CSS Button No Image Css3Menu.com
New! Search comments:
(NOTE: Relevance, Author, and Article are not applicable for comment searches)
I'm not a programmer and have generated "software" in Excel for Z-scores, regressions, projections, et cetera that could not have possibly amounted to 1/1000th the complexity of what Nate created and tweaked over the years. When I've handed my "software" to a programmer to translate it, they've often laughed at the silly complexity involved in using Excel to kludge the calculations. I imagine successfully converting Nate's work lies somewhere between flapping your arms to fly and blowing someone's head off with a thought.
We're stuck with the name PECOTA for branding reasons, but the product is no longer the same. It isn't clear whether it should be exactly the same or not--certainly asking that question is the correct thing to do. PECOTA was *not* a fantasy baseball tool at first. My impression from the beginning is that the focus is on identifying prospects before they identify themselves or prognosticating breakouts and collapses before they happen. It's too bad that building a successful business in baseball analysis requires turning to the fantasy world. PECOTA wasn't and isn't well built for fantasy--it's an aggregate tool being shoehorned into an individual game.
If I could, I'd go through the whole series and read them all again. I did that with the post game 5 and game 6 columns, and both are excellent. Perhaps a complete set of columns for the whole series would read well together.
(Tongue in cheek: after all, no one's written about the 2004 Red Sox, right?)
Maybe I'm reading the post incorrectly, but I don't see any mention of tweaks or refinements in SIERA itself. The error was in coding xFIP for comparison.
Russell, the one point of the article where some of your deeper analysis--assuming as others do, that ESPN Insider demands either shallower analysis or a truncated article--might best be stated here would be to support your signpost, "If the gentleman currently holding the ball is an outfielder, please turn left and run an additional 90 feet." There are allusions to evidence supporting this, particularly "even after controlling for the distance of the fly ball and the speed of the runner [the odds are still better than 73.2%]."
There's other things to control for that are not stated: the quality of the outfielder, the quality of the upcoming batters, the quality of the pitcher.
Could you talk about another type of flaw in the way we observe events? Not only are humans loss averse, we also do a very poor job of recollecting the occurrence rate of events. With a runner on 3rd and no outs, we take great notice of the rarest outcome (the runner getting thrown out at home), take good notice of the desired outcome (run scored), but perhaps fail to appreciate how often the runner gets stranded. Do you have a number for the strand rate of runners held at 3rd during a potential no-out sacrifice fly?
It's 0.025 outs gained per batter, and Lee yields 0.017 outs per *inning*.
Based on the rest of your math and about four batters per inning, the net is more like 6-7 runs. But that's assuming pitchers with average platoon splits against average hitters at 1.00 leverage. All those can improve with good management.
If there's a time for it, it's the heart of the order close and late. You gain more outs with a platoon advantage against above average hitters, and they're far less likely to be pinched.
With the high leverage, 6-7 runs is more than .6-.7 wins.
I speak a English.
Youk's got a best projection.
Different skill set?
The implication is pitchers will inherently have a tougher time pitching the ninth and will see their results suffer unless they have some mystical Closer skill.
How does that explain the countless times mediocre pitchers have functioned just fine as closers? See Aardsma, Sherrill, etc. Or how about the countless Closers (TM) who lose it some years? See Lidge, Wagner, etc.
I continue to lean towards the idea that pitchers (and hitters) who can't handle MLB pressure are weeded out long before this point.
We're picking nits here since I think we agree with the basic premise: the prediction systems for baseball are simply not very good at capturing the whole story when you look at one guy. We can't say for certain that Strand Rate is not included in PECOTA, but there are plenty of things that the system does not capture or can not capture.
Pitchers with Buehrle's basic stats have ERAs and WHIPs around what PECOTA keeps predicting and tend to age the way PECOTA predicts. That's not a flaw in PECOTA; it's a reason to praise Buehrle for defying his innate abilities. PECOTA also has struggled with Tom Glavine...Jamie Moyer...and Ichiro on the other side of the ball. But like anything of this nature, there's going to be outliers at both ends of a normal curve. PECOTA does fairly well, in the aggregate, for most players. I think it's better to look at it as a team-level tool.
scottlong, are you sure PECOTA has missed that badly over the years? PECOTA, 2009 reality, and past history agree on three things: BB/9, K/9, and HR/9. Where 2009 is a break from past history AND PECOTA is hit rate: his H/9 is 8.2 compared to 9.3 for career and 9.7 for the last six seasons. Is it an adjustment of his own, improved defense, or some luck? If he should get all the credit, that'd make him a truly exceptional pitcher. I'll wager on the latter two reasons more than the first. His DER this year is >74% compared to a team average of 70%.
There are some skills that I don't think PECOTA captures, and Buehrle exhibits an important one: pitching with runners on. He tends to improve with runners, and this year that's been particularly strong as evidenced by these triple slash results:
.270/.303/.460 with bases empty
.203/.257/.320 with runners on
There's a huge BABIP disparity here too, but that doesn't explain the disappearance of extra base hits.
The Yanks and Sox are different animals in this regard.
The Yankees have had some awful contracts to deal with in recent years. While it didn't kill them, the huge contracts inexplicably altered Cashman's construction of the non-Big Contract portions of the roster. Depth has reared its ugly head at various times this decade because of poor choices at backup C, backup 3B, outfield, the bench, and the bullpen. The Yankees have put all their eggs in a handful of very highly paid baskets and patched the rest of the roster together.
The Sox haven't had this kind of contract beyond the debatable pain of Manny's albatross contract. It wasn't a killer because the rest of the roster is reasonably balanced, with depth and youth everywhere. The biggest salary on the roster this year is $14m to JD Drew. Only three others make $10m or more. Compare that to the Yankees, with SIX guys paid more than Drew and nine above $10m. To quantify the difference in depth, the Sox pay their 20th highest paid player $1m, more than the Yankees pay theirs ($0.5m). All of which would lead me to be very surprised if the Sox are players in this one. The extension contract for Halladay and the Wells albatross would both be significant deviations from the Epstein era.
Simply drop the "east coast bias" thing. If the western clubs had larger run differentials than the eastern clubs, they'd be ahead on this list. Run differential (1st order), equivalent runs (2nd), and finally schedule quality (3rd) are the criteria used around here for a reason--it's backed up by data and evidence. If you so choose, you *can* debate the merit of using run differentials for analysis and prediction. Since you've read BP for nearly a decade, no one needs to detail for you the dozens of studies conducted here and elsewhere that show run differential (and, moreover, equivalent runs and schedule-adjusted runs) to be a better predictor(s) of FUTURE performance than W-L record.
This list certainly doesn't predict that the M's, Angels, and Rangers are going to crater. Compare predicted W-L based on current W-L vs. the "Hit List Factor" listed here:
Angels: 89.5 wins by current W-L, 87.1 by HLF
Rangers: 88.4 by W-L, 86.2 by HLF
M's: 82.1 by W-L, 81.0 by HLF
W-L predicts these three clubs will end up with 260 wins. HLF predicts 254. A total difference of 6 games is NOT cratering and is well within random chance anyway.
Only the Yankees create a payroll disparity. The average payroll of the other 13 AL teams is within 1-2% of the average of the 16 NL teams. And as I've said, the Yankees IL record alone certainly doesn't account for the W-L difference between leagues. If you only look at a Yankee-less AL, they've still dominated the NL for six straight years. I'm repeating myself here, but I need to provide the above again before proceeding:
Given 90 mil, if a non-Yankee AL club dedicates 10 mil to a DH, they must fill the other 13 (8 fielders + 5 pitchers) regulars and 11 bullpen/bench slots with only 80 mil left. NL clubs have 89.6 mil left for their 13 regulars and remaining 11 bullpen/bench slots. If we assume the clubs have equal talent recognition, the NL team's 13 regulars, bullpen, and bench should end up being 12% "better" than the AL team's. However, what IL records for several years have shown is that, as a whole, the NL clubs are *not* recognizing talent as efficiently as AL clubs.
Bingo. Two words stand out for me: "work harder." It's not the money difference. The Yankees account for just about the entire difference in average payrolls between the leagues but don't account for the difference in interleague performance alone. It's effort across most of the league. For some time, the NL has effectively colluded. Too few teams are stepping up and giving maximum effort at all levels of the organization. There's not enough incentive to try harder because almost every NL club has a shot each year just by showing up--better to relax and wait your turn than to try harder. Consider your favorite team: are they expanding Latin American scouting/development programs? Adding scouts in Asia? Is your favorite team using the six-years of indentured servitude rookies must suffer as well as they could? Does your club have a lot of dead weight contracts?
Despite playing about 50% of their games intradivision, the AL East is 134-106 against non-division opponents. At .558, that doesn't look all that gawdy, but it'd be good for 1st place in half of MLB's divisions. Runs scored backs up the story: they're +173 runs combined. 240 games is hard to argue with.
Quite simply: the AL is at or above .500 in interleague play for several years running...IN NATIONAL LEAGUE PARKS. If the DH really does make a big difference, it can only mean that the AL is even more superior right now than the numbers suggest.
Bah I say. Boyer and Campillo were the THIRD and FOURTH relievers that inning. When Boyer entered, he had the tying run at the plate with the bases loaded. That moment was the epitome of high-leverage: on the road with a chance to stick the first nail in the Phillies coffin on April 8. To make matters worse, Boyer and Campillo hadn't appeared since spring training. Both of your top-flight relievers, Gonzalez and Soriano, pitched in low leverage situations the night before and were unavailable, so you're stuck trying to stop the bleeding with #5 and #6 on the bullpen depth chart. Had the better guys been available, they'd certainly have been able to be ready to be the THIRD or FOURTH relievers in the inning.
The alternative history is: Soriano or Gonzalez warm up for the 8th inning on 4/7 *just in case* while the lesser pitchers handle those last six outs. This opens the possibility that one or both of them remains available 4/8.
As for warming up, human beings can loosen up every day *to a point of no return* without any impact on their availability the next day. It doesn't take a PhD to know that these guys can stretch and soft toss to their hearts content each night if need be.
Yankee fans better hope this is the last reason to link Teixeira and Damon.