Neil Weinberg FanGraphs Q&A – 11/12/14
2:33 |
: Hey all, queue is open and we’ll get going at 3pm! Priority to stat/data/FG questions, but anything is fair game. I’m @NeilWeinberg44 on Twitter if you’re looking for me. See you shortly.
|
3:01 |
: Alright, let’s consider important* topics (*-no guarantee of actual importance)
|
3:01 |
Does Steamer project PAs? I know the projected PAs on the player pages on manually set by FG staff. Say Steamer projects a player for 200 PAs and FG manually adjusts to 400 PAs. Are the rates stats kept the same and counting stats double? |
3:02 |
: Steamer projects PA, but FG also does custom playing time numbers. That said, Steamer is driving the rate stats 100%, so a .340 wOBA projection is a .340 wOBA projection whether we project 1 PA or 500 PA
|
3:02 |
How often do bugs like Nori Aoki’s -10 BsR pop up? |
3:04 |
: I don’t know the official answer, but I seem to notice one or two mistakes a year…usually they are easy to correct, but sometimes not. I’m not sure exactly why we haven’t updated the database on that one, but he’s supposed to be a slightly above average runner this year
|
3:04 |
Have you seen any research on the value of a roster spot? If a platoon and everyday player are projected to be equivalent, what is the added value of having the everday player? |
3:05 |
: Nothing comes to mind off the top of my head….there’s obviously an advantage, but I’m not sure what the order of magnitude is. Kind of depends on how many roster spots are affect. At some point, the every day player gets more important as available roster spots appraoch 0
|
3:05 |
Is there an endpoint for baseball analysis? Is there a point where we say “we can’t squeeze another drop of data out of this here machine” or will we always be able to develop new and improved methods? |
3:07 |
: I think there will be a point at which we won’t be able to accurately measure anything else, but the beauty of baseball is that 1) there’s a new slog of 2400 games every year and 2) there’s enough randomness to keep it interesting, even once we have all the data. Although, I don’t think that day is on the immediate horizon
|
3:07 |
Do you think there will ever be a day when we have complete pbp data for more seasons earlier than 1974? |
3:08 |
: Complete? I doubt it. Record keeping probably doesn’t support it. Could be wrong, but we need to find someone who wrote everything down for every game…which seems like it would have happened by now
|
3:08 |
When will other projection systems become available? |
3:09 |
: I believe ZiPS starts up in a couple weeks, and the full data will be around early January. Dan can speak to this, specifically though. Not sure on Oliver.
|
3:09 |
Why don’t you do a noon chat on Fridays? Could use a Friday chat at noon, and I’d guess that noontime chats gets better participation? |
3:09 |
: Two reasons. Kiley has that slot, and prospects are a popular topic. At least for now, I have a permanent standing meeting at that time and it would be awfully rude to chat during it.
|
3:09 |
Are the Mets “going for it” or is RAJ’s condition contagious? |
3:10 |
: I think they are going for it. If they are willing to spend a little money, they are definitely within range….so that’s just the question we need to answer.
|
3:10 |
How on earth did Nick Castellanos get a 3rd place vote for Al ROY? |
3:11 |
: Courtesy from local writer. I do not see any justification for the vote, but it’s basically irrelevant given the landslide election
|
3:12 |
Does Royals’ BP success make RPs more desirable as FA targets? Nice to be able to count on a safe and sound 7-8-9, strikes me as worth the money (contrary to ca. 2003 Billy Beane wisdom). |
3:13 |
: Problem is that it’s so difficult to predict reliever performance relative to other positions. Teams are very aware of the value of good relievers, but I’m not sure they will start spending more on them as a result of that knowledge. If you have a 50% chance of finding a Holland in your farm system at 500K or a 70% chance of finding him for $8 million on the FA market, what do you do?
|
3:13 |
AFTER DARK CHATS ARE BACK THANKS NEIL |
3:13 |
: I had nothing to do with this but I will accept your gratitude.
|
3:13 |
Why don’t you sit down at that meeting? |
3:14 |
: Zing.
|
3:15 |
How are the run expectancies by count for pitch type LW determined? |
3:16 |
: Essentially, the difference in observed performance from one count to another. So, the difference between going from 0-0 to 1-0 versus 0-0 to 0-1. The math is just like any other LW after that
|
3:18 |
Are baseball’s fundamentals changing? |
3:18 |
http://www.fangraphs.com/bl…
: Referring to this: And I think yes, always. |
3:18 |
: Baseball, like any competition, is an effort to gain a competitive advantage. So it would make sense that teams cycle through different strategies
|
3:19 |
where does one look for the Crowd sourcing results if one were so inclined? |
3:19 | : there was a post, but they are also here: |
3:19 |
might be a dumb question but if the mets go out and sign someone else who recieved a QO, how does draft compensation work? |
3:20 |
: They give up their 2nd round pick.
|
3:20 |
How far do you think a left handed SS could make it through a minor league system before having to change positions? or is that pretty much squelched right away universally? (say it’s a high school draft pick who was the best on his team so he played SS) |
3:21 |
: I can’t imagine he would play a real inning at SS in pro ball.
|
3:22 |
What does a realistic Ubaldo Jimenez salary dump look like? |
3:22 |
Is a bad contract swap of Ubaldo Jimenez and B.J. Upton a realistic framework for a deal? With the Braves kicking in some $ to even the numbers a bit? |
3:25 |
O’s eat half the salary, they’ll have plenty of takers. |
3:26 |
Why wasn’t Billy Hamilton a base-stealing dynamo? |
3:26 |
: Jeff’s explanation was pretty good. When you’re that good, pitchers don’t forget about you.
|
3:27 |
Would anyone be interested in Jose Bautista and if so, whom. What would the return be on a Bautista trade, assuming that the Jays got an offer they could not refuse. |
3:29 |
: Bautista would be highly sought after. He has 2/28 left on his deal and is projected for 5 WAR this year. In a perfectly rational market, there is tons of value there, and this market seems to crave RH power. He’d probably fetch a not-quite-elite prospect and some solid other pieces. Or, a similarly placed pitcher (nearing FA but not there). Tons of value, not sure they are moving him. Last year was the time to do it, IMO.
|
3:29 |
Who designates plays as “Impossible”, “Remote” etc for the inside edge fielding numbers? Do they have a guideline? Is it normalized? I mean, what’s impossible for Morse to catch may be “Even” for Gordon. |
3:30 |
: Impossible is defined as a play that literally no one at that position could make. The rest are based on the percentage of the time that play is made, and binned accordingly.
|
3:30 |
What if you could trade players outside of the industry? Could the Phillies trade Ryan Howard to Ford in exchange for, say, a 1997 Ford Taurus? Used? |
3:32 |
: He has more value that than if the Phillies pick up his contract. I really want them to eat the money, because I want to see someone sign him, have him get 80 RBI and watch people go nuts
|
3:32 |
1974 is when MLB pbp started? |
3:32 |
: It is as far back as we have complete play by play logs
|
3:32 |
better predictor of MLB talent: success at semi-professional volleyball or beer league baseball? |
3:33 |
: Semi-pro volleyball. Although I don’t really know a lot about such a league…so I may be over estimating the quality of athletes
|
3:33 |
Are other sports, such as football, pursuing a world of advanced metrics on player evaluations? |
3:34 |
: Yes. Basketball is pretty invested, hockey is coming along. I believe NFL is behind, but getting into it. Although I am not that well versed in the details of each sports movement, so I am speaking very generally
|
3:36 |
What exactly do you do for Fangraphs? |
3:37 |
: I think my official title is “site educator.” I was hired to update the Library/Glossary and flatten the learning curve for people new to advanced stats.
|
3:37 |
Mankind lands a robot camera machine on a whizzing interplanetary hunk of rock, and I can’t even get published writing for free about baseball; woe is me. |
3:37 |
: Start a blog, write a lot. Develop a portfolio of good work. Share it.
|
3:38 |
Will presumed continuation of increased prevalence/effectiveness of contact hitters make it harder to hide lousy fielders in corner positions? |
3:39 |
: So far, contact hitters aren’t really in vogue. More balls in play would mean more emphasis on avoiding bad defense, but that hasn’t happened yet
|
3:39 |
What does a realistic ROBOT look like? |
3:39 |
: I am not qualified to answer this question at all.
|
3:39 |
Why is it always about eating the salary |
3:40 |
: Most of our metaphors in English are about food or violence.
|
3:40 |
Should Toronto’s next stadium be outdoors? |
3:40 |
: Current one has a retractable roof. That’s probably the likely outcome
|
3:40 |
how did rickey henderson steal so many bases? everyone knew he was going just like b-ham the runnin man |
3:41 |
: Extraordinary ability/reaction time. Hamilton is fast, but it’s only 90 feet. A lot of it is getting moving in the first place
|
3:41 |
To generate more interest in mid-season games, should MLB experiment with having a couple games a year where the players must run the bases clockwise? Would be an entertaining novelty. |
3:41 |
: Imagine the hot takes.
|
3:41 |
Where can we find historical ZIPS projections? For example, ZIPs projections for 2012. |
3:42 | : |
3:43 |
In 2012, outfielders had 9,237 total impossible chances according to Inside Edge. In 2013, outfielders had 9,753 total impossible chances. In 2014, outfielders had 5,272 impossible chances. Why are the 2014 numbers so different than the 2012/2013 ones? |
3:43 |
: Has to be a change in the definition/categorization.
|
3:43 |
Can Ervin Santana do better this time around? |
3:43 |
: Yes, hey wait a second you’re just putting titles in here…
|
3:44 |
I’d like your take on the latest crowdsourcing effort – how much would you pay Ryan Howard the next 2 years? |
3:45 |
: $5 million?
|
3:45 |
Ubaldo Jimenez for Ryan Howard |
3:45 |
: Can’t hear you over the sound of DD hanging up his phone
|
3:46 |
do you buy Marte as a 4 win player? what do you see him doing in 2015? For some reason i dont buy him as a 4 win guy like hes been the past 2 years but then again..i dislike the pirates |
3:47 |
: Don’t see any reason why he won’t be a 3-4 WAR guy
|
3:47 |
Orioles trade Jimenez, Schoop and Tommy Hunter, along with 8 million to the Braves for Upton and Gattis. Thoughts? |
3:47 |
: I think the O’s pass on this, but it’s a conversation worth having
|
3:47 |
Is there a good rate stat to use for defense? |
3:48 |
: UZR/150, or something like a basic RZR (revised zone rating) which is like defensive efficiency. Or you can just scale UZR or DRS to 1400 innings
|
3:49 |
If you could force the MLB teams to compete in a tournament in another sport/game, what would produce the most entertaining results? |
3:50 |
: Hockey, because it’s always funny to watch people fail to ice skate properly, but the actual quality of play would probably be best in soccer?
|
3:50 |
what do you think of the japanese series games so far? i havent paid a lick of attention. im guessing this is geared more towards Japenese fans. |
3:50 |
: Haven’t watched at all. It’s about growing the popularity in other countries. That’s great, but it’s the middle of the night and there are no stakes
|
3:50 |
Why did my wife make me go on this diet |
3:50 |
: I would assume so that you will live longer
|
3:50 |
: But I don’t know your wife
|
3:51 |
Jimenez for Howard + cash to balance out the contracts + an additional $12M from Philadelphia. Whatever value Howard has left could be best captured on AL team in home stadium with shallow LF. |
3:52 |
: There’s a dollar value where UJ for RH makes sense. Don’t know exactly what it is.
|
3:52 |
Do Cubs and Red Sox ever make a trade under their current regimes? Swihart for Russell, why not? |
3:52 |
: Probably, but teams don’t deal prospects for prospects very often, so that’s probably not happening
|
3:52 |
Would it be better to use an xFIP based WAR than a FIP based one in terms of evaluating future contracts and trades (and then blend in whatever % value folks give to RA9-WAR)? |
3:53 |
: Well, we typically use projections to evaluate contracts, so it’s already a projected FIP, which is taking into account various HR/FB stuff. Your point is right, but we already do a better version of what you’re describing
|
3:54 |
Does UZR account for foul territory? AKA Josh Donaldson debate |
3:55 |
: I believe so, but I am not 100% positive.
|
3:55 |
What way would you attack, without looking at scouting reports,trying to figure out if minor leaguer players high strikeouts are due to pitch recognition issues or plate discipline/over-aggressiveness at the plate |
3:57 |
: Assuming I don’t have PITCHf/x data? Would look at his behavior in certain counts, and also his wOBA on contact. If he’s not swinging a ton early in counts, he’s probably not too aggressive, but if he’s doing damage when he does swing, he’s probably recognizing pitches fine.
|
3:57 |
California kid, went to Long Beach, why is Tulo to the Dodgers not a no brainer for them? Younger Seager good, but prime Tulo when they have a mandate to win now, Friedman wanting to make a splash – what am I missing? |
3:59 |
: I think it makes some sense, but there’s the whole same division thing and the fact that people are focusing on the Dodgers trading an OF right now. Only real Tulo rumor so far was Mets, so the night is young
|
3:59 |
For defense I’m looking for something that shows value per opportunity. Does RZR do that? |
4:00 |
: RZR is percent of balls in zone turned into outs. So you probably want to do UZR or DRS scaled to some number of innings/opps. Although UZR/150 scales the player’s performance to a normal number of chances per 150 games, which might be what you want
|
4:01 |
If a player retires while still under contract, then the contract is forfeited. Do you know the rules governing if that player later attempts a comeback? |
4:01 |
: I do not. The team probably has the right to make them honor the deal, but can also choose to let them out. But I’m guessing.
|
4:02 |
You’re a new team owner & can pick anyone to be your GM. Who do you choose? |
4:03 |
: I would probably choose Andrew Friedman, but if I was an actual baseball owner I would have access to more information that I do now. The public things I’ve seen from AF and things I’ve heard off the record would make him my first choice but I don’t have nearly enough knowledge of everyone else to feel strongly about it
|
4:03 |
dodgers could build a package around Joc to get tulo…no? maybe puig |
4:04 |
: Joc could be a centerpiece but you need more too. Puig would get it done, assuming the Rockies don’t hate him or something
|
4:04 |
: Running low on questions, if you want me to stick around keep them coming
|
4:04 |
What would be the specific skill behind fouling off pitches when necessary? And if there is such a skill (and Panda is not but a serious outlier), can and will it be taught? |
4:06 |
: Presumably the ability to react late? You don’t need to square the ball up, but if you can start your swing late you have an idea if you need to foul it off? Just hypothesizing, though. I suspect it can’t be taught because a lot of it is just instinctual, but again, hypothesis.
|
4:07 |
If the Panda does indeed go to Boston, what does SF do at third base? |
4:08 |
: Headley? I confess I don’t know their farm system very well, but otherwise it’s a trade.
|
4:08 |
Why aren’t %’s used in FIP as opposed to a per IP rate? Isn’t their a BABIP somewhat buried in there when looking at HR, K and BB /IP instead of per batter? |
4:10 |
: Yeah, you can create a similar stat with TBF rather than IP, but there’s very little actual difference and an ERA scale is easier to understand. You also do care about outs with ERA or FIP, so it’s not entirely wrong. (Also K% is better than K/9, but there is rarely a difference.)
|
4:11 |
What 3-4 front offices use data & analysis the least? As a Phillies fan, I know they’re #1 on this list, but who else is in the Stone Age of baseball analytics? |
4:12 |
: Tough to say exactly. Basically every team has an analytics staff, but the real question is where the decision makers are listening. Colorado seems like one team for this list. Hard to tell with Seattle at this point. KC yes and no.
|
4:13 |
: Dbacks?
|
4:13 |
: It’s really hard to know how much the analysis makes it up the chain.
|
4:14 |
The Indians have an expected 10-12 million to play with this off season, they have a serious need in RF and at 3B, do you think that Urshela could be good enough for them to really not look into getting Headley? |
4:14 |
: Headley would be an interesting fit for them.Whoever they get needs to play quality defense.
|
4:15 |
Cubs/Nats seems like a perfect fit for a Zimm/2B trade, right? The Cubs have multiple levels of 2B options, so there’s a lot of room for negotiation. Who do you think gets it done? I’m thinking Javy makes the most sense. But what about Alcantara+? |
4:16 |
: Less than Javy. I don’t think they’re good partners here. JZ is owed $16.5 million for one year. We’re only talking about $10-15MM in surplus value, which nets you a good not great prospect. Cubs can get a similar pitcher for only money right now. Can’t see a deal.
|
4:16 |
Scenario: bad baserunner on second, good baserunner on first, and a base hit. Bad baserunner does not score, which also prevents good baserunn from making it to first. Does the second base runner get dinged (in UBR) for preventing the good baserunner from advancing? |
4:17 |
: They are only dinged for what they do. The average effect of the chain reaction is built into the numbers, but not the specific case
|
4:17 |
what pitcher do the red sox sign/ trade for this offseason? your prediction is__________ |
4:17 |
: Shields?
|
4:17 |
UZR doesn’t count plays w shifts, but what about plays where players subtly adjust their positions. I feel like some players/teams would be drastically affected by somehing like this. What do you make of it, any stat that would minimize this affect? |
4:18 |
: If players are just shading a few steps here or there, UZR includes the play and makes no correction for the positioning. This is the biggest problem with current metrics. Presumably it goes away with StatCast
|
4:18 |
One a scale of 1-10 how much are you bothered by the fact OPS is the sum of fractions with different denominators? |
4:19 |
: 4? OPS is a short hand. It’s a rule of thumb that went mainstream, but it’s a fine tool. I’m bothered from an pedantic perspective, but not a real one
|
4:19 |
How are defensive values created for WAR for players like Babe Ruth? |
4:20 |
: Something called Total Zone. Pretty basic measure of balls hit to you and outs made.
|
4:20 |
Any progress on working w quality of competition? I’ve asked his to a couple FG staff and they said they were trying to incorporate that. Seems like it could make a huge difference, especially w relievers. |
4:21 |
: Stephen Lotfus at Beyond The Box Score has an opponent adjusted FIP and RA9 metric. I know we’ve had some people work on it, but I don’t know how close we are to a finished thing
|
4:21 |
do you have any inside mlb sources? do you ever consult them for writing, or just casual conversation. also do you find that “inside” sources are often as shocked as us about certain trades. i.e. cubs- a’s trade |
4:22 |
: A few. I often talk to them to get the inside opinion on a player or see if they have data to confirm/disprove my ideas. I don’t really chase trade rumors.
|
4:22 |
what do you prefer over OPS? wrc+? |
4:22 |
: wOBA and wRC+
|
4:24 |
Looking at the top pitchers for the last few seasons, it’s striking how many seem to be from west coast teams. I know park factors say hey adjust for this, but might we be underestimating how much harder it is to pitch in the summer in aces like Bal, Cin, Tex, etc versus SF, LA, Sea |
4:24 |
: I’m not sure I buy west coast pitchers being at the top of the lists
|
4:24 |
Have you heard anything about StatCast data becoming available to the public? What is its status? |
4:25 |
: Still sounds like they plan to make it public, but no real work on the format. There’s going to be loads of data, we might not get the raw output
|
4:25 |
By true talent is Troy Tulowitzki the best non-Mike Trout player in baseball? |
4:25 |
: Probably
|
4:26 |
Any stat on FG that tries to neutralize stats for era and park? I know wRC+ compares it to the league that year, but what about trying to compare .285 avg w 26 HR in ’73 to the same in ’99. |
4:26 |
: wRC+ does this
|
4:26 |
what do you think of the latest Lester back to boston? i think its interesting. more i think about it, the more i think its actually realistic. definitely moreso than a month ago |
4:27 |
: Boston doesn’t seem to want to pay what he wants. Still skeptical
|
4:28 |
In other news, Tigers apparently close to resigning Vmart on a 4 year deal |
4:28 |
How crazy are the Tigers for giving V-Mart 4/60 (if rumors are true)? |
4:29 |
: Well, I’m going to have to go write about this in a second. But I don’t think it’s terribly crazy at all
|
4:29 |
: Not a bargain, but it’s fine
|
4:29 |
: Take care, see you next week!
|
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.