56 Responses to “WAR Leaders as of Sept. 14th and Changes Caused by Updated Park Factors”

1. shoewizard Says:
   September 15th, 2010 at 10:14 am

   Sabathia jumps from 5th to 1st in pitching WAR

   That will raise a few eyebrows....LOL

2. Dave V. Says:
   September 15th, 2010 at 10:32 am

   Yeah, that it will Shoewizard! 🙂 There are a lot of people who don't think CC has any business being in the Cy Young debate, let alone possible lead, so with WAR now showing he is #1 in the A.L., I am curious to see what they have to say.

3. Matt Y Says:
   September 15th, 2010 at 10:39 am

   These changes are quite significant for sure. It was a pretty significant readjustment for Yankee players. I'll have to look at the other teams. To me a 0.6-0.7 shift is pretty big, park-adjusted or not. It definitely raises some eyebrows when Sabathia goes from, "shouldn't be in debate and 5th, to firmly in the debate and first in WAR". While I would vote for Sabathia over Hernandez b/c the win difference is just too large, Hernandez still seemed to be having a better sabermetric season --that was my impression. Interesting.

4. Brian Says:
   September 15th, 2010 at 11:08 am

   Joey Votto leads the NL in OBP, OPS, and OPS+. He's batting 380/489/672 with runners on. He's batting 388/473/713 in late and close situations.

   I have a real hard time believing he's the 8th best player in the National League (old or new method).

5. Tom Says:
   September 15th, 2010 at 11:19 am

   Seattle has been more of a pitcher's park. How much of that is affected by Mariners' inability to score runs?
   Is it possible that one of the factors that made Hernandez's W-L record so so-so is also bringing him down in WAR?

6. Sean Forman Says:
   September 15th, 2010 at 11:19 am

   Brian,

   Clutch stats play no part in these stats, so that isn't an issue we consider. You can look at WPA if you want a systematic inclusion of clutch.

   Among position players he is third, though so close to first as to be meaningless. If you don't think pitchers should be included in the ratings, you are free to look at just the position players.

   Regarding the NL pitchers. The pitching in the NL is on average so bad (nearly a higher R/G than the AL), that it is boosting the really good pitchers more than usual.

7. Malcolm Says:
   September 15th, 2010 at 11:21 am

   Well 5 of the players ahead of Votto are pitchers. It doesn't seem unreasonable to me that an ace-caliber starting pitcher is more valuable to a team than a superstar position player. Amongst position players, Votto is ranked third, behind Pujols and Cargo--the way these players are ordered is debatable, but I think most of us would agree that these three have had the best seasons for position players in the NL this year.

8. Sean Forman Says:
   September 15th, 2010 at 11:22 am

   Seattle has been more of a pitcher's park. How much of that is affected by Mariners' inability to score runs?

   It really shouldn't. Park factors consider what the teams and their opponents did at home relative to what they did on the road.

9. King Felix Says:
   September 15th, 2010 at 11:23 am

   Crazy finding, and I wonder how the sabermetric community will react to this. Silly question: are ballpark factors applied on an individual game basis? As in, you don't just take all of CC's stats and multiply in aggregate by a positive Yankee Stadium factor, correct?

10. Steve Says:
    September 15th, 2010 at 11:33 am

    All this does is bring to question the validity of a stat like WAR. I know that no stat is perfect but WAR has so many different variables that slight changes in each variable lead to big changes in player value / rankings. Sabathia's move from #5 to #1 is case in point. That's why most people will continue to favor regular stats like HR, RBI, and OPS - we understand them and they don't change without a player playing. WAR a great stat other than the fact that no one understands how it's calculated, the defensive metrics are questionable, it's calculated differently depending on which website you are looking at, and now park stats change mid season. How can Seattle's park be more of a pitcher's park this year compared to others years? Has there been a physical change to the stadium like moving the fences out or maybe it's because the Mariners are on pace to be one of the worst hitting teams of all time. Just something to chew on.

11. DavidJ Says:
    September 15th, 2010 at 11:37 am

    #7: "Amongst position players, Votto is ranked third, behind Pujols and Cargo--the way these players are ordered is debatable, but I think most of us would agree that these three have had the best seasons for position players in the NL this year."

    The Gonzalez who's in the top three in WAR is Adrian, not Carlos. Cargo is all the way down at #10.

12. Devon & His 1982 Topps blog Says:
    September 15th, 2010 at 11:39 am

    The move of Sabathia here does make me think some more about the Hernandez vs Sabathia for the Cy. It also makes me think about Zack Greinke's opinion from a few days ago, where he said that Sabathia was the most valuable pitcher in the league but Hernandez was the best pitcher. Interesting.

13. Mike Says:
    September 15th, 2010 at 11:51 am

    If nothing changes at a park from year to year, except the roster of the teams, then why should the park factor change. If the run scoring changes in the exact same park from year to year it seems odd to somehow attribute that to the park as opposed to the players?

14. AlvaroEspinoza Says:
    September 15th, 2010 at 12:03 pm

    Will this end the debate that wants to give a .500 pitcher the Cy Young?

15. JayT Says:
    September 15th, 2010 at 12:08 pm

    The park factor can change from year to year because things like the weather can change. If it was unusually dry in Seattle last year then the ball would carry better if it was super wet all this year.

    It's not just the shape and location of a stadium that decides how well the ball will travel there.

16. Michael Jong Says:
    September 15th, 2010 at 12:10 pm

    Regarding park factors: The following was quoted from <a href="" title="this page by Tom Tango">http://www.tangotiger.net/parks.html:

    1 - static conditions: the dimensions of the park. You should use 100 years, if the park hasn’t changed, if you have it, because 300 feet is 300 feet.

    2 - dynamic conditions: the weather, the cut of the grass, the wetness of the field, the wind, etc, etc. You should use multi-year if these things are predictable-dynamic, but single-year if they are unpredicatable-dynamic. If the groundskeeper is the same guy, and cuts the grass kinda the same way, then use multi-years. The wind patterns probably change drastically, so you should use single-year.

    3 - other parks: Park factors are relatively to other parks. You can have say the Big O be a hitter’s park 10 years ago, and a pitcher’s park today, even if the park is exactly the same and the climate hasn’t changed, simply because the other parks have become/introduced as hitter’s parks.

    (Note: because of this, regression towards the mean is not necessarily towards 1.00. You should look at the same parks year-to-year, and in some years, that could mean 1.02 or .98, etc, because the other parks that are not part of the analysis do make up the park factor. )

    4 - the tendencies of hitters/pitchers: If you have a team filled with lefties or flyballers or whatever, that introduces a bias. A park that might be optimal for flyballers, and you have a team that is filled with flyballers will not show an accurate park factor. The sample of your players should be representative of your population. For extreme parks, like Fenway and Coors, I’m sure they are not. The degree of which, I’m not sure.

    5 - the quality of hitters/pitchers: Barry Bonds is great no matter where he hits, and he might not be hampered as much as someone else by playing in a pitcher’s park. So, he hits the ball 390 feet instead of 400 feet. If ReyRey hits the ball 320 feet instead of 340, that’s a big difference. Again, I don’t know the degree of impact here, but it is a consideration.

    6 - the game context: Different game states (score/inning/base/out) results in different hitting/pitching approaches. Again, probably small impact, but needs to be determined.

    So, your “park factor” is made up of several factors, each of which needs to be analyzed on its own. For some, you can use multi-years, and others you need single-years, etc, etc.

    Long quote, I apologize, but essentially PF can be affected by more than just the dimensions of the park. Dynamic effects such as local weather can have an effect. New parks can change a park factor (think about how other parks were affected compared to average when Coors entered the fray). In addition to that, there is the year-to-year fluctuations of players that play in the parks (both home team and the visitors).

    That's why multi-year factors are usually used, in order to wash out some of the randomness of these effects. But in some cases, it can be a little difficult, as in the case of CitiField/NYS/Target Field, where there aren't enough years to make good assessments on how a park plays. Even when you use multi-year factors, however, you can still get slight changes from year to year. Remember, this is an estimation, not the "true" PF handed down from the baseball gods.

17. Jerry Says:
    September 15th, 2010 at 12:19 pm

    Interesting that New Yankee Stadium is allegedly more of a hitter's park, since there's been a lot of discussion of how the ball doesn't fly out to right center as much as it did last year (supposedly due to the demolition of the old stadium changing the wind patterns).

18. Michael Jong Says:
    September 15th, 2010 at 12:21 pm

    I think people are getting a little hung up on small differences. I wouldn't consider the difference between Sabathia and Hernandez/Liriano on the new leaderboard that great. To dismiss either player because of this change would be to misuse WAR.

    I would never say a player is definitely better than another one because he has 0.3 WAR more than the other guy. I personally go with increments of 0.5 WAR, with everyone within half a win of each other being pretty close to equal that it's to debate.

19. Darren Says:
    September 15th, 2010 at 12:49 pm

    So if Seattle is now a less hitter friendly park due to inclusion of 2010 park factors how does F.Gutierrez defence WAR jump .2 ?

20. Sean Forman Says:
    September 15th, 2010 at 12:59 pm

    So if Seattle is now a less hitter friendly park due to inclusion of 2010 park factors how does F.Gutierrez defence WAR jump .2 ?

    Probably four things put together.

    1) he had five putouts last night
    2) a reduced run scoring environment decreases the number of runs per win.
    3) probably rounding. He may have gone from 1.04 to 1.16
    4) the rest of the CFers may have had a bad night as tzr is versus the league average. More rookies up could mean decreased quality of OF play.

21. Sean Forman Says:
    September 15th, 2010 at 1:00 pm

    I would never say a player is definitely better than another one because he has 0.3 WAR more than the other guy. I personally go with increments of 0.5 WAR, with everyone within half a win of each other being pretty close to equal that it's to debate.

    I would agree 100% with this and add that round WAR to a whole number is probably defensible, but then we'd have people asking how a player went from 4 to 5 WAR when he was 0 for 4.

22. Matt Y Says:
    September 15th, 2010 at 1:03 pm

    I thought the same thing #17. I understand how all these things can be more dynamic than originally thought, perceived, or enumerated (it's not as easy as the stadium is the same from year to year so why does there need to be such park adjustments) , but please, wind patterns don't change that much from year to year April-Oct at the same location, UNLESS, perhaps an adjacent big building comes down (i.e old Yankee stadium), which then renders stadium less hitter friendly (which is what is perceived), but more WAR hitter friendly. Arrggh. As for #18, I applaud the .5 approach, but many WAR-centrists do make a huge deal out of someone having a WAR of 5.4 to 4.9. It creates a platform for ones to say, CC, for CY, hogwash, it's all King Felix. Then, we have a park adjustment, and that argument completely flies out the window. Again, i will back the WAR as the best single metric out there until I'm blue in the face, but I will also state until I'm blue in the face, that it's a number to be taken with a healthy sized grain of salt!

23. Tmckelv Says:
    September 15th, 2010 at 2:11 pm

    I know the teams/stadiums would never go for it, but as part of the pre-game, someone should use a pitching machine and shoot baseballs to the outfield and infield at various prescribed angles/velocities to formulate some kind of rating for dynamic park factors from such effects as wind/weather/grass length/ground hardness/etc. Then that rating could be used in the calculations for a given year. it would show the actual effects rather than just trying to determine what the effects might be given the temperature/wind gusts/precipitation (if those measuements are even used in calculations).

24. MikeD Says:
    September 15th, 2010 at 2:27 pm

    This is only going to reinforce among some why WAR is not dependable. One data point is changed and the numbers change. Sabathia goes from not being worthy of consideration for the Cy Young, to being the leader. As a long-time baseball stats person, going back to Bill James' Baseball Abstracts and everything since, I am hardly in the Murray Chass school of thinking. I use WAR as one was to evaluate players, but I don't accept it as gospel. On the other side, the fact that WAR is always being updated to reflect new data is a positive.

25. MikeD Says:
    September 15th, 2010 at 2:48 pm

    Question for the B-R guys (or anyone who knows the answer.) When constructing WAR and OPS+ on Baseball Reference, are you using a blended average (meaning multiple years) to come up with the park factor input? Despite the hype, last year the new Yankee Stadium seemed to play neutral to slightly more pitcher friendly. This year, it's the opposite, playing more as an overall hitters' park. That probably reinforces the idea that we'll need three years of data to get an accurate view of the place, or any new/redesigned park. Yet that gets back to my question. Are the updated WAR and OPS+ numbers that I'm seeing for 2010 based on two years of data from the new Stadium (when is it acceptable to drop the word "new"?), or are the 2010 numbers only based on what's happening in 2010?

    Last, I think the main reason the new Yankee Stadium is more offense oriented this year than last relates to doubles. Yankee Stadium (all versions) has always been a good HR park because of the short right field. The new Stadium seems even more so because the fence is shorter and not curved away from the field, making it a shorter HR shot by on average nine percent. Yet Yankee Stadium (all versions) has also reguarly depressed doubles, making it in many ways the opposite of Fenway, which is not as easy to hit HRs as people believe, but is a doubles factory, leading to higher offense. Yankee Stadium in 2009 once again reduced doubles, turning it slightly favorable to pitchers. The old Yankee Stadium was the same, and since the dimensions of the new place are, more or less, the same, it made sense doubles would continued to be depressed. That's not happening this year, which is why offense if up. It'll be interesting to see what happens in 2011, the third year of the park. I'm guessing doubles go back down.

26. Johnny Twisto Says:
    September 15th, 2010 at 2:52 pm

    These park factors are based on the total runs scored in a team's home park compared to total runs scored in its away games. That's it. They are not based on dimensions, temperature, wind gusts, full moons, whatever. Those things may affect how many runs are scored, but they are not directly considered in calculating park factors. Runs scored, home vs away. Nothing more.

27. DavidJ Says:
    September 15th, 2010 at 2:52 pm

    #24. "This is only going to reinforce among some why WAR is not dependable. One data point is changed and the numbers change. Sabathia goes from not being worthy of consideration for the Cy Young, to being the leader."

    Before the park adjustment, Sabathia was tied for fifth in WAR, just 0.6 behind the leader. If some people did not believe that CC was worthy of Cy Young consideration based on that, then that is a problem with them, not with WAR.

    The arguments that CC is not worthy of Cy Young consideration have more to do with defense-independent pitching stats such as FIP, and so never really had anything to do with the B-REF version of WAR to begin with (since that version is based on a pitcher's actual runs allowed, as opposed to the FanGraphs version, which is based on FIP). Anyone who was citing the B-REF version of WAR as an argument against CC's Cy Young candidacy was misusing it, since CC's ranking even before today's adjustment already had him in position as a reasonable candidate. The top few candidates were all very close in WAR before the adjustment, and they're all very close after it; there's just been a little bit of reshuffling. The change really isn't as dramatic as some are making it out to be.

28. Johnny Twisto Says:
    September 15th, 2010 at 2:53 pm

    Mike, see here: http://www.baseball-reference.com/blog/archives/8259

    They use 3-year PFs.

29. Kelly Says:
    September 15th, 2010 at 2:55 pm

    This is sabermetric malpractice! How can you apply one PF to all players on a team, regardless of whether they are RH or LH, power hitters or singles hitters, speedy or slow? To show the folly of this I need go no further than Coors Field:

    Home OPS - Road OPS (Career)
    - Seth Smith, +.321
    - Carlos Gonzales, +.289 (Rox years only)
    - Clint Barmes, +.178
    - Todd Helton, +.145
    - Troy Tulowitski, +.121
    - Colorado Rockies Team, +.220 (2010 only)

    So, Coors helps everyone, but it helps some players a hell of a lot more than others. Other parks like Fenway with their idiosyncratic measurements show even more pronounced differences.

    Folks here have vested interests in OPS+, WAR, etc., so I'm sure I'll get flamed, but anyone who stops to think about this for even a minute will know that I am right.

30. Tmckelv Says:
    September 15th, 2010 at 2:58 pm

    Boy, at this time yesterday Robinson Cano was CLEARLY the Yankees MVP this season with a 6.1 WAR with his next closest teammate being Sabathia @ 4.6.

    Since then, Cano went 3-4 last night (w/ HR & 2B) and the new park adjustment kicked in...now Robby and CC are tied @ 5.4 WAR.

    Also, If this Park Adjustment correction will be made every year in September, then it is probably not a good idea to use WAR as the basis for ALL-Star selections, which I saw many people try to suggest this past July.

31. Michael E Sullivan Says:
    September 15th, 2010 at 3:08 pm

    Anybody who thought Sabathia was out of consideration for the Cy Young was being silly. He was .6 WAR behind the leader with 17 games to play. As others here say, .5 WAR is like the margin of error. So, if you look *solely* at WAR, the only guy you'd have had clearly ahead of him is felix hernandez, everybody else was within that .5. And Sabathia isn't a clear leader now either, he's just the leader, ahead of Hernandez by about the same amount that he used to be behind him.

32. Josh Says:
    September 15th, 2010 at 3:09 pm

    "All this does is bring to question the validity of a stat like WAR."

    The WAR formula/methodology didn't change, and neither did the formula for simpler things like OPS+ or ERA+. This was just a data input issue and not a problem with any of the formulas themselves.

    All that happened was that Sean accounted for new input data all at once (i.e., how the parks have been playing this year instead of just previous years) that hadn't been updated yet this year. Sure, it would have been better to have had that data day-by-day all year long, and then we wouldn't have seen this jump. It sounds like we will get that feature next year. What more can you ask from a free site? Let's also be clear, if it weren't Yankee Stadium that happened to be the big outlier, this wouldn't be nearly the same "issue" in so many peoples' minds.

33. Tmckelv Says:
    September 15th, 2010 at 3:11 pm

    Yikes, I just happened to check Derek Jeter's stats to see if his OPS+ could creep up back toward 100 (from the low 90's) after his first multi-hit game in what seems to be 2 months...but I guess the park factors have kicked in for OPS+ also with Yankee Stadium knocking him down to 86. Now he has gone from horrible to pathetic overnight. 🙂

34. Josh Says:
    September 15th, 2010 at 3:16 pm

    Kelly,
    The reason everyone on a team gets the same park factor adjustment is because the adjustments are trying to measure value toward winning games, not theoretical ability. If a certain player can take extra advantage of his home park (Boggs in Fenway or Damon in New Yankee Stadium) that other players cannot, good for him. His performance led to real runs, which in turn led to real wins for his team. That player deserves credit for the real value he brought even if maybe he wouldn't have been quite so valuable playing in a different park.

35. MikeD Says:
    September 15th, 2010 at 3:26 pm

    @27 and @31, agreed. It's just that people were using that argument, so it's a bit, ummmm, funny, that it's changed from one day to the next.

    @29, Kelly, I had the same thought. It does strike me as odd that one park factor would apply to all players on one team. Since the Yankee Stadium and CC Sabathia have been subjects in this thread, let's continue to use them as examples. When the Yankees have constructed their teams over the years, they always make sure to have left-handed power to take advantage of the short distance to right, and those always had left-handed starting pitchers, because they negate other teams lefty power hitters, while at the same time Yankees lefty starters could use the wide expanses in center and left field to their advantages against right-handed hitter. It would seem based on that a lefty starting pitcher, like Sabathia, would have an advantage at Yankee Stadium, more so than a right-handed pitcher, yet he's being rated with the same park factor. At the same time, going back in history, I doubt Joe DiMaggio's OPS+ truly reflects the hitting conditions he faced at Yankee Stadium. CF and left/CF took moon shots, and it's reflected in DiMaggio's home/road splits. Bill Dickey, from what I understand, became a master at hooking the ball down the right-field line for HRs. Now it's still a skill on Dickey's part to do that, yet Dickey and DiMaggio were both being rated with the same park factor, even though they had very different hitting environments.

    BTW None of this is knock on B-R, which is the best site out there. Discussion leads to improvements, or at the least, better understanding.

36. Djibouti Says:
    September 15th, 2010 at 3:53 pm

    I've never understood this argument that adding precision to a process and getting different results somehow invalidates the process. Anyone who takes WAR as truth and declares the value a complete and infallible summary of a player is an idiot. WAR is not perfect, it's merely better. Because it is not perfect, it will be continuously refined. That doesn't make it wrong, that makes it a work in progress. But even with the old PFs, WAR was vastly superior than W/L or ERA or SOs. These are stats that measure a singular aspect. WAR tells a story. To use a loose metaphor, most stats are like headlines; they give the basic information and that's it. You might get an idea of what's going on in the world from just headlines, but you won't know everything. WAR is a newspaper; takes all the headlines, expands on them and gives you all the news. If you see a note from the editor saying they got somebody's name wrong and they're correcting it, do you declare the paper as a whole to be useless as a source of news?

37. Charles Saeger Says:
    September 15th, 2010 at 5:33 pm

    Kelly@29: One year does not a park adjustment make, even for a whole team. Not even for a whole career will have enough data for a correct park adjustment. Coors in particular changed with the humidor.

    "Sabermetrics" is objective knowledge about baseball. Todd Helton's home/road split is less than Seth Smith's; we want to know why. Is it luck? There is assuredly a great deal of that in Smith's home/road split, as he has career 887 plate appearances. Both men are left-handed hitters. Both have identical career GB/FB ratios, infield fly rates, pull rates, home runs per flyball rates. Why should we have custom home/road splits for the two men? Aside from quality and line drive percentage, they look to have the same characteristics to me.

    To say that Seth Smith has had a bigger home/road gap than Todd Helton is a fact. To say that Coors Field has helped Seth Smith more than Todd Helton is conjecture.

38. Todd Says:
    September 15th, 2010 at 6:46 pm

    Just a random thought, I don't see where anyone else mentioned it in the comments: Really, the 2010 WAR "leaders" won't be determined until after the 2011 season, if I understand correctly that a three-year average of park factors has been used to determine previous seasons' WAR numbers. Am I correct with that statement?

    I'm not saying it in a smart-alec way, because I'm sure the changes would be rather small and not all that significant. But, I would assume that, if so-and-so finished just a tiny fraction short of the all-time single-season WAR record as calculated by this site, that could possibly change based on what happens next season?

39. Johnny Twisto Says:
    September 15th, 2010 at 8:19 pm

    Todd, correct.

    Park factors are complicated and there are many different ways to use them or apply them. B-R chooses to use 3-year factors, bookending the season in question. So in real time, it seems a little funny for the numbers to change a year later. When looking back though, presumably the numbers using PFs will be more "accurate" for having the additional information. Ideally you'd want lots of years to know the "true" PF for a park. But that's not really possible, because the park gets altered, other parks (to which it is being compared) change, other parks close and open......

40. Fireworks Says:
    September 15th, 2010 at 9:03 pm

    I like how people are complaining about the update. I was just wondering what would happen with the park factors the other day based upon a comment about NYS's PF being much higher than it was last year. And then Sean goes and updates the PFs and pledges to devise a formula so that the current season's PF isn't merely a mirror of last season all year.

    I love you Sean.

    Sabathia!

41. tim Says:
    September 16th, 2010 at 9:08 am

    These WAR stats love Justin Morneau, yet two years in a row the guy has missed more than a month of play and his team has played better without him.

42. jiffy Says:
    September 16th, 2010 at 9:44 am

    Re #41, it's weird, it's almost like baseball is a team game or something.

43. Fireworks Says:
    September 16th, 2010 at 9:45 am

    @41

    Are you suggesting that the Twins are better off without Morneau and thus that WAR "lies" about to value of a player?

    You know, the Yankees have an excellent record without A-Rod in the lineup this year. I suppose you think that they are a better team when they start Eduardo Nunez or Ramiro Pena at third base rather than A-Rod?

44. Lee Panas Says:
    September 16th, 2010 at 2:37 pm

    Such a big swing in park factor in one year suggests to me that the Yankee Stadium PF is still very volatile and needs more years of data. 4.6 to 5.4 is too big of a swing for me to trust the Yankee Stadium PF at this point. If we look again after the 2011 season, who knows where it will be then? I usually like to see five years of data for a PF. Perhaps park factor needs to have a smaller influence for pitchers pitching in stadiums that have not been in existence for a long time.

    Lee

45. MarkW Says:
    September 16th, 2010 at 4:59 pm

    Poster #41 has obviously never seen Cuddy play first base.

    Though he's probably one of those Twins fans that thinks Delmon Young is the best hitter on the team, and ripped Morneau during the first half of the year for not driving in runs and walking too often.

46. Fireworks Says:
    September 16th, 2010 at 5:20 pm

    @44 Lee

    Such a big swing in park factor in one year suggests to me that the Yankee Stadium PF is still very volatile and needs more years of data. 4.6 to 5.4 is too big of a swing for me to trust the Yankee Stadium PF at this point.

    There was one year of data and now the average of two years is being used. And it's a false concept that there's a "swing". Sabathia's WAR, like everyone else's all year, wasn't using updated park factors. A proper understanding of the implementation of this update is that 5.4 is his WAR as B-R calculates it, and 4.6 was not. Period. There's no 'swing'. An old variable, using limited data below the minimum required for B-R's WAR merely skewed the data. Also, you say that 4.6 to 5.4 is too big of a swing, so does that mean if the 'swing' was less then you'd be perfectly fine that NYS is only using two years of data?

    If we look again after the 2011 season, who knows where it will be then?

    I don't get your point? The 2011 data might be different from the 2010 data or 2009 data? Okay. So? I get the feeling you want the park factor to be consistent year-to-year, as if there's some sort of regression to a mean and park factors really perfectly describe the park itself instead of telling us what happened in that park in that particular year, which is subject to variation and volatility naturally.

    I usually like to see five years of data for a PF. Perhaps park factor needs to have a smaller influence for pitchers pitching in stadiums that have not been in existence for a long time.

    Okay, so you like five years of PF data. Fine, that's your preference. But you know what? We don't have five years of data on the New Yankee Stadium. I don't get what you're proposing when you say it should have a smaller influence? What park factors would you use for players playing in new stadiums? Whether or not the data is volatile due to small sample is irrelevant--that's all the data. You can't pull park factors out of your butt because it may cause unsexy variances in the data. It is what it is.

    Further, I like that Sean uses three years. I don't think you need to try to dampen park effects by using a huge number of years. If a park comes off as a big hitter's park one year, that's what it was. Having it being bookended by four or six or eight more 'normal' years only serves to devalue what happened in the year when it wasn't that. Three years is a great compromise between outliers having a huge effect and many far-removed years dampening what a park actually was in a particular year.

47. Lee Panas Says:
    September 16th, 2010 at 7:04 pm

    Fireworks, I think I posted too fast and didn't explain myself well! I'm fully aware of what caused the change from 4.6 to 5.4. My point is the big change in park factor makes me think that we really don't have a good feel for what the actual Yankee Stadium Park park factor is as of yet. Thus, I don't have great confidence in either the 4.6 or 5.4. I'll probably have more confidence when another year is added next year.

    I understand your point about three years versus five but I prefer more rather than less. Just because a park seemed to play more like a hitter's park one year doesn't mean it actually was more of a hitter's park. It could just as easily be a fluke as a real effect. Because of their volatility, I like to err on the side of caution with park effects. Thus, I don't mind the potential "dampening" effect. That's just my preference though and an argument could be made for either side.

    I don't mean to be critical of rWAR. It's another very useful data point along with pitching runs, fWAR and some other measures. The more measures we have, the more informed we will be. I'm glad rWAR is here and I think Sean's doing the best he can with limited data for some stadiums.

    Lee

48. DavidRF Says:
    September 16th, 2010 at 7:16 pm

    @46
    I agree 100% with what you are saying about 2010. That seems be an issue with infrequent in-season park-factor updates that Sean is planning on remedying in future seasons.

    If there's anything to worry about, its what has happened to 2009 numbers. The Yankee Stadium park factor jumped by the same amount there. So, after the season was over and after all the postseason awards are debated, voted on and handed out... all players on one team get a ten point shift in their OPS+ (down) and ERA+ (up). The first year of a park is more susceptible to this one-year-later shift in adjusted numbers because you're doubling the park information instead of adding 50%.

    Its the right thing to do and there's nothing one can do about it. But its something to keep in mind when evaluating players playing in a new park's inaugural season. The error bars are a little larger than usual.

49. Neil L. Says:
    September 16th, 2010 at 7:21 pm

    Thank you for the data, Sean. It will be great to have up-to-date park data on an ongoing basis on this site.

    Am trying to absorb all the different debates in here and respond intelligently.

    It is interesting how the updated park factors decrease Bautista's WAR numbers compared to Hamilton's. Hamilton stayed at his "old" value while Bautista went down 0.5.

    My issue with park factors is that a good hitter can raise the number to a small extent by having an awesome offensive season at home. Park factor and home offensive performance are inextricably intertwined.

    Cause and effect cannot be separated. So because Jose has hit 30 of his 47 HR at home he has helped make the Rogers Centre a hitters park this year.

    Perhaps Rogers Centre has not changed in any way, but the home team's home runs are a temporary abberation. I must be misunderstanding how PF is calculated....

50. Fireworks Says:
    September 16th, 2010 at 10:13 pm

    @47 @49

    I remember how a poster was saying OPS+ or a similar stat was terrible because it penalized a guy like Richard Hildalgo, who slugged more on the road than in Enron Field in 2000. He talked about how it penalizes right-handers in Yankee Stadium and made other such comments. And then someone posted that these stats don't attempt to (nor could they) quantify ability. They attempt to quantify value. I look at park factors the same way. Doesn't matter if a park should be a pitcher's haven or a bandbox. What happened is what happened. Several years of park factors may tell you what sort of park it is in general, but that doesn't mean that for a short period it can't be something other than what it is normally, whether that is due to an identifiable variable or an unidentifiable one. The way I see it park factors for a spread of years tell you what sort of park it is, and park factors from one year tell you what it was that year. In both cases, you just use that number to determine players' values based upon their environment.

    If the Padres and their opponents knock the crap out of the ball at Petco and Petco somehow is a 115 for that one year, that's what it is. If it's partially because Adrian Gonzalez hit 74 homers and somehow hit 50 of them in Petco, oh well. It is what it is.

51. John Q Says:
    September 16th, 2010 at 11:38 pm

    Seeing these new updates just makes me scratch my head as to why the Mets built Citi Field to be such a pitcher's park. It's especially odd when you consider that David Wright was one of the best hitters they've ever developed and he was going into his age 27 season in 2009. Why wouldn't you build a field that was a slight hitter's park or a least a neutral park?

    Then to make it more odd, they signed Jason Bay who seems like the prototypical Anti-Citi Field player.

52. How Not To Use Park Factors | U.S.S. Mariner Says:
    September 17th, 2010 at 1:08 am

    [...] level for New Yankee Stadium and significantly lowered it for Safeco Field. As Sean Forman noted in his post announcing the move, the data indicates that Seattle has “become more of a pitcher’s [...]

53. Johnny Twisto Says:
    September 17th, 2010 at 7:32 am

    Neil, I think you are understanding it correctly. It is definitely true that a PF could be affected by a team's players having a particular ability (or difficulty) to take advantage of the park.

54. Jim Says:
    September 17th, 2010 at 11:09 am

    Does war take into consideration that C.C. never faces the Yankees lineup, and that Felix never faces the worst lineup in baseball? There has to be some competition consideration when you're talking about Yankee pitching.

55. Sean Forman Says:
    September 17th, 2010 at 11:16 am

    Jim,

    It does. It computes the average quality of the opposition faced.

56. Neil L. Says:
    September 17th, 2010 at 5:11 pm

    @51
    John Q, do the architects of a ballpark really know how it is going to play out?

    I suppose raw distances to the fences are an easy predictor of a pitchers' paradise but then a huge outfield might make it a good doubles or triples park. Who could how Citi Field would shake out?

    With respect to designing a stadium for a current star, in this day and age of free agency and baesball as a business, the player turnover is too great for a franchise to reap much benefit from a roster-designed park.