Neil Weinberg FanGraphs Q&A – 10/22/14
2:42 |
I’m @NeilWeinberg44 on Twitter if you want to get in touch that way, too. |
3:02 |
: Alright, let’s get started.
|
3:03 |
: Remember that we have live blogs during each world series game if you need more of this kind of thing. I will likely be hosting Saturday/Game 4.
|
3:04 |
If AL teams were given the option to forego having a DH (their pitcher must bat) in exchange for being able to play an extra fielder (giving them 10 men in the field at once) would it be worth the tradeoff to do so? |
3:04 |
: Let’s do some quick research
|
3:05 |
AL P: .101 wOBA |
3:08 |
: Now let’s assume that they would only get 2 AB per game instead of 4 and PH would be around .320. So let’s call it a .210 wOBA compared to .327…. then let’s assume that batting more often would make them a little better, call it .227 for ease. So it’s .100 points of wOBA. That’s like 50 runs of lost offense. Can’t imagine the fielder could bring that back, even factoring in the compound effect.
|
3:08 |
what contract do you think brandon mccarthy gets? I was just doing the contract crowdsourcing and having a tough time. |
3:08 |
: Probably 2/25 or 3/30.
|
3:09 |
Will there be /600 PA projections for all projection systems? |
3:09 |
: Probably, but I don’t know for certain.
|
3:09 |
Doesn’t scaling statistics to the league average of each year ignore the fact that the quality of competition is highly variable between eras? Are Babe Ruth’s home runs really more impressive than say, Bonds’ just because no one else was hitting them at the time? Ruth wasn’t playing against a worldwide talent pool that included blacks, latinos, etc.. |
3:11 |
: We can’t really say. We want to compare a player to his era, so hitting .400 wOBA in 2014 is way better than in 2000 given the offensive environment, but is .400 wOBA in 2014 the same as a .400 wOBA in a similar environment in the 1930s or whatever….obviously not. But it’s kind of hard to nail that down. Babe Ruth wouldn’t be BABE RUTH today, but I’m not sure how much that matters.
|
3:11 |
Will per 600 PA projections be added to player pages? |
3:11 |
: I believe so?
|
3:11 |
From a theoretical perspective do you think park effects impact all players equally? I tend to think that great players are great regardless of environment, but marginal players may need every advantage to be successful. |
3:13 |
: Players have different styles, so one size fits all park factor isn’t going to be perfect. Think about extreme pull hitters, for example. It doesn’t matter how the ball carries to the opposite field for them, but that might have a real impact on the PF.
|
3:13 |
I know this has been answered before but I couldn’t for the life of me find anything. How is historical UZR calculated? How reliable is it? |
3:15 |
: There is no historical UZR. Our old defensive data is based on Total Zone, which is an approximation based on play by play data when available and box score type data at the minimum. So it’s mostly, like UZR except we don’t know how hard the ball was hit, the location beyond the player it was hit to, the weather, etc. So it’s worse than UZR. Over a career, it works well enough, but for one season it’s basically a hint. A +10 TZ means he’s more likely good than bad, but it doesn’t tell you much more
|
3:15 |
Does a larger outfield, like the K, really lead to more hits just because there’s more grass? Parks with smaller dimensions create hits (HR’s) that would sometimes be fly outs I would think. |
3:16 |
: More space = more hits, typically speaking. But you can’t ignore the weather and atmosphere for any park either. The way the ball carries and the amount of space both matter
|
3:16 |
How many years is someone going to give Hanley? |
3:16 |
: Five?
|
3:17 |
Anytime anyone asks about the guts of some metric on Fangraphs, whoever is running the chat says that’s not their department. Who actually runs the guts of these things (aside from UZR which I know is from MGL and DRS which is from BIS)? |
3:18 |
: Certain stats aren’t FG created (such as UZR, UBR, etc), so the writers don’t know all the technical details. David Appelman handles most (all?) of the actual coding, with input on the creation and adjustments from people like Dave and some of the more technical stat people on staff. Or sometimes they talk with outside people like Tango, MGL, etc.
|
3:19 |
: For example, no one at FG invented wOBA, but FG implemented it. wRC+ was sort of created as a more accurate OPS+ and that came out of conversations with Tango, for example.
|
3:20 |
If Barry Bonds was coming off of his 10.5 WAR 1993 season and was set to hit free agency at age 29- in 2014 what contract would he get? Miguel Cabrera Esque? |
3:21 |
: Bigger. 10/400? $35MM per season would be the floor.
|
3:23 |
Are scoops incorporated into 1b UZR? Is there any correlation year to year? |
3:24 |
: Scoops are not in UZR. It is a skill, but the talent spread is pretty small. I think we’re talking about 5-7 runs between worst and first per season. Small blindspot, but only because the data wasn’t available. Probably gets folded in at some point.
|
3:24 |
If baseball went to an automated strikezone, do you think teams would then be ok with catchers back there who are not great receivers? |
3:25 |
: To some extent. Receiving also influences how well you hold runners, but it would change the game. Also, pitchers have preferences about how a guy receives a ball. I don’t know how much that would change their real performance, but it’s something
|
3:25 |
: You wouldn’t just put stone hands behind the plate, but obvious it would hurt Lucroy’s value relative to the average catcher
|
3:25 |
Thanks to your help I built a cool WAR calculator for hitters. Are you interested in taking a look? |
3:26 |
: Gladly. Tweet me a link if it’s public or if it isn’t ask me for my email on Twitter. I have an old one build, but it needs to be updated. Planning to release it again once I get to updating pitcher WAR too.
|
3:26 |
Also, where is player of the year results? |
3:27 |
: Delayed, briefly!
|
3:27 |
Why is the rank order for RE24 and REW sometimes different? I always thought it was because the runs-to-wins conversion included a park adjustment, but I’ve seen you state in previous chats that a park adjustment is already included in RE24. So what is it, then, that sometimes results in a player having a higher REW than another despite having a lower RE24? |
3:28 |
: Wait, really? I don’t see why that would happen, although I don’t really know if we do anything special to create REW. Can take a look.
|
3:29 |
How do you think Jay Bruce will play next year? Bounce back well or below his past performance? |
3:29 |
: Better than this year, but he’s already peaked. 110 wRC+?
|
3:29 |
Do you buy the notion that outfielders from Japan have found more success than infielders because “the MLB game is faster” in the infield and the imports can’t keep up? |
3:30 |
: That’s possible. I don’t know enough about the NPB to say for sure. Also not sure if the success rates are really that different or if we’re just blinded by a couple of very successful OF
|
3:31 |
If you assume MLB was pro-robot umpires, how ready is the technology to implement that? Could it be a possibility for next year (if the MLB wanted it)? |
3:32 |
: I believe so. They would have to figure out exactly how to deal with communicating the call and decide on an official strike zone, but the technology can locate the pitch accurately and quickly enough
|
3:32 |
Are park factors from later years incorporated into previous year figures? Ie are Colorado player stats from the first year or two in Coors unreliable because they only use one or two years of PF figures? (Not to mention Mile High…) Can we expect this year’s WAR numbers to change if PFs in future years alter the multiple-year averages? |
3:34 |
: Park factors are regressed more if there is less data. So new parks, for instance, are treated as average until they demonstrate otherwise, if that makes sense. Park factors are the easiest thing in the world to explain conceptually and tough to do in practice. Will do a thing over the offseason about how we do them
|
3:34 |
![]() |
3:34 |
: This is not me. He likely has more money that I do, but I have more hair!
|
3:34 |
Do you think we’ll ever see a legitimate two-way player? |
3:35 |
: Like pitcher and hitter? I don’t like to say never, but not in the current paradigm. Maybe in 40 years
|
3:35 |
Are all of the “writers” for NotGraphs creations of Cistulli? Are all of the “writers” for FanGraphs creations of Cistulli? Is everything I perceive a creation of Cistulli? |
3:35 |
: If I said no, wouldn’t you wonder if my answer was just generated by Carson to throw you off the trail.
|
3:35 |
Seems like having four outfielders for your flyball starters or five infielders for your groundball starters would be pretty drastic |
3:37 |
: In reference to the earlier discussion, I’m just not sure there are 50 runs to save using this strategy, and that’s assuming you PH effectively and the pitchers learn to hit a little better, before factoring in an increase in injuries from hitting. You’d have to save two runs a week, basically. That’s hard to do.
|
3:37 |
Jeff mentioned yesterday that Bumgarner’s 3.6 WAR v. 2.99 xFIP over 217 IP could be explained by park and league adjustment. But isn’t xFIP already park-adjusted? |
3:37 |
: No
|
3:37 |
: Here, let’s look at Madison’s line.
|
3:39 |
: 217 IP, 3.05 FIP. 3.6 fWAR. Plus 23 IFFB, which go into WAR.
|
3:40 |
: Strasburg for example, 215 IP, 2.94 FIP, 4.3 fWAR (just 12 IFFB)
|
3:41 |
: So same innings, but basically the same FIP when you consider IFFBs. Why such a big gap?
|
3:42 |
: Giants have a 93 PF and Nats have 100.
|
3:42 |
: So Bumgarner’s FIP- is 89 and Strasburg’s is 79!
|
3:42 |
: Basically all park.
|
3:43 |
I feel bad for the Rockies. Their hitters appear to be at a disadvantage on the road adjusting to breaking pitches at normal altitude. It’s hard for them to attract pitching free agents while at the same time position players do not give them discounts just to play at altitude. Seems like a really tough environment to constuct a baseball team? |
3:43 |
: It is. I wonder if MLB had any idea this would happen when they put the team there, and if an indoor park would have helped.
|
3:44 |
Why doesn’t Michael Jordan have a Fangraphs page? Because he never made it to the majors? |
3:44 |
: Think we only have minor league player pages back to a certain year.
|
3:44 |
Ubaldo got 4Y/$50M. JDLR got 2Y/$25. How crazy is the market for bad pitchers? |
3:45 |
: Teams dream on the past. Only takes one.
|
3:45 |
If you had to guess, how many fielders would it take before we start seeing diminishing returns by adding another fielder? |
3:46 |
: I think probably six infielders and five outfielders, so 13? Although, you could probably push that to 15 if your goal was to put three guys on the bases and tell them not to field anything that makes them move their feet.
|
3:46 |
Which is your preferred basis for pitcher WAR: FIP, xFIP or SIERA? |
3:47 |
: FIP.
|
3:47 |
UZR takes weather into account? |
3:49 |
: Each park’s average temp/altitude/etc are included in the UZR park adjustment, I believe.
|
3:49 |
Asdrubel Cabrera a fit for the Tigers? |
3:50 |
: Probably not, unless Iglesias has a setback.
|
3:50 |
Would you support a penalty for teams that roster steroid abusers? It seems like franchises (and the league) both profit from players drug-bolstered stats (winning more games, selling more tickets) … then profit again when they get to write off aging ex-roider’s salaries (think A-Roid). Seems hypocritical. |
3:50 |
: In theory, I think the team should have to pay the salary to a charity, or disperse it to the rest of the league or something.
|
3:51 |
10/400 for barry, and they didnt even know he wasnt done putting up 10 WAR seasons…haha unbelievable. |
3:52 |
: He put up 86 WAR over the next ten years after our hypothetical. So a conservative estimate of his value if that happened starting today would be $500 million.
|
3:52 |
: As always, WHAT.
|
3:52 |
Jon Lester Over/Under $147M that Greinke and Hamels got. Hamels was extended, inflation is happening, and the pitching market is a joke this off season. I mean he isn’t as good as them either. |
3:53 |
: I think that’s actually about what he gets.
|
3:53 |
Is there a chance Chapman becomes a starter? |
3:53 |
: Very small.
|
3:53 |
As someone who semi-regularly watches NPB, the infielders are generally awful. I don’t know why that would be the case, but it’s shocking how bad the infields are given how prevalent contact-type hitting and small ball tactics are in NPB. |
3:54 |
Yes, for example: Ortiz 39.04 RE24, Cano 38.00. Cano 4.35 REW, Ortiz 4.19. Look at the leaderboards and you’ll find lots of other little examples like that. |
3:54 |
: I will check into this.
|
3:54 |
Do you see any value in using park factors by handedness for park adjusted stats(either on the pitching or hitting side), or would that do more harm than good? |
3:56 |
: Problem is that you’re then making an assumption that just divides the players into two bins. Slightly more information, but then you would want to know the opposing pitchers and spray tendencies and it gets really complicated quickly. There’s room to improve,but is the end goal a park factor for every pitch?
|
3:56 |
How do you think your tandem starter idea for the Tigers would have worked in the playoffs this year? |
3:57 |
: I think it would have helped, but I don’t know if it would have helped them win three games. There was no way to predict a 33 ERA bullpen.
|
3:57 |
Is Masterson the next Pittsburgh reclamation project? I think he would be a great fit there |
3:57 |
: Makes some sense. Lot of teams would be wise to gamble on him
|
3:57 |
Why do park factors vary so much year to year? |
3:58 |
: They don’t vary that much, but it’s only 81 games. That’s not actually a huge sample.
|
3:59 |
By the way, that would be genius of an already genius Cistulli if all the NotGraph writers were in fact pseudonyms of Cistulli |
3:59 |
: I actually have no evidence that could disprove your hypothesis.
|
3:59 |
So then, what’s going into PF that isn’t in xFIP? Just BABIP, since xFIP is already correcting for a depressed HR environment? |
4:00 |
: xFIP regresses HR rate to league average, not to park average. HR aren’t the only reason parks play differently. It’s hard to hit bombs in Minnesota, but there are a decent number of runs scored there because doubles and triples are way ore common
|
4:01 |
As someone who doesn’t watch a lot of NL baseball, why is SF’s park so pitcher friendly? Just the wind off the water or is there another issue? |
4:01 |
: Large dimensions, but it’s pretty cold there relative to the rest of the league in the summer, plus the wind/water. Also the high wall in RF.
|
4:02 |
: Quick break. Back in a minute and will go to about 430p. First time I’m going to get nowhere near all of the questions, but dump a few more in and we’ll see what happens
|
4:04 |
re: indoor park in Denver. I love this idea, but Coors is one of the more beautiful parks in the country, I don’t know if I as a fan could bear to see them go to a dome. Though I’ve often thought that a pressurized dome might be the answer to the Rockies problems |
4:05 |
: Right. Trade off between the view and the impact on the team.
|
4:05 |
Why do we assume the Rockies’ hitters are at a disadvantage on the road rather than an advantage at home? with the extreme PF isn’t it possible they’ve adjusted to playing at Coors way more than their opponents and have figured out ways to exploit it (like how the Sox LFs have adjusted to the Green Monster)? |
4:06 | : Purple Row did a piece about Coors earlier in the year. |
4:07 |
Why can’t we calculate park factors just using the physical characteristics of each park? Dimensions are static, average humidity/temperature are all very easy to calculate. Why do we need to depend on the players’ stats at all? |
4:08 |
: We need to know how the park influences outcomes. Need the outcomes to do that. I think there are potentially some other ways to go about it, but you can’t just say here are thirty parks with these dimensions and air temp and assume you can figure out how they will play. Need to calibrate
|
4:08 |
What’s your preferred method of allocating ottoneu arbitration dollars: spread the love between several players? max out a single target player’s salary and hope the owner drops him? |
4:09 |
: I generally look for players who are on the keep/don’t keep bubble and go all in
|
4:09 |
Is it pronounced CIS-tul-li or cis-TUL-li |
4:09 |
: The second one, although now that I’m saying it out loud I can’t tell the difference
|
4:10 |
Actually jumping off BONKERS, I really would like to see some new mega-formulas to compete with WAR. Not to replace it, just to delve into some different methodologies. |
4:10 |
: If StatCast really dumps the data the way they say they will, we could really have some fun because everyone would be on the same playing field with data access. Would love to host a “best WAR” contest
|
4:11 |
Using player stats to determine park factor seems like needlessly introducing a confounding variable to me |
4:11 |
: You have to know how the dimensions/weather influence the numbers. Can’t do that without the numbers.
|
4:11 |
Is James Shield literally a line drive machine? I don’t have time to read the article. |
4:12 |
: Yes. You put quarters in and then there are line drives
|
4:12 |
True or false … the Tigers would be better of with Billy Butler and Denard Span over VMart and Rajai Davis |
4:13 |
: Hmmmm. That would be close. Would probably go with VMart/Davis, but it’s not a crazy question at all
|
4:13 |
LOL. I had the same response to that article and thought it might be a bit biased! Still not sure why the opposite isn’t potentially just as valid. Opposing players are at a disadvantage trying to adjust to Coors’ quirks, especially pitchers. |
4:14 |
: Other teams only have to play in Coors a few times, so they only have to adjust one or two times per year. Rockies have to do it a ton
|
4:14 |
Can you explain why early poor performance from a pitcher does not predict late poor performance? I would guess that poor performance early in the game leads to incurring the times through order penalty sooner than expected. In turn, later performance will also be worse because of the higher penalty. Why is that logic wrong? |
4:15 |
: Are you talking about MGL’s piece? He adjusted for TTOP in the projections.
|
4:15 |
If you had to guess, where does Lester end up? |
4:15 |
: Baltimore.
|
4:15 |
Thanks for chatting, Neil! I know this might not be your wheelhouse, but what are the rules regarding non-rostered guys like Christian Colon and what they can and can’t do around the dugout to help their team when they can’t actually play? |
4:15 |
: I think the rule is that they can not be on the field during the game.
|
4:16 |
Giants in 5 sound realistic? |
4:16 |
: Sure. Anything but Royals in 4 is realistic. Not a jerk response, but there is no outcome of any single series that would surprise me.
|
4:16 |
Given how much of pitching velocity is genetics/innate talent, do you feel that certain regions of the US are underscouted? Wouldn’t you expect distribution of origins for arms to be more evenly distributed than for position players (more of a learned skill even at lower levels)? It seems like both pitchers and batters share the same Southern US/California biases in distribution. |
4:18 |
: They don’t play as much ball in the cold weather states, so even if you scout them, you don’t know as much. Teams are risk averse like that, but velo/etc also requires training and in Minnesota the athletes focus on hockey and never really find out they have pitching talent.
|
4:18 |
Would having a couple parks with unusual dimensions like the Polo Grounds or Baker Bowl in the majors today make things more or less interesting? Seems like the league is cooler because Coors exists. Or would today’s savvier teams be too good at building a team around unusual home factors? |
4:19 |
: More! I love parks that play weird.
|
4:19 |
: I think the Coors problem of elevation is an issue because it affects the pitches, not just the flight of the ball, though
|
4:19 |
When will Wendy get her own chat? These things always feel like boys’ clubs, so I like/am inspired by what she does and it would be fun to have her do a chat. |
4:20 |
: Wendy has done a few chats and did a live blog this postseason. Don’t know if she has interest in permanent chats or not, but I endorse your positive reviews of her work!
|
4:21 |
The funniest part of that 10-year deal for Bonds is that it still wouldn’t include all of his 10+ WAR seasons in the remainder of his career |
4:21 |
: In theory that $400 million would be a bargain.
|
4:21 |
Your last name means “wine mountain.” Is that the opposite of a wine cellar, or one of those displays at the entrance to a wine and spirits store? |
4:21 |
: I have always assumed it was like a vineyard.
|
4:21 |
Who is your favorite player who is better in fantasy than he is in reality? |
4:21 |
: VMart
|
4:22 |
You are made commissioner of baseball and as a one-time event are given the power to force one owner/ownership group to sell their team. Who do you choose and why? |
4:22 |
: Marlins. Seems like they’ve poisoned the well there
|
4:22 |
I asked Jeff this the other day, but didn’t get a clear response: When I look at the Giants defense including pitchers, their Def is like +35 or something. When I click NP it becomes like -4. Is pitcher defense that heavy or inaccurate? |
4:23 |
: Pitchers have a huge positional adjustment
|
4:23 |
Would baseball be more popular if someone discovered a more entertaining way to gamble on it? If so, any ideas? |
4:24 |
: A friend and I used to gamble on the outcome of each at bat. Like 50 cents at a time, so winner would wind up with like $5 at the end. But I don’t think it would really help much.
|
4:24 |
If bumgarner’s era takes a hit b/c of park adjustments (sf pitcher’s park), does it get offset by the fact he pitched worse at home? |
4:24 |
: No. At least not from a calculation standpoint
|
4:25 |
Which formerly common (decades ago) prediction about the future of baseball looks dumber today: 1) everyone will play on turf/domes in the future or 2) everyone will use aluminum bats in the future? |
4:25 |
: Turf.
|
4:26 |
What’s your favorite all time movie? By the way, Your chat is tied for my favorite with Eno. I like his beer answers, but I like that you seem to answer all of my questions and i enjoy your responses. #mancrush |
4:27 |
: Thanks, I don’t usually look at the names until I post it. So that’s just a coincidence. A lot of the time I answer 85+% of the questions, but this week is flooded. Shawshank? Up?
|
4:27 |
It’s “pretty” cold in SF in the summer? hahahaha |
4:28 |
: I am not sure why this is funny.
|
4:28 |
RE: Park factors: and it’s not just weather and dimensions, either. Things like the quality of the hitter’s backdrop can make a difference. Don’t know how you’d account for something like that except in seeing how it actually shows up in the outcomes. |
4:28 |
: Yup. Good point.
|
4:28 |
Please expand on the Lester to Baltimore thing. Don’t the Orioles already have a crowded rotation with 5 decent guys + Ubaldo + Bundy? |
4:30 |
: O’s need to add pitching and have the resources to afford him if they want. Was a guess. Don’t feel strongly
|
4:30 |
Where are all the Mookie Betts trade offers this week? I’ll start things off: Angels trade Trout + Pujols & Hamilton for Betts, straight up. |
4:32 |
: FG Mookie talk has been quiet for a while! That’s one of those deals that theoretically makes sense from a value standpoint (given all the negative value that Pujols and Hamilton have and Trout no longer makes the minimum) but makes no sense because the Angels want wins, not efficiency right now. So it’d be 13 WAR for 3-4 WAR, even if the finances balance. Can’t take that money and buy a new Trout
|
4:32 |
Pablo or Alcides Escobar? |
4:33 |
: The context of this question is very important!
|
4:33 |
People in my town recently lined up to purchase tickets for Foo Fighters, in the middle of the afternoon on a work day. Is it 1999 again? |
4:33 |
: Heh.
|
4:33 |
Your chat is more fun than studying for the CPA examination. |
4:33 |
: Presumably not a high bar, but appreciate it.
|
4:33 |
Would baseball be improved if the players on the team that finished with the worst record in all the majors were sacrificed so that the crops would yield a good harvest for next year? |
4:34 |
: I would actually like to see some sort of penalty for the worst team. Not human sacrifice, but something!
|
4:34 |
Please confirm my fears as a Phillies fan: no team will trade for an ace (and pay the contract) and give up prospects, when they can just sign one in free agency. It’s gonna be a long decade is it not…? |
4:34 |
: Won’t get both things, but Phils eat some cash and there will be prospects for Hamels or a healthy Lee
|
4:34 |
“The coldest winter I ever saw was the summer I spent in San Francisco.” -Mark Twain |
4:35 |
: Ah…so this is why reading the name would help.
|
4:35 |
Would it be easier or more difficult to hit if you doubled the distance to the mound but guys could throw twice as hard? |
4:36 |
: Harder. Reaction time would be the same but presumably picking up the ball would be harder. But that is a question for an actual physicist and not me.
|
4:36 |
Are fake Mookie Betts trades enough of a meme in FG chats that people will continue submitting him even after he actually gets traded? |
4:36 |
: I hope so!
|
4:36 |
I think you mean “physician” |
4:36 |
: No I mean physicist!
|
4:37 |
: Alright, this was fun. Sorry for the questions I didn’t have time for. Dave and Jeff will be here during the game and I’ll be back Saturday to do the same, I believe.
|
Neil Weinberg is the Site Educator at FanGraphs and can be found writing enthusiastically about the Detroit Tigers at New English D. Follow and interact with him on Twitter @NeilWeinberg44.