LambeauLeap1250 WSSP


  
Go to page Previous  1, 2, 3, 4, 5  Next  [ 83 posts ]  New Topic   Add Reply

Is Fangraphs a reliable website? Project 2019 Brewers to finish last?

Author Message
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#21

Posted: January 02, 2019, 10:02 PM Post
User avatar
Posts: 7327
Fear The Chorizo said:
jerichoholicninja said:
At this point, Ben Gamel and Claudio are the only additions the team has made that will likely impact the major league roster. We've lost our best SP and a starting infielder and replaced neither, and last year relied on Yelich and a bullpen full of guys playing out of their minds. If the season started today I could see this being a .500 team.

I'm not trying to be a Debbie Downer, I'm just pointing out Stearns hasn't gotten to work yet, so I can't blame anyone for looking at the current roster and not seeing many wins there.


Fangraphs' 2018 brewers record projection, after cain and yelich were added - 79 wins. Actual result - 95 wins in 162 games

Fangraphs' 2017 brewers record projection - 73 wins. Actual result - 86 wins.

Based on their recent track record, a 77 win projection means the crew may threaten 100 w's without adding anyone else


Fangraphs may have its value, but this proves that it's record prediction algorythm is broken.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#22

Posted: January 02, 2019, 10:27 PM Post
Posts: 3677
Fairly easy to project that Christian Yelich will not hit like steroid era Bonds in September of 2019. Difficult to project whether other players will pick up the slack, whether Jimmy Nelson will be productive, how valuable Burnes & Hiura & Dubon will be, and whether our relief aces will regress or not.

The David Stearns era: Controllable Young Talent. Watch the Jedi work his magic!


 Top
 
Quote   Reply 
Online  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#23

Posted: January 02, 2019, 10:57 PM Post
User avatar
Global Moderator
Posts: 5934
Good point, and if production wasn’t already tough enough to predict, correctly projecting player usage is a nearly impossible task. Right now these projections have Hernan Perez with 513 plate appearances, Cory Spangenberg with 302 PA, Eric Thames with 211 PA, and Tyler Saladino with 191 PA. I will take the under on those four hitters combining for 1,200 plate appearances.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#24

Posted: January 03, 2019, 8:19 AM Post
Posts: 3077
Fangraphs takes into account player values but they seem to struggle with the massive impact of Counsel in the totality of our performance.

I’ve never seen a baseball manager get more out of a roster than he does.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#25

Posted: January 03, 2019, 8:41 AM Post
Posts: 1947
CrewFanBrewMan69 said:
I think it definitely has a blind spot for us because how we are built and how we win is a little different. Last year required a crazy bullpen year and defense, those are volatile and hard to predict year to year and it’s probably hard to sustain the bullpen success we’ve had (in the projections eyes). Our rotation is also weak in their eyes, partly because the guys who have major league innings are league average or so and/or guys with only minor league track records. Also our defense/shifting was so good that it probably won’t project us to be as good there again because either it’s not likely to replicate or has not adapted correctly to account for the shifting. We also had like 14 wins out of Yelich and Cain, I don’t think it’s crazy that the projections don’t see that happening again and the rest of the offense is guys with not huge samples or impressive ones.


I agree. I mean, what's the FG projection for manager/gm/analytics WAR? They're milking a lot of extra wins out of these guys. Not a lot of guys who contributed the last 2 years are coveted individual talents if you think about it, so I don't think FG's pessimism about the sum of individual talent on the team is actually that far off. It just misses the part of their value that goes with being able to fit into the larger strategy they're employing. That includes having options, being a good fit in a platoon, being able to handle long relief, and being a versatile defender.

As for next year, I think it's a combination of uncertainty about the young pitchers and some possible regression from guys like Aguilar, Yelich, Chacin, Cain, and Jeffress. I personally think Burnes, Peralta, and Woodruff will help them have a pretty strong staff, but I seriously doubt they'll have the most all-stars in the NL once again next year. I'm kind of bracing myself for a temporary step backwards, but I support them wanting to keep Hiura and their young pitchers instead of trying to keep pace with some of the other moves teams are making for the short term.


 Top
 
Quote   Reply 
Online  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#26

Posted: January 03, 2019, 9:26 AM Post
Posts: 4267
Eye Black said:
Good point, and if production wasn’t already tough enough to predict, correctly projecting player usage is a nearly impossible task. Right now these projections have Hernan Perez with 513 plate appearances, Cory Spangenberg with 302 PA, Eric Thames with 211 PA, and Tyler Saladino with 191 PA. I will take the under on those four hitters combining for 1,200 plate appearances.


This is the same recurring problem with their projection methodology - it does a terrible job of factoring contributions from good roster depth that can produce better than what the projection system indicates when put in favorable situations. I remember diving into it last offseason, and they had a full season's worth of DH ABs factored in for a combination of players for the Brewers and other NL teams...problem is they maybe play 12 games where they need to use a DH. Their team record projections are literally trying to fit square pegs in round holes, and they are totally unreliable in trying to assess how the group of individual players fits together as a unit to win games.

Seems like even their staff spend more time discussing the reasons why their team record projections are flawed, particularly for rosters like the Brewers who lean heavily on almost their full 40 man roster during a season and often add pieces that fit their needs using all available talent acquisition approaches - maybe they should save themselves a little time and just stick with projecting individual players and scrapping the team record projections.

In 2018, many Brewers who heavily contributed to the 2017 team that won 86 games regressed - yet they wound up winning 95 games anyway. I don't buy the argument that they should be projected for significantly fewer wins in 2019 using the same "regression" argument - the roster is deep and talented enough where it's too difficult to project where all the production leading to winning games will come from. Just because Yelich won't spend all of 2019 hitting like Ted Williams doesn't mean the Brewers will automatically win almost 20 fewer games than 2018.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#27

Posted: January 03, 2019, 9:48 AM Post
User avatar
Posts: 1194
Fangraphs has a wealth of data and they do a great job a organizing data and putting it in one spot. They also have a group of pretty talented writers. The projection systems are just that, systems. Plug in the historical stats and let it do it's thing. Taking the individual WARs from those projections might give an indication on individual talent per team but it doesn't do a very good job of accounting for things like platoons, depth, and situational adjustments.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#28

Posted: January 03, 2019, 9:55 AM Post
User avatar
Posts: 2097
Fear The Chorizo said:
Seems like even their staff spend more time discussing the reasons why their team record projections are flawed, particularly for rosters like the Brewers who lean heavily on almost their full 40 man roster during a season and often add pieces that fit their needs using all available talent acquisition approaches - maybe they should save themselves a little time and just stick with projecting individual players and scrapping the team record projections.

I'm not sure they should even go back to the individual projections! Just look at the steamer projection for Chacin vs the previous actuals:

2016 1.7 WAR (1.181 WAR/100IP)
2017 2.4 WAR (1.333 WAR/100IP)
2018 2.6 WAR (1.353 WAR/100IP)
2019 1.2 WAR (0.642 WAR/100IP)

This is just one example of why I shake my head in disbelief at these projection systems. For three years with 3 teams he has achieved a greater than 1 WAR/100IP, but in 2019 he's going to DECOMPRESS to 0.642 WAR/100IP. That's less than half what he's accomplished just the last 2 years... It's these types of complete head scratching projections that make me wonder why I can't sell my snake oil...

JosephC said:
Stearns probably had no interest in getting a C because the Brewers need a C. It makes much more sense to trade for 3B when it's not needed, and then move the other 3B to 2B, then trade for a 2B, but since the 3B is now at 2B, then the new 2B goes to SS


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#29

Posted: January 03, 2019, 10:01 AM Post
User avatar
Posts: 7327
Fangraphs predicts every Brewers starter to have an ERA in the high 4s. Plus pretty much every offensive performer regressing.

That, my friends, is proof enough that it is far from reliable. In fact, looking at the level of regression that they point to several Brewers players having even leads me to believe that there is an anti-Brewer bias. Case in point ... Chacin put up a very solid season for the Padres in 2017, followed by a terrific season with the Brewers last year, in a park tougher to pitch in and in the midst of a nerve-wracking pennant chase. Fangraphs has his ERA projection jumping by more than a full run. I am struggling to understand why.


 Top
 
Quote   Reply 
Online  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#30

Posted: January 03, 2019, 10:29 AM Post
User avatar
Global Moderator
Posts: 2205
Joey Meyer Bombs said:
Fangraphs predicts every Brewers starter to have an ERA in the high 4s. Plus pretty much every offensive performer regressing.

That, my friends, is proof enough that it is far from reliable. In fact, looking at the level of regression that they point to several Brewers players having even leads me to believe that there is an anti-Brewer bias. Case in point ... Chacin put up a very solid season for the Padres in 2017, followed by a terrific season with the Brewers last year, in a park tougher to pitch in and in the midst of a nerve-wracking pennant chase. Fangraphs has his ERA projection jumping by more than a full run. I am struggling to understand why.


FanGraphs isn't predicting every Brewer starter to have an ERA in the high 4s, Steamer is. Fangraphs just hosts those projections.

Guerra (3.88), Nelson (4.20), Peralta (4.39), Woodruff (4.44) & Burnes (4.52) are all projected in the high three to mid four range.

The reason why Steamer projects regression for Chacin is because historically pitchers of his age/profile get worse.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#31

Posted: January 03, 2019, 10:54 AM Post
User avatar
Posts: 2097
sveumrules said:
The reason why Steamer projects regression for Chacin is because historically pitchers of his age/profile get worse.


Except the regression predicted for Chacin is falling off a cliff. Anibel Sanchez (bbREF top 10 match by age 30 to chacin) who is 4 years older has a better projection (significantly) than Chacin. Tanner Roark, who has 3 continuous years of actual regression in his performance and is a year older than Chacin, all of a sudden for 2019, is projected to barely get worse! Really? I won't say theirs a definite Brewers bias, but it's clear the adjustable metrics they feed into their projection system is caca...

Edit: Kyle Gibson (same age) tiny dropoff. Mike Fiers (2 years older) tiny dropoff. Jordan Zimmerman (1 year older) small dropoff... I'm sure I can keep going on... They Predict Chacin to be significantly worse in his age 31 season compared to players who are similar or older - that ain't no age/profile adjustment...


Last edited by xisxisxis on January 03, 2019, 11:11 AM, edited 1 time in total.

JosephC said:
Stearns probably had no interest in getting a C because the Brewers need a C. It makes much more sense to trade for 3B when it's not needed, and then move the other 3B to 2B, then trade for a 2B, but since the 3B is now at 2B, then the new 2B goes to SS


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#32

Posted: January 03, 2019, 10:57 AM Post
User avatar
Posts: 7327
sveumrules said:
The reason why Steamer projects regression for Chacin is because historically pitchers of his age/profile get worse.


I get that. But a rise in ERA of 1.18 runs per 9? For a 31-year old? Nope. Also, Chase Anderson (age 31) followed up his breakout 2017 with a 2018 that wasn't as good, but still saw a very respectable 3.93 ERA. But yeah, he'll put up an ERA of 5 this year. I get that Anderson will never be a world-beater, but he is a respectable mid-rotation starter, not a replaceable 5-ERA guy. But you look at the Cubs, and the projections indicate that Quintana (age 30) is going to get better, while Hamels (age 35) will continue to pitch like a #2, as if his trainwreck 1st half with the Rangers last year isn't even a blip on the radar. Also, Darvish, coming off an ok 2017 and a lost 2018 (and at 32 years old), is suddenly going to be Darvish again. At least they finally have Lester, at age 35, slowing down.

Also, Fangraphs, Streamer ... whatever. If Fangraphs hosts it, they own it.


 Top
 
Quote   Reply 
Online  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#33

Posted: January 03, 2019, 11:16 AM Post
User avatar
Global Moderator
Posts: 2205
I'd imagine that projection systems are going off of FIP much more so than ERA, since FIP has a higher year to year correlation than ERA.

The last three seasons Jhoulys has FIPs of 4.01, 4.26 & 4.03. Projection: 4.68

The last three seasons Chase has FIPs of 5.09, 3.58 & 5.22. Projection: 4.96

The last three seasons Quintana has FIPs of 3.56, 3.68 & 4.43. Projection: 3.94

The last three seasons Hamels has FIPs of 3.98, 4.62, 4.49. Projection: 3.94

The last three seasons Anibal has FIPs of 5.06, 5.33 & 3.63. Projection: 4.21

The last three seasons Roark has FIPs of 3.79, 4.13 & 4.27. Projection: 4.49

The only projection that seems wonky to me out of those is Hamels, but he also was a top 7-8 pitcher in MLB for a decade from 2006-15 so maybe Steamer is somehow cooking that in.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#34

Posted: January 03, 2019, 11:43 AM Post
User avatar
Global Moderator
Posts: 10123
Steamer, Zips, Bill James, PECOTA are all projection systems. Fangraphs does not do projections - they list other people's. They are all about the same in regards to their methodology with some minor differences.

Like other projection systems, Steamer uses past performance and aging trends to develop a future projection for players. It also uses pitch-tracking data to help forecast pitchers. On Fangraphs, the projections are updated daily and predict each player's numbers over the course of the remainder of the season.

Obviously, no one is claiming that every one of Steamer's predictions will come true, but it is widely regarded as one of the most accurate predictors in the industry.


http://m.mlb.com/glossary/projection-systems/steamer

"Dustin Pedroia doesn't have the strength or bat speed to hit major-league pitching consistently, and he has no power......He probably has a future as a backup infielder if he can stop rolling over to third base and shortstop." Keith Law, 2006


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#35

Posted: January 03, 2019, 12:52 PM Post
User avatar
Posts: 737
I don't think Steamer knows how to factor defensive positioning and effective use of shifts in their pitching projections. Remember, in 2018 a Brewer pitcher with the Orioles defense behind him would have had an ERA around 1.5 runs higher. So basically a pitcher could throw the same pitches to the same batters, generate the same swings and misses, but have dramatically different results from batted balls in play. In Milwaukee the batted balls went to the gloves of defenders much more often than where Baltimore positioned it's defenders. Chacin could have thrown the exact same pitches in 2018 with the same velocity and location and have had an ERA pushing 5.00 with teams like Balt, Phi, NYM playing behind him. These projection models have a really hard time with that.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#36

Posted: January 03, 2019, 2:07 PM Post
Posts: 8289
Projections miss too many variables, so I pay little/no attention to them. First, most teams really only have a handful of players where you can feel comfortable making a fairly exact projection. Too many inconsistent players, young players, etc. Use of defensive shifts was another good example.

Biggest issue though is they don't take into account how a player is used. For example, using Shaw almost exclusively vs RHP would give you different numbers than vs all pitching. But where this really goes off the rails is use of the pitching staff. Last season Chacin, Miley, Suter, etc. destroyed their projections due to how they were used. Stats are projected using the old definitions and use of starting pitchers. It doesn't take into account limiting exposure the 3rd time through the line-up, where ERAs really bloom.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#37

Posted: January 03, 2019, 2:25 PM Post
User avatar
Posts: 737
FVBrewerFan said:
It doesn't take into account limiting exposure the 3rd time through the line-up, where ERAs really bloom.


Yes that's another biggie. They haven't caught up to the way the Brewers deploy their pitchers. And in 2019 they'll be exposed to repeat trips even less.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#38

Posted: January 03, 2019, 6:51 PM Post
User avatar
Posts: 3302
That projection seems fairly ridiculous, but I do think it is fair for Brewers fans to hedge their 2019 expectations a bit given the team's current weaknesses on paper. The projection system means very little, but so does simply saying "we did well in 2018 so what is going to change?"

Projection systems have given up on Arcia (hopefully they are wrong), and our current starting 2B is probably Perez. Aguilar is a risk to plummet back to earth and Pina at his best his below average. That means our entire infield except Shaw (who has his platoon issues) could be bad. Meanwhile the rotation has obvious faults.


 Top
 
Quote   Reply 
Offline  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#39

Posted: January 04, 2019, 12:51 AM Post
User avatar
Posts: 60
Fangraphs is pretty good at projections from the view point of no player has a break out ever and the one that did last year that's nothing more then a fluke.

They try to put them all in a neat little box where no player will hit more then 40 Home Runs or have 40 Stolen Bases. Where no player hits much better then .300. Although there are many that do every year but for the players that do fit within the box those positions can be some what good in most cases.

But overall it's a usefull website if you understand they can be a bit flowed and if you question what don't make sense.


 Top
 
Quote   Reply 
Online  Re: Is Fangraphs a reliable website? Project 2019 Brewers to finish last?
#40

Posted: January 04, 2019, 7:30 AM Post
User avatar
Posts: 9780
SRB said:
That projection seems fairly ridiculous, but I do think it is fair for Brewers fans to hedge their 2019 expectations a bit given the team's current weaknesses on paper. The projection system means very little, but so does simply saying "we did well in 2018 so what is going to change?"

Projection systems have given up on Arcia (hopefully they are wrong), and our current starting 2B is probably Perez. Aguilar is a risk to plummet back to earth and Pina at his best his below average. That means our entire infield except Shaw (who has his platoon issues) could be bad. Meanwhile the rotation has obvious faults.


Man, you just bummed me out...

All true though, which is pretty scary. We haven't done a thing yet to address the weaknesses that you mention.

We caught lightning in a bottle with Miley and Chacin last year, can that even happen again?


 Top
 
Quote   Reply 
Display posts from previous:  Sort by  
Go to page Previous  1, 2, 3, 4, 5  Next  [ 83 posts ]  New Topic   Add Reply
  


Who is online


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search this forum (phpBB search):
Jump to:  
Search entire board (Google search):
Google
Powered by phpBB® Forum Software © phpBB Group
Test