Monday, October 5, 2015

The Best ERA Doesn't Mean You're the Best Pitcher, the Best Record Doesn't Mean You're the Best Manager, and More End-of-Season Lessons

Obviously, I can't vote for MVP, Cy Young, Rookie of the Year, or Manager of the Year; I'm not a professional baseball writer. But the great thing about the internet is that you can find a club no matter who you are. The Internet Baseball Writers Association of America lets even us amateur bloggers into its ranks, and one of the perks is that we vote on the IBWAA's own versions of these end-of-season rewards. The weekend I spent researching these awards made me question how I believe runs are prevented; they forced me to pick sides in the debate over defensive metrics. They led me to uncover the secret brilliance of Sergio Romo and to quantify how "clutch" Josh Donaldson is. And they made me angry—again—over how stupidly people interpret the Manager of the Year awards.

Here's how I voted for the IBWAA and how I would have voted for the official, BBWAA awards:

MVP


American League

It's a close one between Josh Donaldson and Mike Trout. Although WAR is a great guide for asking the question of who was "most valuable," it can be subjective. My heart wants to give it to Trout out of sympathy for getting boned by Miguel Cabrera for a few years in a row—and he has been better offensively. Trout has a better slash line at .299/.402/.590 despite playing in a worse hitter's park; Donaldson's gaudy runs scored and runs batted in totals can be discounted as having to do more with the excellent lineup that surrounds him. But Donaldson has a few subtle advantages. He was an excellent baserunner, going 6 for 6 in stolen-base attempts (indeed, he hasn't been caught since 2013); Trout, meanwhile, was a sloppy thief, going only 11 for 18. Also, despite Trout's reputation as a fielding wizard, Donaldson played better defense according to both UZR and DRS—Trout, meanwhile, was merely average. Finally, Donaldson, in context, was a more valuable hitter. That means that he got hits when they counted the most, as measured by stats like Win Probability Added and Run Expectancy. The Blue Jays' prolific offense also helped Donaldson score better in these stats, but Donaldson also has a comfortable lead over Trout in Context-Neutral Wins (WPA/LI). That means, simply, he contributed more to his team's wins than Trout. That's a pretty good definition of MVP to me.

Of course, contrary to popular belief, MVP isn't just an award for hitters—it's Most Valuable Player, and pitchers are players too. Therefore, we mustn't forget about the best pitchers in the AL, which I'll explain in more detail in the Cy Young section below. Dallas Keuchel and David Price are extremely close in that race, but for MVP, Keuchel's surprising 13 Defensive Runs Saved—more than double any other AL pitcher this year!—put him over the top in overall value. Downballot, Lorenzo Cain may have been better on a daily basis than Manny Machado, but Machado had a hundred more plate appearances. Kevin Kiermaier deserves recognition for his historic defensive season—contributing to lowering the ERA of an entire pitching staff certainly qualifies as valuable. On the flip side, sure, Nelson Cruz gave away some runs on defense, but his OPS+ was third-best in the league, and no one but Trout and Donaldson added more win probability with his at-bats. I think he deserves to be squeezed in.

1. Josh Donaldson
2. Mike Trout
3. Dallas Keuchel
4. David Price
5. Manny Machado
6. Lorenzo Cain
7. Kevin Kiermaier
8. Mookie Betts
9. Sonny Gray
10. Nelson Cruz

National League

Amid all the uncertain votes elsewhere, there's great comfort in the fact that there's no way I can screw this up. Bryce Harper is having an undeniably historic season. His OBP of .460 and SLG of .649 reveal a hitter who simply belongs on a higher plane. In historical context, it's even more impressive. Given that this is an era of low offense, he has the best adjusted runs-created score (wRC+) of anyone not named Barry Bonds since Mark McGwire in 1998—and the best score of any clean player since 1994. He's also played run-saving defense and been a positive force on the bases. Spare me any talk about how the Nationals didn't make the playoffs; without Bryce Harper, they'd have lost 90 games. He has been the most valuable—the most value added—to any team in baseball.

It's a good thing, too, because if he weren't around the race would be too close to call between Paul Goldschmidt and Joey Votto. Votto's second half has pushed him to an insane .427 wOBA with more walks than strikeouts—even though Goldschmidt plays in a much better hitter's park. But Goldschmidt had 18 Defensive Runs Saved this year—damn impressive for a first baseman—and he also stole 21 bases. In the name of consistency, I erred on the side of defense and baserunning here too—but ask me next year and I might have a totally different answer.

AJ Pollock and two stars from Chicago finish up the ballot, edging out Jason Heyward. Heyward's defense was great, but Kris Bryant's ability and willingness to play four different positions, plus the Fielding Bible's surprising bullishness on Rizzo's own defense, were nothing to sneeze at, either. Finally, I'll explain my treatment of pitchers in the following section, though again it's worth noting that Zack Greinke was the NL's best defensive pitcher and that factored in here.

1. Bryce Harper
2. Zack Greinke
3. Jake Arrieta
4. Clayton Kershaw
5. Paul Goldschmidt
6. Joey Votto
7. AJ Pollock
8. Max Scherzer
9. Anthony Rizzo
10. Kris Bryant

Cy Young


National League

So much ink has been spilled on this award, including by me. Suffice it to say that it's effectively a three-way tie between Zack Greinke, Jake Arrieta, and Clayton Kershaw. Each one looks the strongest depending on how much control you believe pitchers have over run prevention vis-à-vis their defense. There are so many complicating factors here that I've attempted to use to create some daylight between the three. Take, for instance, the happy coincidence that Greinke and Kershaw are on the same team. That means that, generally, they've had the same defense behind them. It's not literally the same defense, of course, but it's more directly comparable than the fielding between Kershaw vs. the fielding behind Arrieta, at least. And yet, for some reason, Dodgers fielders are helping Greinke a lot more than Kershaw. That could be a tipoff that Greinke really is having some extra effect on hitters. Or it could just be luck. Or it could be Greinke's choice of catcher, and it's unclear whether that should count as something under his control.

Greinke has also been more consistent from start to start, but, on the flip side, that has meant he hasn't soared to the same heights. Arrieta and Kershaw each has four complete games and three shutouts; Arrieta has a no-hitter, and Kershaw has three games with Game Scores over 90. Greinke hasn't worked as deep into games and, as Dave Cameron of FanGraphs crucially notes, has much worse numbers against hitters the fourth time through the order—signaling that taking him out relatively early has been a key strategy by the Dodgers to keep his ERA depressed. That's a strike against him in my book. Kershaw and Arrieta have met more challenges.

After much agonizing over how I wanted to assign credit to Greinke's stinginess, I came to a compromise/copout. When asking the question of who was the most valuable player, you have to consider what the player accomplished, in the past tense—regardless of whether they "earned" it. But the Cy Young Award is different—it's meant to reward the best pitcher. Skill at pitching is measured in other ways too—strikeouts, lack of walks, durability, etc. When evaluating who has the most pitching talent, it is desirable to strip away defense as much as possible. So while Greinke topped all pitchers on my NL MVP ballot—because, when he was on the mound, he (one way or another) gave up the fewest runs and put the Dodgers in the best position to win—I cannot place him first for Cy Young. His strikeout rate (8.08 per 9 innings) is inferior, and he lucked into favorable scenarios like stranding 86.5% of runners on base. Instead, I decided to vote for the man who everyone knows is the best pitcher on the planet. If you put things in perspective, it's clear who the consummate artist on the mound is, the man who can exert the most will on the baseball when it's placed in his hand. That's Clayton Kershaw.

1. Clayton Kershaw
2. Jake Arrieta
3. Zack Greinke
4. Max Scherzer
5. Jacob deGrom

I also puzzled over whom to give my last vote to. Gerrit Cole and Jacob deGrom both had strong cases, and Cole had pitched more innings, which I usually look favorably upon. But deGrom was skipped in the Mets rotation several times in a conscious effort to keep his arm fresh, so it can't be held against him. There are also several ways in which deGrom was slightly but noticeably superior to Cole: K/9 (9.66 to 8.74), xFIP (2.92 to 3.16), hard-hit percentage (25.7% to 29.5%), percentage of strikes thrown (68.4% to 65.5%), and swinging-strike percentage (12.7% to 10.2%).

American League

This race is actually harder to decide than the NL's. There were at least clear differences between Greinke, Kershaw, and Arrieta—it was just a question about how to weight them. In the AL, you have two dominant starters with nearly identical pitching lines in Dallas Keuchel and David Price. Keuchel has a 2.48 ERA; Price is at 2.45. Keuchel allowed 185 hits, Price 190. Each allowed 17 home runs and twirled three complete games. Keuchel has 216 strikeouts to 51 walks, and Price has 225 strikeouts to 47 walks.

At first I leaned toward Price. He struck people out at a slightly higher rate and walked them at a slightly lower one. Following my logic with the NL Cy Young, Price's 2.78 FIP should give him the edge over Keuchel's 2.91. But something interesting happens when you normalize home run rates. Because Keuchel allowed a flukish and awful 13.6% of fly balls to go for home runs (a ratio that can usually be chalked up to luck), Keuchel's xFIP of 2.76 is much better than Price's of 3.24. And, judging by Skill Interactive Earned Run Average (SIERA), which assigns different values to hits depending on how well they were struck, Keuchel has a 2.84-to-3.27 advantage. The reason, methinks, is Keuchel's extreme ground-ball tendencies; 61.7% of the hits off him were on the ground, best in the AL. Other than strikeouts, ground balls are the best thing a pitcher can produce, because they are most easily converted into outs and rarely go for extra bases. Price, on the other hand, gave up a lot more fly balls (36.4%) than Keuchel did (only 19.6%). While strikeouts were an approximately equal part of both pitchers' games, Keuchel had the added weapon of the ground ball. He should get credit for this more than the typical fielding event. To put the icing on the cake, Keuchel also had the AL's lowest hard-hit percentage (21.2%), while Price's was worse than Kyle Gibson's and Chris Tillman's.

Otherwise, Chris Sale clearly deserves recognition for his 6.52 K/BB ratio. The man led the league by striking out 32.1% of batters, and his fielding-independent numbers were best in the league. His 3.41 ERA was a tad high, especially by his standards, but still ranked 10th in the league—and no one other than Keuchel, Price, and Sonny Gray had an ERA under 3.10 anyway. Speaking of Sonny Gray, sabermetrics aren't a huge fan of his 2015, but I still am; his ratios may have been merely solid (7.31 K/9 and 2.55 BB/9), but, similarly to Keuchel, he used a high ground-ball rate to avoid hits. Given that a low hit rate has been a career-long trend for him, it's safer to say it's not all good luck. (He also allowed fewer line drives than Keuchel, Price, and Sale.) Finally, I compared Chris Archer and Corey Kluber head to head for the last slot. Archer has a better ERA, but much of that is due to the park he pitches in. Kluber proved superior in most peripheral stats, especially walks allowed, strike percentage, and hard-hit percentage (Archer's was a surprisingly poor 32%).

1. Dallas Keuchel
2. David Price
3. Chris Sale
4. Sonny Gray
5. Corey Kluber

Rookie of the Year


American League

Let's play a game. Player X is a rookie shortstop. He played 99 games this year after being called up in June. He hit double-digit home runs in about 435 plate appearances and stole 12–14 bases, and he's now an American League Rookie of the Year favorite. Is it Carlos Correa or Francisco Lindor? Mwahaha, trick question! It was both! The pairing that many have suggested are the new Jeter and Nomar have had eerily similar stories this season. On offense, Correa has the upper hand, since he's jacked 22 home runs and has walked more (Lindor has a higher OBP, but he's been quite lucky on balls in play). However, Lindor has been a revelation with the glove, saving twice as many runs as any other AL shortstop in half the playing time. According to UZR, Correa has also fielded his position terribly, and that dooms him.

The choices after these two are slim pickings, with Billy Burns leading in both rWAR and fWAR. However, this has as much to do with his 555 plate appearances as his 26 stolen bases; he's been a compiler, not a particularly outstanding player (a .317 wOBA). The truly impressive rookie, for me, has been Miguel Sanó, who debuted on July 2 and hasn't stopped hitting home runs since. In a lot fewer at-bats, he's been almost as valuable as Burns, hit almost as many home runs as Correa, and shown insanely good plate discipline. If he qualified for the batting title, his OPS+ would be between José Bautista's and Chris Davis's.

1. Francisco Lindor
2. Carlos Correa
3. Miguel Sanó

National League

Whither Joc Pederson? His bizarrely bad second half dropped him out of contention for this award, even though I'd wager his career will eventually surpass all but my top NL rookie's. I voted for Kris Bryant for MVP, so it should be fairly obvious that he's my pick here. The following are the categories he led NL rookies in: home runs, runs, RBI, doubles, and, despite some service-time shenanigans, plate appearances. A couple of Cardinals, Stephen Piscotty and Randal Grichuk, matched him in rate stats, but you'd have to smush them together to reach Bryant's counting stats. Piscotty and Grichuk did it in spurts; Bryant has sustained his excellence for a full season.

Matt Duffy was thrust into an everyday role this season and responded by hitting .295 in a tough park for hitters along with 12 runs saved on defense; that puts him in a comfortable second. And, although his season was sadly cut short by a hard slide, Jung Ho Kang still outhit the likes of Odubel Herrera and Addison Russell with a .287/.355/.461 slash line.

1. Kris Bryant
2. Matt Duffy
3. Jung Ho Kang

Manager of the Year

Here's the thing about the Manager of the Year award. It's always used as a proxy for "most surprising team"—when a team significantly outperforms its preseason expectations, it's assumed that the manager's leadership got them there. That's baloney. When a team significantly outperforms its expectations, it's because the pundits were wrong and the players themselves overperformed. Furthermore, that assumption implies that who's a good manager changes from year to year. Last year, Matt Williams was a genius because the Nationals made the postseason. This year, he was run out of town on a rail. But do we really think he just forgot how to manage? No—he was never that good to begin with.

The people who are the best managers—and who are truly worthy of this award—do not change much from year to year. They're the people who are loved and respected by their teams, who avoid controversy, and who generally make shrewd strategic decisions. These traits don't ebb and flow depending on how well his team plays baseball. So, to sum up, my Manager of the Year ballots won't be conforming to the conventional wisdom.

National League

Joe Maddon has been the best manager in baseball for quite some time—beloved by his players, new age in his tactics. That didn't change this year in his first season with the Cubs, where he blended in seamlessly and got top-notch effort out of a young team. Even though his Giants underperformed, Bruce Bochy is still one of the shrewdest tacticians in baseball, and his staying power through good times and bad has to be respected. Finally, Clint Hurdle has quietly become one of the stat-savviest managers in baseball, and his adept use of the bullpen led the Pirates to strongly overperform their Pythagorean record in 2015.

1. Joe Maddon
2. Bruce Bochy
3. Clint Hurdle

American League

This is the rare case where my pick is probably the same as everyone else's. Jeff Banister has really impressed in his first season as Rangers skipper, quickly earning the trust of players while implementing unorthodox and saber-friendly tactics. It's a close call, though, between him and Joe Girardi, who has quietly done excellent work in New York for eight years now. In the toughest media environment in the sport, he has avoided major controversy—no small feat, considering the return of Alex Rodriguez this year—and is also the manager who uses his challenges most efficiently (a study conducted last year found he got 79% of challenges overturned, the best in baseball by a country mile). Finally, I was torn several different ways for third place. If there were an award for Most Improved Manager, I'd definitely vote for Ned Yost, who has shown shocking development as a tactician since last year's playoffs. AJ Hinch's Astros, meanwhile, have made smart use of the defensive shift, but that is probably a directive from the front office. Bob Melvin is also typically one of the game's steadiest hands on the tiller—even earning a contract extension this year despite a terrible record—but he shares with Hinch a team that drastically underperformed its Pythagorean record. And Terry Francona is the public face of probably the most positive, well-run workplace in sports, and he's managed to maintain strong leadership through a disappointing season and turnover in the front office. Perhaps against my own better judgment, I decided to go with the guy who made everyone look silly when they questioned even his most bizarre moves all season long.

1. Jeff Banister
2. Joe Girardi
3. Ned Yost

Reliever of the Year

Unlike the BBWAA, the IBWAA also votes on an award for best reliever. There is no stipulation that it has to be a closer, however, which is going to lead to some unorthodox choices here—be prepared.

American League

Dellin Betances pitched 84 innings out of the bullpen, striking out 14.04 batters per inning with a 1.50 ERA. His walk rate was high and he was lucky, but he still had elite fielding-independent numbers. Orioles closer Zach Britton and Yankees closer Andrew Miller were arguably even more dominant, though over fewer innings. Miller had the lower ERA at 1.90 and struck out a whopping 40.7% of batters, but Britton offered better control at 1.92 walks per nine. The tiebreaker ended up being how good Britton was at getting ground balls (79.1%), contributing to a 1.21 SIERA. Miller also allowed a high 29.8% hard-hit percentage. I wish I could've found room for David Robertson, with his elite 6.62 K/BB ratio, and Carson Smith, with his high innings total in higher-leverage situations than the rest of this crew. Some might ask where Wade Davis is—his 0.94 ERA by far led all relievers—but he pitched in relatively low-leverage situations, his K/BB ratio is merely solid, and his swinging-strike percentage trailed all the aforementioned names, indicating his stuff wasn't as sharp. Instead, he lucked out with a .200 BABIP and 92.2% left-on-base percentage.

1. Dellin Betances
2. Zach Britton
3. Andrew Miller

National League

If only Carter Capps had stayed healthy for the full season—he'd be a lock here. Instead, the NL doesn't quite have the same caliber of reliever as the AL. Kenley Jansen pitched the best among the remaining crop, but he too was injured for a chunk of the year. The question is whether his 10 K/BB ratio (first in the NL), 4% walk rate (first), 0.78 WHIP (first), 2.14 FIP (fourth), 1.43 SIERA (first), 11.1% line-drive rate (first), and 16.6% swinging-strike percentage (fourth) are good enough for his low innings total (52.1). Sergio Romo also had awesome ratios and an uncannily low FIP (1.91), but he pitched fewer than 60 innings and gave up too many runs to go with it (2.98 ERA). Jeurys Familia and Francisco Rodríguez were about equally stingy when you adjust for park factors, but Familia threw 20 more innings. Thank goodness there's a clear number one, at least: sometimes the best ERA belongs to the best pitcher. Like Betances, Aroldis Chapman had his struggles with walks, but he also struck out 15.74 (!) batters per nine innings to make up for it.

1. Aroldis Chapman
2. Jeurys Familia
3. Kenley Jansen

No comments:

Post a Comment