Wednesday, November 14, 2012

Just Win (and Have Your Team Lose), Baby!

W. Ross Clites
Your City Sports-Cleveland


Saint Louis, MISSOURI -- At the end of the day, the biggest push back I received from my "Mike Trout for MVP" campaign centered on the indisputable fact that Miguel Cabrera led his team to the World Series and Trout did not even make the playoffs. While I will not dwell on this debate, nor carry it over to this article, it is the fundamental basis of this piece. 

Playoff appearances are so vital to that award and yet so trivial to others, primarily the Cy Young Award. I want to investigate why that is and if it needs to change.

Recent history has shown that pitching for a successful team is not a prerequisite for postseason awards. Since 1995 -- when making the playoffs got "easier" -- actually doing so seemingly became antiquated. Ironically, you are now better served to miss the current expanded playoffs in order to be labeled the game's best pitcher. Is this a coincidence? Have we unknowingly created a new stat: the conditional win, which overemphasizes (and rewards) individual wins on a garbage team?

David Price of the Tampa Bay Rays and R.A. Dickey of the New York Mets were just named the 2012 winners of their respective league's Cy Young Awards. This mirrors what I projected using Pitcher Rating, meaning both are also taking home the (still fictitious) Walter "Big Train" Johnson Award. The latter award is strictly objective; the result of a complicated formula that does not take team record into effect whatsoever. The pitchers involved have a full season to put in the work. Come October 4, 2012, the numbers are what the numbers are. I take the pitcher in each league with the highest value... end of story.

Meanwhile in crazy land, the Cy Young is still voted on by the Baseball Writers' Association of America (BBWAA). This annual process had me assuming Justin Verlander of the Detroit Tigers would capture the AL award. Human instinct remembers the most recent performance and values the eye test. The show that Verlander put on in the 2012 postseason was unworldly. He was everything an ace should be, and looked the part of the game's best pitcher.

But performances in the playoffs always have been, and always should be, discounted from the Cy Young equation. This is easier said than done. Unless a writer's ballot was cast before Game 1 of the Division Series, the greatness of Verlander had to be fresher in the mind.

My assumptions were surprisingly wrong. The BBWAA got it "right." In the tightest vote since 1969, by a margin of only 4 points (or the repositioning of a single first-place vote), Price won the award. Whereas I should be happy, I am actually the complete opposite. This is because I am a contemporary BBWAA contrarian. Writers are making it harder to oppose what they stand for -- and vote for -- when their credentials are a roaming target. 

Just when I thought I had the BBWAA figured out, I am now left wondering -- along with all Detroit fans, and most educated baseball fans -- why it wasn't Verlander. In an even two-man race, he had the upper hand in WHIP, strikeouts, innings pitched, complete games. You know, everything recent votes have shown the writers value. It should have been (and thus should not have been) Justin Verlander with the Cy Young. How can I argue against it when they pull a statistical 180? They are "getting it wrong" all wrong.

Is this not the same BBWAA that voted for Felix Hernandez and his 13 wins two seasons ago? They are the reason I set out to make an objective, statistically-based postseason award in the first place. It seems that each year what Cy Young voters value changes. Notify the press, wins are back in vogue in 2012. 

Perhaps the writers understood how hypocritical it would be to award Miguel Cabrera the MVP and Justin Verlander the Cy Young. You cannot have the offensive Triple Crown stats carry an MVP candidacy, while the pitching equivalent is overlooked. A voter that says WAR and Runs Saved are not applicable in postseason awards cannot come back at me with WHIP or Run Support. Plus, wins and RBI are equally dependent on extrinsic factors, mainly the quality of the team around the player. So, I see a situation where votes were backed into a corner. If they wanted Cabrera's RBI supremacy to have merit, they begrudgingly had to give Price's 20 wins their due.
    
Forget Obama/Biden vs. Romeny/Ryan, 2012 was a baseball fan's ultimate showdown. The American League Cy Young and MVP votes were a statistical crossroads. The Cabrera/Price ticket spoke to the old guard, where batting average and wins both maintain meaning. The Trout/Verlander camp was a sleek machine of sabermetrics, where hitting .400 means a player had an atrocious season (using OPS standards) and a pitcher's win-loss record has been eliminated from the baseball vocabulary.

I, for one, will admit I have been talking out of both sides of my mouth; pulling for Price and Trout. I admit it because I see the day when the Jumbrotron at a Major League stadium displays the lineup of a team with Player Name, Number, Position, and OPS (not batting average). But I cannot allow myself to envision a future where pitchers -- warming up in the bullpen -- do not have their win-loss record displayed. 

It would be nice if the BBWAA followed suit and admitted double-speaking. All we ask is that they share their vision of the future of the sport they care so much about. Pull back the curtain and let the fans see what the rules of the game are. It sure feels like writers wanted Cabrera and Verlander, but could not pull the trigger. If that is true, say so. Help us better understand the process, instead of releasing the voting results (in a zero explanation, take-it-or-leave-it fashion) and disappearing until the Winter Meetings. 

Like Congress, baseball fans must work across the ropes, figure out what standards they want, and demand it of the writers. Taking the guess work out of the vote begins with a simple survey. What consensus statistics do we value the most? The BBWAA will never come out and tell the general public what things they look for, so we either need to lead them to water or change the entire secret-vote process. 

To show you how far we are divided on these issues, I brought in the opinion of a friend -- currently working for the St. Louis Cardinals. He recently wrote to me "if we didn't have the wins stat now, and somebody proposed it as a new idea, most people would think it was useless." The unfortunate thing is I agree with his assessment. I do not agree with the stat's fall from grace, but I certainly cannot argue the statement's validity.

I want to use wins, both by a pitcher and their respective team, to dive deeper into the comparison of Cy Young past vs. Cy Young present. After all, we have to see how we got here before we can plan a course of where to go. Let's start by looking at the team these players pitch for.  

Postseason awards and non-playoff baseball:

Unknowingly, today's writers have laid out a new formula for winning both major postseason MLB awards. Since the days of Andre Dawson (1987) and Cal Ripken, Jr. (1991), the tides have slowly receded back to the norm in the MVP discussion. You now "have" to be on a winning team, and more-times-than-naught be playoff bound. 

The data backs this up: from 1992-2012 (excluding the 1994 season in which there where no playoffs), there have been 40 MVPs chosen in Major League Baseball. Of the 40 that were selected over that span, only seven times did the recipient fail to make the playoffs. Those players were Albert Pujols (2008), Ryan Howard (2006), Alex Rodriguez (2003), Barry Bonds (2001, 1993), and Larry Walker (1997). Of those seven occurrences, Bonds in 2001 and Howard in 2006 would have both made the playoffs in the newest format. Today, with ten playoff teams, you need not apply if you could not get your team into the top third of the league. The sample size is large enough to say with certainty, the voters officially have a "No playoffs, No value" stance. This should not be news to anyone, as it was the thrust of most people's Cabrera counter-argument to my Trout campaign.  

But with the MVP now requiring the player's team to be in the postseason, the Cy Young has strangely gone in the opposite direction. Of the last five seasons, only Roy Halladay (2010) and Justin Verlander (2011) pitched their way to the postseason. That means a mere two out of the last ten recipients (20%).  

Fitting, that in a year when more teams than ever make it to the playoffs, two pitchers from middle-of-the-division teams took home the top pitching prize.

Take it back further -- to when the playoffs expanded to eight teams -- and the numbers only slightly improve. In these 18 years (now counting Price and Dickey) both leagues are dead even with nine playoff-bound Cy Young winners and nine guys that played golf all October. This 50/50 split shows the playoffs are not on the mind's of contemporary voters as much as previously thought. At least not for pitching accolades. 

This is certainly not how it used to be. The first eight American League Cy Young Award winners played on a playoff team, including 1967 and 1968 where there was still only one AL "playoff" team -- back when it was pennant or bust.

In the first 19 years of the Cy Young's existence the recipient(s) came from a postseason contender 21 out of 27 instances -- a 78% correlation. The Cy Young was like the MVP voting of present day; you better have some kind of season to win it on a non-playoff team.

This is why I contest Nolan Ryan was snubbed out of the Cy Young for his entire career. The modern ideology surrounding the award had not been fully embraced yet. In 1973 and 1974, Ryan had the best pitching seasons, hands down. But his California Angels were not in the World Series and, at the time, playing in those games (on national television) was a prerequisite for individual awards. 

It took voters one more year, 1975, to finally honor an American League pitcher who did not appear in the postseason (Jim Palmer). Their "No playoffs, No value" stance softened as talk of only four postseason contenders (out of 26 franchises) started to resonate. After all, how can a guy who plays in less than 25% of his team's total games supposed to assure a league title? His greatness must stand alone -- set aside of team win percentage -- as it can only go so far to determining the outcome of a 162-game season (cough, cough, why we there should never be an MVP that is a pitcher).  

So what has changed? How have qualified candidates, ordinarily buried in the standings, risen to elite status? Exposure has sure evolved; our 24/7 sports networks grant an ability to watch every team coast-to-coast. But the answer is actually simpler than that. For several decades, the league continued to expand, while the playoff participation stayed the same. In the era of at least thirty teams per professional sport -- and even with its second Wild Card expansion -- Major League Baseball is not even close to NBA or NHL in its postseason representation. In short, writers would be stupid to shut out 66% of qualified Cy Young candidates simply because their team failed to make it to the playoffs; something a little too far out of their individual control.  

If the NHL chose the Vezina Trophy (best goaltender) off the same criteria, 53% of the starting goalies would be eligible. This is why hockey can get away with an unspoken "No playoff, No value" policy; no matter how terrible the offense is, no one can be the best goalie if their team is in the bottom half of the league. 

With baseball there is a grey area of being above the midpoint of teams in the league and still not appearing in the playoffs. Being above .500 is not guaranteed a thing in MLB standings.  

There is also the issue of utilization. For the second straight season, the Major League lead in games started was only 34 games (21% of team's games played). Once again seeing similarities between hockey and baseball, this would be the equivalent to an NHL goalie making 25 starts in net. Essentially, this means a starting pitcher is like a back-up goalie; neither of whom should be held all that accountable for the team's end-of-year record.

Not his fault, but Justin Verlander was utilized less often than Jhonas Enroth (an athlete you have likely never heard of) was for the Buffalo Sabres in 2011. This is tongue-and-cheek and certainly not a comparison on their value/quality of play, but something to ponder. The point is, there are other players in the lineup every single day that shoulder the responsibility of team wins far more than pitchers. The past had it wrong; non-playoff pitchers can have the best statistical seasons. 

Have we swung too far in the other direction?:

Are voters now starting at the bottom of the standings when looking for the best pitchers in each league? Modern sabermetricians say "a game is a game is a game" and that drives me bananas. In how many more meaningless games do pitchers out of contention take the mound? SABR does not account for this, nor the term "clutch", because they would like you to believe that a game started by CC Sabathia in late September against the Orioles (in Camden Yards, on Sunday Night Baseball) is the same level of stress as Felix Hernandez feels pitching at home against Minnesota (throw in some shadows, making it even more impossible to hit). Scenario 1: both teams in playoff contention. Scenario 2: neither team has people caring for anything but their own contracts.  

Playing for something is a huge deal. We need to pump the brakes before we start handing out the award for "most outstanding pitcher" solely to the individual with the lowest ERA, WHIP, BABIP, blah blah blah. They are all nice in explaining the nuanced facets of the game, but fall short in explaining a simple outcome: who won. 

When a pitcher does not have his best stuff on a given day, what is the one -- and only -- statistical category that can still benefit? Picture a tough divisional opponent in late September, your team is two games back in the loss column, hostile crowd, and an unfavorable match-up for your offense: that ERA is going up, that WHIP is going up, that run support and strikeout totals are going down. But if you somehow do even the bare minimum to qualify for a win, you did your ultimate job; what owners pay you to do. These days balance with the hard-luck 1-0 CG losses. It is just how baseball has "worked" for centuries. 

Meanwhile, across the country, the pitcher chasing you in the Cy Young "race" mopped up a comparable cellar dweller to his own -- 7.2 IP, scattering 1 run on 4 hits. If this were the Heisman Trophy, analysts would say his stats jump ahead of yours. But why? This isn't football. You both won the game, and the fact that your win was far more stressful/meaningful to the team has serious meaning. Or should.        

I am not advocating that we dissect every Cy Young candidate in terms of who they pitched against and when. That is not worth anyone's time. The writers simply need to make the assumption that anyone pitching for a playoff contender ran into more of this hypothetical adversity along the way and move on. A game is not a game and wins do matter. 

I am all for the greatest pitchers coming from losing teams, as long as they put up a healthy win total. The 2010 Felix Hernandez Cy Young, heralded as a victory for baseball nerds like me, was short-sided and frankly a travesty. Pitching in meaningless games is pitching in meaningless games; you cannot change the opponent or your own team to up the adrenaline level. What you can do is go out and earn the "W". Hernandez (2010) lost a majority of his starts, while Dickey (2012) won his fair share.  

The pitching victory is back in the SABRmetric vernacular... conditionally

These days, the modern Cy Young is evolving into something that resembles Andre Dawson's infamous NL MVP win in 1987. Wins matter, but only when talking about percentage of the team's total. There seems to be a new focus on a single starting pitcher's contribution as it relates to the whole. It is as if writers are saying "look at what this guy could do on an inept team; no one on his team was even half as good." 

This is especially ironic when it comes out of the mouth of a sabermetrician who despises the pitching victory stat. They are all about rewarding a 20-game winner on a 60-win team, but hypocritically quick to yawn at 20 wins from a member of a division-winning team. It proves that writers are looking for pitchers to come from non-playoff teams, period. 

Gio Gonzalez was not that far behind Dickey's accolades, but the vote was not even close (93% to 42%). That large gap shows voters bit hard on win shares as the "it" topic for this year's Cy Young. Dickey winning 20 of 74 (27.0%) team wins crushes Gonzalez's 21 of 98 (21.4%). Dickey's Mets were also 24 games behind Gonzalez's Nats; that makes him a much sexier pick by today's standards. 

Option 1 is to win often and have your team lose a lot (with decent supporting stats). Option 2 is to have dazzling SABR stats on a good team, where your personal win total won't matter. 

If that is really the new formula, why did Verlander not win?:

Justin Verlander was the classic case of Option 2. He beat out David Price in all the gobbledygook figures that you would see under Baseball-Reference's "Player Value" tab, including everyone's favorite, waaWL%. His team was also the AL Central Champions. So what happened? How did the voters go against their own set of made-up rules? 

You could argue that Price winning 20 of 90 (22.2%) and Verlander winning just 17 of 88 (19.3%) would give the edge to Price using the Dickey Option 1 theory. I discredit that because both teams were in playoff contention throughout. Neither were on teams that lost enough for their individual win-share to drop jaws.  

Despite the win total being greater, Price's Rays missed out on the playoffs, while Verlander's Tigers got in. Did Price win because he did not make the playoffs? Perhaps the bias against pitchers headed for the postseason is real. The jury is still out on this one. I am not going there; future votes will tell.

Either way, Verlander vs. Price goes down as the second-tightest race in the history of the American League. It was obviously a split-decision by a large majority of the writers. If Verlander had the edge in all the SABR stats and we already touched on the fact that neither won over 23% of their team's total games, so there is only one thing to believe. There is a new Option 3 that is slowly building steam in the underworkings of the BBWAA. And it is something that many of us have been calling for. 

Option 3 is to have a superior win-loss record. It will definitely take more time for this one to gain traction. Based on this year's vote total it appears it was used by many to break a virtual tie. 

The various arguments against David Price that are overrated compared to his record

It is a given that a pitcher's win-loss record has evolved into something out of their own control. Well here are some things that get tossed around nowadays that I feel are overblown.

Fewer Complete Games: Fans become restless if the last nine outs of the game are not quick and painless. You could argue this is no different than baseball played 100 years ago. The difference lies in who records these final outs. A starter, going through the opponent's lineup for the fourth (maybe fifth) time, rarely gives a manager the best option to win anymore. The complete game, as it pertains to a stat that boosts a Cy Young candidate's stock, is ridiculous. It is even more out of the control of the starting pitcher as their record. The CG is more of an admission that a team has seven other pitchers -- with established late-inning roles -- that are not qualified to get major league hitters out. 

The nightly decision weighs a fresh pitcher, with potentially a new arm slot or handedness, against that of a laboring pitcher, with the stamina tank ~75% full (or ~25% empty) and no ability to throw up anything that the batter has not already seen. It is why nine times out of ten managers opt for the bullpen. And this is not the fault of the starting pitcher. One player should not get overly rewarded for a gambling manager or a fledgling pen. As the game gets more and more specialized, the belief is that slew of bullpen pitchers can more effectively hang zeros on the scoreboard. This is why the CG is on Major League Baseball's endangered species list.   

Fewer Innings Pitched: 27 1/3 more IP do not prove to me that Verlander is a better pitcher than Price. It is a nice footnote, but also out of Price's control. This is the same argument as the CG, but with a twist. Complete games do not show up in the formula for computing any other stat. It is an independently "dumb" column on the stat sheet; it tallies up instances that apply and nothing more. Innings pitched, however, have their hands in all sorts of statistics: ERA, WHIP, SO/9, H/9, BB/9, HR/9, on and on. IP is like garlic, an overpowering ingredient that shows up in so many different dishes, but never needs to stand alone. It is redundant to reward someone for the most innings pitched and then also a low ERA. If someone throws 20 more innings, he is feeling positive effects of a buffer to protect the earned runs. The extra IP reward him the flexibility to give up a run or two without the ERA moving. He doesn't need the additional credit for the IP themselves. It is the same argument surrounding the double-counting of dependent statistics with power hitters, namely RBI and HR.  

Total innings pitched, among starter Cy Young candidates, are annually so bunched up to really mean anything. People are quick to ask "what is the big deal with WHIP? What does a difference between 1.10 and 1.15 really say about the pitchers?" That is a fair point; one that I believe is more poignant with IP. For the last two seasons, every National League starting pitcher in the top 10 of the Cy Young vote had over 199 innings pitched. Dickey led that field with 233 2/3 IP, with an asterisk that he featured a knuckleball -- that is less-taxing on the arm -- and a really poor bullpen. So you have 15 top-tier pitchers separated by a handful of total innings over the course of 162 games. It is splitting hairs to give a player the edge over all others for one more start made.      

Fewer Total Strikeouts: In the case of pitchers, a strikeout is most definitely not "just another out." In fact, the ability to create outs for one's self is among the most important skills a pitcher can possess. The godfather of all baseball statistics, Bill James, once stated, "There is simply no such thing as a starting pitcher who has a long career with a low strikeout rate." However, the key word there is rate. Strikeouts are not like 200-hit seasons, where accumulating a massive amount is a standalone feat. Sure, a ton of strikeouts are nice, but when did they occur, how many innings did it take to get there, and how many walks -- the negative flip side of the pitcher accountability coin -- were issued? 

Price and Verlander have been strikeout pitchers for most of their big league careers, and 2012 was no different. Some talking heads were drawn to Verlander's 34 strikeout edge over Price, saying that the distance in the statistical standings was equivalent to "four average games' worth of strikeouts." True, it may have taken David Price four more starts to get to Verlander's final total, but the strikeout rates paint a far better picture of their similar effectiveness. The two aces were separated by a mere 0.3 strikeouts-per-nine (SO/9): Price 8.7, Verlander 9.0. 

The strikeout-to-walk ratios (SO/BB) also help to shrink the 34-strikeout gap. Just as total strikeouts are overblown, there is nothing to be gained from Price's "victory" in walks surrendered -- 59 to Verlander's 60. With some grade school math, the BB/9 pendulum swings the numbers back in Verlander's favor: 2.3 to 2.5. The moral of this section is to remember that innings pitched show up in so many more useful stats than the provide on their own. Stop tallying IP just to do it. And don't go chasing totals of SO or BB; bring all three stats together in the form of practical rates. It levels the playing field, showing Price and Verlander were nearly identical in walking guys to first base and walking them back to the dugout.

Better Bullpen in Tampa BayA starting pitcher can pitch brilliantly, leaving a game in-line for a win, and then helplessly watch from the dugout as the bullpen blows the hold or the save or the "pre-hold" stat that hasn't been formally isolated yet. But a good pen is worth its weight in gold, at -- what should be -- no fault to the starter.

Price did have the better back-end staff, which is partly to "blame" for his average of 6 2/3 IP per start (to 7 1/3 IP for Verlander). In 2012, any Rays starter that got through the eighth inning was a cherry on top. They, more than most in the league, did not need a 120-pitch Herculean effort from their ace. Price could save his bullets to spread throughout more dates on the schedule.

Fernando Rodney had a stellar season -- eighth best Pitcher Rating in Major League Baseball -- closing games for the Rays. He had only two BSV in 50 chances. Yet, one of those two missteps actually cost David Price a win. 

Meanwhile in Detroit, Jose Valverde had a rockier year than his recent past would suggest and he blew five saves. But only once did a blown save come in a Verlander start (Opening Day). So you can add one victory to each Cy Young candidate's speculative win total and/or chalk it up as a wash. The bullpen did not solely account for Verlander reaching only 17 wins.  

Even if you said all of Valverde's blown saves cost Verlander directly -- meaning a potential 22 wins -- I would still give the edge to Price. Postseason awards are not for extrapolation or speculation. It is like comparing pre-tax income and take-home-pay. It doesn't matter what you would have without deductions; it is about what you walk away with. Bullpens strip every pitcher of wins. Price dodged the minefield of a long season with the most to show for at the bottom line. It shows that the modern Cy Young is more like the NFL rushing title; it takes a broader team effort than ever before. 

If anything, fewer innings pitched left Price more susceptible to exiting with a lead and not getting the win. The fact that Price still got to 20 victories, despite this greater vulnerability, proves that Rays' manager Joe Maddon routinely made the correct call. If Verlander pitched the 211 innings that Price did, he could have actually had a worse win-loss record. Ultimately, Verlander has very little excuse for not winning 20 games in a weaker division, with a playoff-bound team. And thankfully, people began to pay attention to this again in 2012. 

Conclusion:

While win total seems to be the theme of the day, the losses are equally important. It is not like Price went 20-10 on the season. His .800 individual winning percentage was far superior to Verlander's .680. Where was Verlander's powerful offense -- led by the MVP -- to spare him a loss here and there? 

The next argument I pose plays more to "the baseball gods" than it does hardline evidence. Baseball has natural ebb and flow; for every time that a starter helplessly watches a bullpen blow a win, there is a occasion of a late-game rally pulling a loss out of the fire. Or so it seems (I have no proof that it did, in fact, even out for both American League Cy Young finalists). Even if Price benefited from more "bail-out" no-decision games than Verlander, it was not enough to entirely account for a 12% edge in win percentage. 

Pitchers makes their own breaks and their individual win percentage is the barometer. Most times the win-loss record gauges a player's success rate of wiggling off the hook. Price did what he needed to do, within the criteria necessary to rack up 20 wins.   

I do not care what external variables are involved; I want the pitcher that has the "it" factor to win a majority of his starts. Call it luck, call it a strong bullpen (one that guarantees a victory after 6 innings pitched), call it a potent offense that "wakes up" when a certain pitcher is on the mound. Call it beating up on weak-division teams and collecting "meaningless games" after being mathematically eliminated. Call it whatever you want. A win is a win is a win. Some pitchers know how to manage the game better than others. 

In a professional sports landscape, with super-economics at play, is it not the ultimate return on investment for a pitcher to win the games they start? Forget the ERA and the WHIP, the SO/9 and the SO/BB ratio. If you are paying a player five million dollars for 34 starts, winning 17 does not cut it. Simple as that. You pitch to win the game. Hello!

This 2012 AL Cy Young vote was encouraging to me. It is one in a series of recent results that show steps in the right direction -- away from those jibberish stats that exist only in a vacuum. Granted, writers have been bailed out by some no-brainer candidates. At least they are acknowledging wins again, even if they will never admit it. Here's to someday actually piercing the veil and getting concrete explanations as to what the BBWAA is looking for. Until then, I cannot wait to see what happens when we get our next 15-game winner vs. 20-game winner.

Pitcher Rating Finale 2012