2010 End of Year Park Factors and Similarity Scores Updated
Posted by Sean Forman on October 4, 2010
I've re-run the stats with end of year park factors applied, so if you've been holding onto your MVP ballot, you can now fill it out with full information available.
Next, I've updated the sim scores for players who played during the 2010 season.
Please let us know if you find any issues with the site.
October 4th, 2010 at 3:00 pm
Not quite full information available....ROE and GIDP components of WAR still appear to be zeros...
October 4th, 2010 at 3:11 pm
Speaking of end of year awards, an interesting thing I dug up...
CC Sabathia has 21 starts with a positive WPA and is 19-0 with 2 NDs (including his highest WPA of the year, ironically) in those games.
He has 13 starts with a negative WPA and is 2-7 with 4 NDs in those games.
Felix Hernandez 24 starts with a positive WPA and is 13-4 with 7 NDs (including 2 of his highest 5 and 3 of his highest 7) in those games. It is also worth noting that in 3 of those loses, he had a WPA of less than .010.
He has 10 starts with a negative WPA and is 0-8 with 2 NDs.
Sabathia's WPA range from .573 to -.340.
Hernandez's WPA range from .581 to -.362.
If Hernandez, like Sabathia, won 90.4% of his WPA+ games and 15.4% of his WPA- games, he would have had 23 wins.
If Sabathia, like Hernandez, won 54.2% of his WPA+ games and 0% of his WPA- games, he would have had 11 wins.
I don't know if this is necessarily the best application of individual game WPA for starters, but it seems, intuitively, to be a good one. And I realize that WPA still has its fair share of kinks to work out. However, this is just one more piece of evidence that the relative strengths of their teams had a massive influence on their W-L record.
Basically, Sabathia won 2 games where he actually hurt his team and nearly every game where he helped them. Hernandez won 0 games where he actually hurt his team and barely more than half of the games where he helped them.
October 4th, 2010 at 3:12 pm
I realize we could go through every pitcher and play what if, and may find even more incredible findings, but I went with these two because of the huge debate going on. If anyone else wants to run further with this idea, I need no credit. Just do it justice!
October 4th, 2010 at 3:21 pm
Completely off topic, but the Pirates were so bad they almost had a reliever lead their team in strikeouts (Hanrahan 100 vs. Maholm 102).
October 4th, 2010 at 3:35 pm
Looks like most of the park factors regressed a bit in the final three weeks. Target Field dropped 3 points from moderate hitter park 104/103 to near neutral 101/100. That's the most volatile because its only one year of data. Yankee Stadium ticked back down one after the big increase from last year (106/104->105/103). Seattle's PPF notched up one (93/94->93/95). Coors Field came down a tick (116/116->115/115).
October 4th, 2010 at 3:37 pm
Looks like Felix Hernandez is back on top of the AL Pitchers WAR leaderboard.
October 4th, 2010 at 4:45 pm
For the 2nd year in a row, Felix Hernandez was the best pitcher in the A.L. In a squeaker, my figures show that Jered Weaver was the 2nd best starter in the A.L., finishing between Hernandez and Sabathia. What do you want to bet that Weaver doesn't even finish in the top 5 of the Cy Young.
October 4th, 2010 at 5:23 pm
@7 I am pretty sure last year Greinke was a much better pitcher than King Felix, it is almost like this years debate is similar to last years, King Felix 19-5 on a very good team vs Greinke 16-8 on a very poor team, this year CC 21-7 on a very good team vs King Felix 13-12 on a very poor team.
October 4th, 2010 at 5:56 pm
[...] 2010 End of Year Park Factors and Similarity Scores Updated … [...]
October 4th, 2010 at 9:46 pm
I don't think Hernandez quite merits a comparison to Greinke's 2009. Greinke's gap on the field was immense, and his record was much better.
October 4th, 2010 at 10:12 pm
Jared Weaver's year has completely flown under the radar. He just had a great season with basically no media attention. 1st in k's, 3rd in whip, 2nd in bb/k, 5th in era & era+. He's just another great example of the relative pointlessness of W/L.
October 5th, 2010 at 6:04 am
@ #2
That's a really interesting way to compare Sabathia and Felix but with all of the talk that Felix was on a bad team, it's important to note that he was the beneficiary of having the best fielding team (according to WAR) behind him while pitching in a pitcher's park. That has a positive influence on Felix's pitching peripherals. I don't think it's fair to credit Felix for his poor run support and not give him demerits for his great fielding behind him. I can't see rewarding him for his team's management setting him up to lose low scoring games because they thought they didn't need any hitters.
I'm sure Felix would be a big winner on the Yankees but I also think his ERA would be over 3.00. As imperfect as they are, wins aren't meaningless or arbitrary, there's a reason CC has about 20 every year.
My ballot;
Price
CC
Felix
Weaver
Lester
October 5th, 2010 at 10:37 am
While CC certainly benefits from pitching on a good team I also have an issue with Felix getting the nod considering he didn't pitch a single meaningful inning the whole year. Context works both ways! Yes, we can play what-ifs, but there's a reasonably good chance Felix gets tanked a few times while pitching in the heat of a pennant race thus inflating his ERA to high 2's to near 3.00. Because the Rays won the east and Price outpitched CC, Price gets my nod. While certainly not the best stat to rank a pitchers effectiveness, I'll also state wins aren't arbitrary or meaningless. They do matter. The difference in WAR b/w Felix and CC or Price isn't great enough to make up for the 6-8 win difference.
My Ballot:
Price
CC
Felix
Lester
Weaver
October 5th, 2010 at 11:18 am
he didn't pitch a single meaningful inning the whole year.
Indeed, I'm surprised Seattle didn't just cancel the entire season. Certainly the Yankees would like back those three games when Felix shut them down.
October 5th, 2010 at 11:39 am
Maybe that's slightly harsh, but yes, Seattle got off to an awful start and started dumping players in June 60-70 games in! They were out of it, despite thinking they were going to be good, 20-25 games into the season. Context matters both ways. As other saber-focused people have mentioned, a .5 difference in WAR is basically a wash. So I encourage everyone to apply that mentality. The difference b/w Felix and others was .6.
October 5th, 2010 at 11:54 am
I see Felix, CC and Price as having equally good years and when looking for a tie-break I'm going to look at what their teams did -- Pretty reasonable and in-line with #12 and probably the majority of others. Me stating Felix didn't pitch a meaningful inning is no different or any more narrowly-focused than some saber-focused types making their decision solely based on WAR.....and, I supported Greinke's pick last! If the difference in WAR was more like 1.0 I might be persuaded to Felix... or even if Seattle was relevant until late July-early August it might be different --they were out it by May!
October 5th, 2010 at 1:40 pm
Matt Y is a reasonable man.
October 5th, 2010 at 2:18 pm
As Keith Law points out, pitching in games where allowing 1 run could very likely lead to a loss is a HELL of a lot of pressure.
Yes, context works both ways. Felix has contextual elements in his favor and opposed to him, just as every player does. Right now, the stats that best take into account those elements all say that Felix is best, both here and at FanGraphs, which seems to have a slightly more reliable and consistent methodology for defense and park effects (not a knock on the guys here at all, just saying).
October 5th, 2010 at 2:20 pm
"Me stating Felix didn't pitch a meaningful inning is no different or any more narrowly-focused than some saber-focused types making their decision solely based on WAR.....and, I supported Greinke's pick last!"
You can't even compare the two. You saying something 1000000% wrong does not compare to folks using what they feel is the strongest criterium to make their determination. You can certainly disagree with them on the criterium. But there is no basis in fact to say that Felix faced no pressure. Absolutely none. Pressure can be defined in a multitude of ways.
October 5th, 2010 at 3:03 pm
Not that pressure should be the focus of the discussion on CY, but a pennant race definitely has more pressure involved than pitching for the worst team in the league. Pressure is a matter of degree so I'd drop that million % wrong to about 25%.
I'm curious, are there any sabermetric pitching stats that take into consideration what the defensive fielding metrics are for the fielders in that particular game? Or how about opponent strength, especially against individual batters?
According to advanced metrics, Jeter and Posada are terrible and Cano is south of average. Shouldn't CC get a big WAR "credit" for getting so many guys out with those bums up the middle?
Plus, the East has better teams than the West and better hitters, granted, Seattle doesn't get the pleasure of playing itself. I would look at those things to separate what Price, CC and Lester accomplished compared to Felix.
I also find it odd when guys hate errors but embrace ERA. I would think those guys (error haters) wouldn't have any use for ERA.
Anyway, maybe WAR(or something like it) does take all that into consideration, I'm just wondering.
October 5th, 2010 at 3:45 pm
FanGraphs, which seems to have a slightly more reliable and consistent methodology for defense and park effects
Why do you say that? (I don't have a strong opinion on the matter, just wondering your thoughts since I think those are tricky issues.)
According to advanced metrics, Jeter and Posada are terrible and Cano is south of average. Shouldn't CC get a big WAR "credit" for getting so many guys out with those bums up the middle?
Plus, the East has better teams than the West and better hitters, granted, Seattle doesn't get the pleasure of playing itself. I would look at those things to separate what Price, CC and Lester accomplished compared to Felix.
I also find it odd when guys hate errors but embrace ERA. I would think those guys (error haters) wouldn't have any use for ERA.
Raker, B-R's WAR does "credit" pitchers for the quality of defensive support, adjusts for their strength of opponents faced, and does not discriminate between ER and UER.
October 5th, 2010 at 3:45 pm
And I screwed up the italics tags again...sorry for the confusion in what I'm quoting in post 21.
October 5th, 2010 at 4:16 pm
It's all relative, and when there's three nearly equal guys I'm going to look at peripherals and at what their teams did and what each pitcher did in big needed games. Sorry, last I checked baseball is still somewhat a team game. Yes, WAR tries to get at all these parts, but it's not going to be the tie-break for me in this situation. Maybe some other situation I would let the WAR break the tie, but not here. Last year the WAR was the tie-break for me. To give even more context from where I'm coming from, Greinke last year was 2.2 WAR above the rest I would have voted for him b/c of this. As I said above, I have seen at least a few very saber-focused types say themselves that a WAR difference of 0.5 is nearly negligible. Yes, statistically, King Felix has a very, very slight edge, but team-wise, both Price and Sabathia have a huge edge --tie-break therefore goes to Price and CC. That's the breaks of the game. Pretty logical I think. King Felix isn't that far head and shoulders above with the peripherals to get the nod IMHO. If King Felix had a WAR 1 or so above the rest he'd get my nod. I don't think it's my viewpoint that's missing the bigger picture here.
October 5th, 2010 at 7:37 pm
Thanks for the heads up Johnny Quisto.
October 5th, 2010 at 7:40 pm
*oops, I meant Twisto
October 5th, 2010 at 8:27 pm
JT-
People far smarter than I, namely KLaw and Neyer, favor FanGraph's WAR over BR's WAR, for the reason mentioned. So, if I put them up next to each other and they are saying wildly different things, I'd give those two the benefit of the doubt and go with FG. That being said, it's really a minor quibble. Generally speaking, I use BR's because A) I find it more accessible and B) BR and FG are usually close enough anyway. My point was that, to the folks who complain here about those issues and therefore dismiss WAR, they may want to take a look at another calculation of WAR that SOME seem to think address those issues.
October 5th, 2010 at 8:32 pm
Here's the big question: What evidence is there that CC Sabathia pitched better than Felix Hernandez? Wins? Anything else? Anything? Buehler?
October 5th, 2010 at 11:19 pm
BSK,
I believe those concerns are more on the offensive side than the pitching side. On the pitching side FG considers only the pitcher's BB, SO and home runs allowed. That's it. They don't include team defense or the quality of the opposition. Rally's pitching WAR that we use here does. FG had a couple of posts about it last Friday.
October 6th, 2010 at 5:17 am
Sean-
Thanks. I'll try to look into it more deeply. I generally only go SO deep into the numbers behind the advanced stats, and instead lean on those with a better understanding to guide me. That includes many folks here and elsewhere. Whatever quibbles I may have with WAR, be it BR's, FG's, or another one, I realize that the concept behind the stat is a huge step forward and that even a slightly flawed WAR is better than a lot of other stats out there. Keep up the good work!
October 6th, 2010 at 7:26 am
Here's another way to look at it. Actual runs instead of earned runs +/- Rdef(runs above or below replacement that the defense is worth).
Price 71 + 2 = 73, 23.186 games pitched, 3.15 runs/gm
Felix 80 + 8 = 88, 27.74 games pitched, 3.17 runs/gm
CC 92 - 1 = 91, 26.41 games pitched, 3.45 runs/gm
Lester 81 + 2 = 83, 23.11 games pitched, 3.59 runs/gm
Weaver 83 - 3 = 80, 24.93 games pitched, 3.21 runs/gm
Felix's 2.27 ERA is misleading because 21% of his runs allowed were unearned, by far the most of any top pitcher. I find that to be a major incongruity with the notion that he has the best fielding team behind him but if people want to insist that advanced defensive metrics are a better gauge than errors, than Price let up less runs than Felix in a much more difficult division and park. When you look at actual runs allowed, Felix's 13-12 record makes a lot more sense.
As far as high leverage situations, 4 of the 5 are 1.0 (average) with CC below average at .9. I have no idea how they came to the conclusion that Felix pitched under more pressure than CC.
October 6th, 2010 at 8:11 am
Leverage relates to ingame situations. C.C. was pitching with a 5-2 lead a lot more than Felix.
October 6th, 2010 at 8:14 am
Sean, I figured that was the case with leverage but with everyone so close to average, you can probably throw that stat out. I'm really more curious what people's take is on the rest of my post. Thanks.
October 6th, 2010 at 8:23 am
As you said it is a tiny difference, so the impact on the analysis and on the pitcher's WAR is tiny as well.
October 6th, 2010 at 8:41 am
And CC actually pitched in games that mattered. I know, that's a senseless comment, but really, it's not. Context works both ways. I notice Fangraph's WAR has the difference b/w CC and Felix at 1.1 and BR has it at 0.6. and the difference b/w Price and Felix is 1.7 and not 0.7. I concede that Felix would certainly be a deserving choice, but I personally would still vote for Price or CC --yes, wins do matter and being in a pennant race matters to me. M prediction is Felix will get nosed out by CC.
October 6th, 2010 at 9:52 am
Still, Felix had 17 unearned runs on the best fielding team and CC and Weaver only had 8 on the worst fielding teams. Price also had 8 and Lester only 6 on slightly better than average fielding teams.
Somethings wrong, probably the advanced fielding metrics.
Felix has the best WHIP but if you added the errors as base runners, it's probably very close.
Felix is great, no doubt, but with all things being equal, I would have to go with wins as the tie breaker for CY in this case. Felix will get his votes but I doubt he will win.
Wins can be misleading but often times they're not. Take Steve Carlton for example, he was able to win 27 games on a team that was far worse than the Mariners.
October 6th, 2010 at 10:47 am
1. How do you conclude the '72 Phillies were far worse? They have the same winning %.
2. The Phillies were a bad offensive team. The Mariners are a historically terrible offensive team. The Phillies scored as many runs as Seattle, despite losing 6 games to a strike and playing in a league that overall scored half a run less.
3. The Phillies were not a bad offensive team when Carlton pitched. They scored 3.6 runs per game when he was in the game, which was slightly below average. The Mariners scored 3.1 RPG for Hernandez (in a higher-scoring league, remember), almost the worst in the league. Seattle scored 0 or 1 runs in 29% of Hernandez's starts. Philly did that 20% of Carlton's starts.
4. Carlton's '72 is one of the great pitching performances ever. Hernandez '10 is an excellent season but not historically great. Of course it will fall short in comparison.
Hernandez didn't win a single game this season when allowing more than 2 runs. He won three games when allowing more than 1 run. He had to pitch a great game to win. If he was slightly less than great, he could not get a win. You can argue that a pitcher may not deserve much credit if he gives up more than 2 runs, but pitchers DO get credited with wins all the time when they do that, and those numbers are what makes you think Hernandez's meager 13 wins look bad. Carlton '72 won 5 games when allowing more than 2 runs. Sabathia '10 won 7. Price won 6.
October 6th, 2010 at 10:52 am
And Raker, you can (and should) have suspicions about advanced fielding metrics, but are you seriously arguing that errors and unearned runs, in all their random glory, are a better measure? On August 15, Hernandez gave up 6 runs, none earned. This was a classic case of an error made on what should have been the third out, and then all the runs scored. They are all counted as unearned. Clearly the defense is not responsible them. I am not bothering to check the other 12 unearned runs but the flaws in what is essentially a bookkeeping measure should be obvious to anyone who watches baseball and is arguing in good faith.
October 6th, 2010 at 11:22 am
OK, the reason I had the Phils worse than the M's was simply because the M's had a better record when the respective team's aces didn't pitch.
Other than that, you make very good points in favor of Felix. You're right, the M's are pathetic with the bat.
As far as earned runs are concerned, you make my point. I'm not arguing that earned runs are a better measure for fielders as much as I'm arguing that they can be a bad measure for pitchers, and in this case a misleading ERA for Hernandez. I understand that the error continued the inning and directly let up a run but the other 5 were because Hernandez took a beating before and after the error and none of them counted against him. Are you saying that because he should've been out of the inning he's not responsible for anything that happens afterward? In Hernandez' case, the total runs allowed are closer to the reality of how he pitched than his earned runs are.
October 6th, 2010 at 11:52 am
No, that's my point, Hernandez (and all pitchers) bear some responsibility for unearned runs allowed. But you seemed to be using those all those UER to say that Seattle's defense must not have been very good.
October 6th, 2010 at 11:54 am
I will say, it is possible Seattle's defense is good (I think they are), and yet they did not play their best behind Hernandez. But I don't think error/UER totals would be a great way to show that. Anyway, if Hernandez's defensive support was actually below-average, that should be another point in his favor for the CYA. (I have no horse in that race.)
October 6th, 2010 at 12:08 pm
How does a Runs per Game of 3.17 make a 13-12 record make a lot more sense...? Allowing 3.17 runs per game would net an average team a far better winning percentage than .500.
October 6th, 2010 at 1:41 pm
It makes more sense because 3.17 is quite a bit more than his 2.27 ERA and even considering how bad Seattle's offense is, a guy with a 2.27 ERA should have better than a 13-12 record. As Twisto pointed out though, Seattle was even worse than normal when Felix was on the mound.
After considering all of the comments and evidence, he probably was the best pitcher in the AL.....uhh, and on cue, Price gives up a bomb to Nelson Cruz.
October 6th, 2010 at 2:07 pm
Raker-
Gotcha. Yes, more runs per game would generally yield a worse record. In the context of Seattle's offense, I could see a RPG around 3 correlating to a roughly .500 record. Which doesn't make sense, in terms of an assessment of the pitcher's skill. But does make sense from a team perspective. Duly noted.