs but the evidence that this is a skill worth a lot is very strong.

3 hours ago, eaterfan said:

Why?

Because saying unintelligent, provocative things is fun for some people?

3 hours ago, Erstad Grit said:

As an umpire I believe very much in the skill set.  Some catchers have bad habits which makes strikes look like balls.  Others excel at "sticking" pitches to make them look like strikes.  I don't question the skill but the difficulty in grading it. 

I offered up CI as a case study.  He is horrible yet the stat rated him high.  

Do we say that a hitter is terrible even though he has a high OPS+ or wRC+? No, of course not. That would be a sure-fire way to earn a one-way ticket to the lunatic asylum. So I don't understand why anyone, if they had bothered to go to any effort to understand them, would do that with pitch framing metrics. I consider it one of the most reliable statistics in baseball for two key reasons.

Firstly, as I mentioned earlier in the thread, there has been a history of an extremely high correlation in year-to-year framing results. The year-to-year correlation coefficient for individual framing metrics has been as high as 0.82. Admittedly, that has declined a lot in recent years but I put that down much more to framing being a teachable skill than any problem with the statistic. Anyway, the 'average' year-to-year correlation coefficient for pitch framing stats from 2008 to 2015 was 0.65. By comparison, for wRC+ it was .512. For OBP it was .546. For OPS it was .549. Put a simpler way, pitch framing metrics have been a better predictor of future pitch framing success than the most widely used offensive statistics have been of future hitting success. I refuse to believe that could be true unless pitch framing numbers were accurately reflecting a true, meaningful skill.

Secondly, pitch framing numbers are produced by thousands of events. We use 500 plate appearances - or events - to judge someone's hitting ability. But for everyday catchers we are using up to 7500 events to make a judgement. So the statistic is being produced with a sample size as much as 15 times larger what we're using to judge hitters. In offensive stats you occasionally get statistical outliers (after all the difference between a .250 hitter and a .300 hitter roughly works out to one hit per week over a full season) but when the sample size is so much bigger, outliers aren't really going to exist.

Pitch framing stats have been criticized in some circles and people like to criticize their findings, based on nothing more than their own limited subjective assessments. It's all well and good to say "he is horrible yet the stat rated him high" because the numbers don't say what you want them to, but if you wouldn't do that for offensive metrics then I'd suggest it's awfully silly to do it for pitch framing data.

40 minutes ago, ettin said:

Sorry I'm just not understanding this line of thought that the closer he gets to free agency the more likely an extension can occur. If I were Trout and only 1-2 years from free agency I'd probably start leaning, like a lot of players do, towards testing the waters and choosing where I'd like to end up.

Better to pay him now rather than in that chaotic free agent market environment. You will save money long term by sealing the deal sooner even if it is at $500M (or more).

I disagree with this line of thinking. The reason I think an extension is far more likely to happen in the final year or two of the deal is arriving at a mutually reasonable number now is practically impossible because the market hasn't set. If they started negotiations now, they're probably starting at figures more than $100 million away from each other and that would be pretty damn hard to reconcile. If I were Trout, I wouldn't entertain the discussion seriously until Harper and Kershaw sign because they're going to set a big-time precedent. An extension now is just totally unrealistic.

13 minutes ago, Erstad Grit said:

I recall last season when mariners signed him many here were using this stat  saying CI was an elite framer.  My eyes told me he was one of the worst.  

He had one season where he was elite, sandwiched around a career of being terrible. That is extremely rare but you do see it from time to time, in the same way shitty hitters occasionally have one-off excellent seasons.

I consider another pre-free agency Trout extension extremely unlikely. Why would you sign another one in his position? You're the best player of your generation, don't you want to test the market at least once? Having said that, it's at least possible in the final couple of years of his deal. At the moment, it isn't. Firstly, if you're his agent you're begging him to wait until after the mega free agent class of the 2018-19 offseason, because the bar is going to be raised by the mega deals Harper and Kershaw (and even Donaldson) will sign. Secondly, how do you come up with a reasonable number now? To sign Trout to an extension now, you'd need to add what he would get in free agency (or at least close to it) to our existing commitment. Trout's agent is probably telling him - or at least thinking - that 13 years, $500 million is possible as a free agent. It might well be, too. At the very least, 10/$400 is realistic. So let's split the middle and say 12 years, $420 million is his market value and we sign him to that extension, on top of what we've already committed to, then we would have him signed for effectively 16 years and $542 million. Yes, that sounds absurd. It sounds like a video game number. It's also entirely possible that my number is too low (or the years might be too high for that dollar figure). But to make it worth it for Trout to sign an extension now, it would have to be something like that. From both perspectives it is so impractical now and any extension is at least a couple of years away.

You've done an epic job in this thread of twisting peoples arguments and dodging questions.

37 minutes ago, Stradling said:

Jesus, I guess in some peoples minds there are two types of teams, "All in" and "Team Rebuild", this team is neither.  

There are three types of teams. Those that are trying to win now, those that are trying to win later and stupid teams making half-assed attempts to do both. Which category do we fall into?

39 minutes ago, Stradling said:

 ...when you don't have players to trade to really improve the system?

That is totally untrue.

6 minutes ago, Troll Daddy said:

It doesn't really challenge anything ... the game hasn't changed. Just more food for thought. 

As great as this fortune cookie insight is, it literally challenges what you had just posted.

4 hours ago, eaterfan said:

Every pitch is recorded. It uses stat cast to track each pitch. It's then compared to other pitches in the same location and seen how often is called a ball and how often is called a strike. 

Getting a pitch called a strike more often is better. I know a lot of you don't believe in stats, but they show his tend to do better when ahead in the count and pitchers tend to do better when the hitter is behind. 

Obviously getting a strike on a pitch that is called a strike 90% of the time isn't as valuable as getting one that's called 10% of the time. It's factored into how much credit the catcher gets. 

Good explanation. When you put it like this, it's silly that the statistic is at all controversial. Many people just don't like it because they don't understand it or because it challenges their previously held views and they don't want to change.

7 hours ago, Stradling said:

i don't think it's controversial but every team has short term guys.  We haven't mortgaged any type of future to acquire these guys and your entire premise is that these guys were acquired in some kind of win now philosophy.  My opinion is they were acquired because the cost was low, they felt they were better than the alternative we had, and it's nice to have actual major league players.  We don't have a ton of farm talent to trade for long term solutions.  My point with Myers is he was the reason we acquired Nolasco, because they view Myers as a long term solution but to get him they had to bring Nolasco as well.  

Again, this isn't complicated. Somebody said we hadn't recently acquired any stopgaps or guys whose purpose was to help us for 2017. I pointed out players who undoubtedly fit that category. I don't get how you've interpreted all this extra meaning from a four word post.

58 minutes ago, ukyah said:

WAR is a highly flawed stat and needs serious reconfiguration before it's going to be the end all be all metric that many want it to be. it leads to some serious overvaluation of some average players and some drastic undervaluation of some good players.

Can you elaborate on this? Is your issue with the methodology or the principle? I obviously disagree, strongly, but I'm curious about your reasoning.

59 minutes ago, Stradling said:

29, Gott, Myers, $1 million. 

What the hell is your point? He said we hadn't acquired any stop gaps or players unlikely to fit into our long term plans - I pointed out examples to the contrary. I'm absolutely lost as to how it's controversial that those guys fit that category.

2 hours ago, Vegas Halo Fan said:

I don't agree with this at all. The recent acquisitions have been guys who will likely be here well beyond 2017. "Win now" mode, to me, means that we are picking up guys with the intention of helping us for the upcoming season, but they don't really fit into our plans after that. I haven't seen any pure stopgaps in our recent acquisitions.

Espinosa, Escobar, Nolasco, Bailey...

1 hour ago, Erstad Grit said:

Im suspect of the stat if it still says iannetta is good.  

It says he's terrible.

16 minutes ago, nikkachez said:

Everyone seems wary of it because they don't know what it is.  

This x 100.

It reminds me of WAR in the 2000s or deserved run average now.

15 minutes ago, ukyah said:

i have serious doubts as to the validity of the "pitch framing" stat, much less it's accuracy. i would be interested in hearing an educated argument for it.

i'm willing to believe in the base principle, but i have a very hard time discounting the importance of the caught stealing percentage, while fully acknowledging that it's far from an individual statistic, meaning the pitcher and receiving fielder play a part in it.

This is a long read and a couple of years old, but this will address much of what you're saying - http://www.hardballtimes.com/the-state-and-future-of-pitch-framing-research/

Choosing to ignore pitch framing is choosing to ignore a massive part of the game. Each ball or strike has real value and, as the article points out, it's not hard to see how a catcher can be worth 20+ runs per year through framing. One framed strike per game will end up being worth that much. It's practically impossible to generate that much value from a catcher's influence on the running game or really any other part of their game.

As for the accuracy of the statistics, the year-to-year correlation has been pretty strong. Even though it has declined in recent years, the year-to-year pitch framing correlation is still higher for the framing rungs metric than it is for a lot of widely accepted statistics. Put another way, 'Year 0' framing runs is a better predictor of 'Year 1' framing runs than 'Year 0' batting average is for 'Year 1' batting average. When there is an ongoing year-to-year correlation in a metric, it is hard for me to accept that it isn't accurately reflecting a skill.

5 minutes ago, nikkachez said:

I don't think Perez has a spot long-term. Just another thing Eppler's going to have to piece together. 

I'm starting to agree with this. His 2016 was awful - he was a Mathis-like hitter and an average defender. Catchers like that aren't hard to find. To have a future he's going to need to come much closer to matching his 2015 offensive performance but I'm guessing his leash is pretty short.

One strange thing about Perez is his AAA catching stats, before coming here, were pretty good. He was +6.2 in 2013, +8.6 in 2014 and +2.7 in 2015. But he has been below average in both his MLB seasons. Maybe that ability is still there and with some good coaching he can get it back.

3 minutes ago, Troll Daddy said:

Prove it with details for every DH

thanks!

Umm, I already did. Like everything else, you just seem to lack the ability to comprehend it.

For the record, Pujols drove in 16.7 per cent of the runners on base for his at bats - among DHs, that rate was behind Ortiz, Butler, Beltran, Morales and Encarnacion. Among all hitters with at least 300 plate appearances, that was the 53rd best rate in baseball.

8 minutes ago, Troll Daddy said:

Yada yada yada

Facts suck, right?

3 minutes ago, AngelsLakersFan said:

Some of the earlier work in this field had the number of runs affected much higher. I wonder what adjustments have been made that move the results closer to 0?

There has been a bit of research out later suggesting that the field is narrowing. The worst framers aren't anywhere near as bad anymore, which of course drags the average level up. I expect we'll see this trend continue. At most teams have been monitoring framing for a decade, so the focus on improving it is pretty new. The longer that goes on, the more condensed the field should get.