Andrew McCutchen: more valuable than we may have thought

Since baseball’s most valuable player award was introduced in 1911, writers, fans, players and coaches have debated award winners and what it really means to be “most valuable.” It’s pretty incredible that over 60 years later, baseball people still talk about Joe DiMaggio beating out Ted Williams for the 1941 AL MVP, despite the fact that Williams hit .406.

MLB Network’s show, Prime 9, did an entire episode on the top nine MVP snubs of all-time. Even the merits of last season’s award winners, Ryan Braun and Justin Verlander, weren’t enough to make them consensus selections. The debate over who should or shouldn’t have won the MVP trophy is part of what makes the award great.

In today’s game, the debate rages on with baseball’s various schools of thought arguing over what makes a player the most valuable in his league. There’s the numbers side of the spectrum, with sabermetricians brandishing their laptops and advanced metrics, and then there’s everyone else, who use subjective “reasoning” to decide who’s been the most valuable, in each league.

A quote from a Ken Rosenthal article, which was cited in the first chapter of Baseball Prospectus’ Extra Innnigs, does a pretty good job of revealing where the debate currently stands:

I understand why some sabermetricians freak out over the MVP voting every year, howling for the mainstream media to get a clue.

But you know what?

Those analysts need to get over it.

Ignoramuses in the MSM, including yours truly, continue to make greater use of sabermetrics; heck, we even elected 13-game winner Felix Hernandez the American League Cy Young Award winner in a landslide last season.

The MVP, though, is different. Always will be different. And heaven help us if the voting ever disintegrates into a reflexive regurgitation of the Wins Above Replacement (WAR) rankings.

Sabermetrics is the analysis of baseball through objective data. An MVP vote is subjective by design. Voters are instructed — yes, instructed — to vote any way they darn please.

“There is no clear-cut definition of what Most Valuable means,” the ballot says. “It is up to the individual voter to decide who was the Most Valuable Player in each league to his team.”

There you have it, a virtual invitation to mayhem, or at least, an intense debate involving fans in four countries — the United States, Canada, the Dominican Republic and Venuezuela — when the race is as hotly contested as it is this season in the AL.

Perfect.

It’s not like we’re debating how to fix the economy here. We’re debating an award that has a flexible definition, and anyone who pronounces his or her own definition superior to any other misses the point.

I think I speak for most sabermetricians when I say WAR should not be the end-all, be-all of any MVP debate, but it also should be considered and weighted heavily. There are other factors that, of course, go into the decision of who has been most valuable. Factors such as quality of teammates, quality of opponents, if the player’s team made the playoffs, as well as WAR and other metrics, should be considered, but settling for whoever the voter decides has been most valuable to his team is in many ways ridiculous.

I do not want to propose a new way of selecting the most worthy candidate for the MVP award, but instead I would like to propose a different way of evaluating which player brings the most overall (considering all factors) value to their team.

A Hardball Times Update
Goodbye for now.

For much of this season, I have been attempting to evaluate which players have been most economically valuable. I’ve used both a WAR per dollar analysis, as well as a slightly more accurate cost-profit analysis.

A huge aspect that was missing in both of those analyses was the overall impact a player has on his team. A simple WAR leaderboard does not encompass the typical MVP-question of “where would they be without him?” Knowing how many wins a player has been worth for his team is important, but understanding how much impact he has in comparison to the rest of the team is even more vital to the analysis of which player is the most valuable.

So I decided to add percentage of total team WAR into the analysis. Below is the graph of total WAR for all 30 MLB teams, ranked from left to right by the highest percentage of WAR contributed by a single player.

image

The first thing to note is the distribution, which is slightly skewed to the right, towards teams who don’t receive an above-average percentage of WAR from one player (or superstar). This skew becomes more apparent when the fact that teams well below .500 like the Mariners, Padres and Twins finish towards the top, while teams well above .500 like the Nationals, Yankees and Rangers finish near the bottom.

This makes sense abstractly, as a team like the Mets with only one real superstar would expect to get the large chunk of their contributions from David Wright; while a team like the Yankees or Rangers, who have a bunch of All-Stars, would expect their contributions to be spread around their roster. Despite this distribution, a team that is 14 games over .500 and currently in line for a playoff spot is at the top of this chart.

That team, of course, is the Pittsburgh Pirates, with star center fielder Andrew McCutchen leading the way.

Much has been written about McCutchen and Mike Trout and their runs at the league’s respective MVP trophies. When I did a cost-profit analysis of baseball’s stars, in the post I referred to earlier, Trout and McCutchen finsihed 1-2, in the rankings. Both are young, cost-controlled players who have brought a ton of on-the-field value to their teams this season, but there’s a chance McCutchen may be even more valuable than we originally thought.

Bradley Woodrum, of FanGraphs, wrote in May about how McCutchen was the only valuable hitter for the Pirates’ horrible offense. Woodrum updated the post in July to show that Pittsburgh’s league-worst offense had improved, although he still made the same point.

Andrew McCutchen, for most purposes, IS the Pirate offense.

There’s a chance that, for most purposes, McCutchen is the Pirate team.

image

McCutchen is the only player in baseball who is worth a quarter of his team’s total WAR (24.77 percent).

But his lead does not look very distinct on this chart. Each team has to have a best player or highest contributor; all 30 MLB teams have one player who is worth over 10 percent of their total team WAR. Thus far in 2012, teams have received, on average, 17.09 percent of their WAR from their best player. I modified the last chart by making that percentage the baseline to show just how far above average McCutchen’s contributions have been.

image

This chart gives us a very good idea of just how valuable McCutchen has been to the Pirates with his on-field production. The 25-year-old’s production has been well above what an average team would expect from its star player. What’s even more amazing about what McCutchen has done is that he does it for millions less than the typical superstar.

McCutchen has yet to accrue three years of MLB service time; thus, he is still eligible to make the league minimum salary. When the signing bonus of McCutchen’s latest contract is pro-rated across the deal’s duration, his 2012 salary rises slightly above the league minimum ($708,333).

His new contract is an incredibly team-friendly deal—six years, $51.5 million with a $14.5 million club option—that has him signed through at least his first two seasons of free agency. The Pirates are lucky to have McCutchen locked up at a below-market rate for the coming seasons, but I’m sure they are even happier to have him for less than $1 million this season.

Economics and the business of baseball have no impact (and shouldn’t) on actual voting for the MVP award. But the goal of this post was to propose a different way of evaluating a player’s total impact on hisr team, not a change to the actual selection process. Thus, the final piece of this value analysis is salary, but not salary in a traditional sense.

I decided against using a formula that incorporates the market value of WAR (FanGraphs traditionally uses between $4.5 to $5 million per WAR) to convert WAR into a dollar value that can be compared to salary. Instead, I chose to use the percentage a player’s salary is of total team payroll.

Raw salary figures are very useful when comparing players on separate teams, but like a raw WAR leaderboard, it fails to encompass value on the individual team level. Robinson Cano’s $14 million salary is very different for the New York Yankees than it would be for Oakland Athletics. For an organization like the Yankees, Cano can produce less for that amount of money, while he would be expected to put up incredible numbers for a team like Oakland whose highest-paid player, Yoenis Cespedes, makes less than half of Cano’s salary ($6.5 million).

For example, Felix Hernandez makes up almost a quarter (23.27 percent) of the Mariners’ total payroll. Given the fact that the Mariners expect to get some surplus production from pre-arbitration players, they probably don’t expect Hernandez to be worth a quarter of their production on the field, but their expectations are most likely pretty close to that number. Interestingly enough, King Felix has been worth 21.54 percent of the Mariners’ total WAR, pretty close to what they are paying him.

I used Baseball Prospectus’ Compensation tables to find the percentage of team payroll for each player with the highest WAR on his team. I then subtracted the payroll percentages from the production (WAR) percentages to create a chart that showed the surplus (or negative) value that each star has brought to his team thus far in 2012:

image

Only four of the 30 teams (Cubs, Rangers, Mariners and Cardinals) have received lower production than percentage of overall payroll from their best player. It shouldn’t be a shock that McCutchen and Trout far and away outstrip the competition when their pre-arbitration salaries are included in the equation. McCutchen still leads Trout in this category, as McCutchen’s salary is only 1.36 percent of the Pirates’ team payroll while he is worth almost a quarter of their team’s production.

Many have pointed to McCutchen’s ridiculously high .414 BABIP (batting average on balls in play) and 22.9 percent HR/FB (home run-to- fly ball rate) as reasons to believe his incredible production is not sustainable. They are correct in that assumption; since 2003, zero hitters who qualified for the batting title finished the season with a BABIP above .400. Also, only Giancarlo Stanton finished the 2011 with a HR/FB higher than 22.9 percent, which in all likelihood means McCutchen’s HR/FB is due for regression.

While McCutchen’s other-worldly numbers are due for regression over the course of the last two months of the 2012 season, as well as next season, that should not take away from what he has done. His “unsustainable” success, has been just that; success. The crazy thing about McCutchen’s crazy-high production and WAR is that a key factor, his defense, has detracted from it.

McCutchen’s UZR (Ultimate Zone Rating) currently rates his defense as seven runs below average, essentially costing his WAR over half a win. The debate over how useful small sample defensive metrics, including UZR, are continues to rage on, but for many, the seven run total that UZR claims his glove has cost the Pirates is not very trustworthy. In a fantasy world—one that assumes every major leaguer is a league-average defender—McCutchen would have the highest WAR this season. Also interestingly, UZR rated McCutchen’s glove in center to be above-average last season.

McCutchen is a superstar who has been everything for Pittsburgh this season, especially offensively. But his low-cost contract makes him even more valuable. Having a superstar who makes under a million dollars gives Pittsburgh a ton of payroll flexibility to acquire more talent at the deadline to surround him with to make a playoff push.

And this past week, that’s exactly what they’ve done.

References & Resources
All WAR data comes courtesy of our friends at FanGraphs and updated through Monday July 30.

Baseball Prospectus’ Compensation tables


10 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Glenn DuPaul
11 years ago

That’s a really interesting point you bring up about salary above replacement and makes a lot of sense.. it doesn’t affect things too much. A guy like Carlos Ruiz, mid-level salary on a high spending club, goes from 2.15% to 2%. And Alex Gordon a “higher” salary guy on a low spending team goes from 9.3% to 10.48%.

Also, I’ll definitely look into doing the same type of analysis with the highest paid players in the coming weeks.

Chris
11 years ago

Really interesting write-up thanks.  And yet more fodder for getting rid of UZR.  Defensive metrics are nice and all, but I have consistently seen backwards data from UZR.  Players who are clearly AWFUL in the outfield getting Gold Glove-caliber ratings at UZR (Alfonso Soriano and Carlos Lee to name a few recent examples) and other players who bounce back and forth – one year they’re great, the next they’re awful, back and forth…  UZR – you are my nemesis…

Chad Evely
11 years ago

Excellent article and a very interesting idea. I really enjoyed it.

Hank G.
11 years ago

The definition of “most valuable” player may be open to each person’s interpretation, but using some form of relative value, including the more sophisticated version that you are using, just doesn’t make sense to me.

$1000 is relatively more valuable to me than $1 million is to Bill Gates, but there is no way he would exchange his million for my thousand.

aweb
11 years ago

Great article – I’d love to see the same type of analysis done on each team’s highest paid player – who is contributing the least among the biggest paid players?

One niggling point on team salary percentage. If you are going to compare team WAR % to team salary %, you need to cancel out the baseline minimum salary – that is, you need to figure out salary above replacement. For each player. In theory, a team could pay 25 total rookies $480,000 each as an absolute floor (unless there is a sneaky way around paying even the minimum). That’s 12 million. Hernandez is making ~18 million more than that, and the team is paying out ~67.5 million above the floor, so Hernandez represents ~26.7% of the salary above replacement.

ksw
11 years ago

pretty good research, and fairly good presentation.
you need to define what is ‘replacement level’, not just parrot other vague summaries.
you need to define how many hypothetical runs, theoretically, constitute a ‘win’, how defense, running, and base runner advancing outs, are involved.
seems to be a quick piece, with little understanding of the underlying concepts.

Glenn DuPaul
11 years ago

@aweb I posted a quick live piece with a response to what you had asked for about each team’s highest-paid player.. http://www.hardballtimes.com/main/blog_article/the-value-of-each-teams-highest-paid-player/

@Chris and Chad thank you for the compliments I’m glad you enjoyed it.

@Hank I’m not certain I get what you’re saying. Obviously Bill Gates wouldn’t trade his $1 million for your one thousand, but at the same time, if that thousand gets you to playoffs, while Bill Gates is in the same league and can’t make the playoffs with it, isn’t your one thousand more valuable?

@ksw.. I’m not sure what you’re talking about with me parroting other vague summaries.  I think replacement level and how WAR is calculated by many writers over time, that it’s necessary to stamp out its underlying concepts in each piece.  The readers would just become bored, if the author did that each time they cited WAR.

Here’s reference on replacement level. Also, 10 runs usually constitutes a win, and a team built of entirely replacement level players would win about 43 games (based on FanGraphs WAR), so each WAR is a win added above that number. Here’s a link that explains how FanGraphs calculates their defense, running and base runner advancing, as well as positional adjustments.  McCutchen’s positional adjustment in center offsets his low UZR (defense), in case you were wondering. 

Also I would say my understanding WAR is just slightly above a “little”.

Glenn DuPaul
11 years ago

@Gaudy, 10 runs is not arbitrary at all. Taking the average runs per is how they do it, and almost every year it comes out to 10 runs.  Whatever that total is though is what they use; this season scoring is down so it’s around 8-9 runs per WAR.

Take the Texas Rangers who have scored 533 runs and won 61 games, they have needed 8.7 runs per win.  So if Ian Kinsler has been worth say 70 runs and the calculated replacement AL 2Bsmen is worth 20 runs, than Kinsler WAR would be ~5.7

Hank G.
11 years ago

My point is that players are fungible. If I could make the playoff with my thousand (this is really stretching the analogy), I’d be even happier if I had the million instead. I would make the playoffs and have $999,000 to spend on booze and broads. And yes, I’d probably waste some too.

For the most part, players don’t control who else is on the team. Trying to ascertain value by relative value could lead to some absurd results.

To me, most valuable is the player whose season you would want for your team if you could predict all players’ performance at the start of the season. Of course, that leaves very little to write about, since you could take the league leader in WAR (or your favorite other metric) and have an excellent choice for MVP.

Gaudy
11 years ago

I’m still curious as to how 10 runs came to represent a win. It seems ridiculously arbitrary, and it also seems to me as though it’d be easy to calculate the actual value by taking the average run differential of games or something. I dunno. Im not a scientist. Anyways, i’ve just never seen any proof of this 10 runs = a win rule. arbitrary.