Yoenis Cespedes will probably sign for more than he’s worth

After a month of watching the playoffs and considering different moves the team could make, the off-season is here and things are happening. One of the things that happpened was Yoenis Cespedes opting out of his contract with the Mets. Last year the Cuban slugger took home a $75 million deal over three years, due to the fact that it contained an opt-out after the first year.

After posting another strong year, Cespedes will opt out of the two years and $47.5M remaining on his deal with the Mets and enter a free agent market much weaker than last year. In that market, seven players received $100M+ contracts. Two of them were outfielders: Jason Heyward and Justin Upton. According to leading transactions site MLBtraderumors.com, Cespedes may be the only player to receive as much as $100M this time around.

Let’s start how my articles typically start, with a look at the player’s recent production:

Cespedes’ WAR has declined quite a bit from last year, but it’s almost entirely a gigantic swing in defensive value. He split his time pretty equally between center (495 ¹⁄₃ innnings) and left (550 ²⁄₃ innings) in 2016. For his career he is a much better defender in left (13.3/UZR 150) than center (-19.3/UZR), to the point that he’s been more valuable in left despite a ten run difference in position adjustment between the two positions.

There’s also the fact that Cespedes suffered a quad injury in 2016. It’s not hard at all to see how such a thing could affect his defense. It could have also affected his base-running, which was below-average for the first time of his career.

The most interesting thing to me was the increased walk rate. Cespedes’ walk rate was at its lowest in 2015, and at it’s highest in 2016. The change allowed him to post a similar wOBA despite a 25 point drop in BABIP. Let’s look at his plate discipline stats from fangraphs.com:

His O-Swing% is down, as is his Zone%. So he’s seeing less pitches in the zone, and he’s laying off more often when it’s not in the zone. Those differences seem rather small, but of the 90 players that had a qualified amount of PA in 2015 and 2016, he has the 14th biggest drop in O-swing% and the 34th biggest drop in Zone%. That led to the 4th biggest gain in BB% and 35th biggest drop in K%.

A couple weeks back, I looked at the improvement Kolten Wong made in terms of walks and strikeouts. I found that Wong had the 17th best improvement in non-contact wOBA (calculating using only unintentional walks, strikeouts, and hit by pitches) of 210 players with 300 or more PA in both seasons.In the graphic I posted in that article, you’ll see that Cespedes placed ninth.

Going forward, the projections essentially expect his walk rate to be an average of the two previous years, with a slightly increased K%. That about matches what I found when looking at other players who made similarly large gains when looking at Kolten Wong’s numbers.

Those gains in 2016 completely made up for a drop in BABIP, which brings us to his contact quality. In case you haven’t read an article of mine where I dissect a player’s contact quality, here’s an article where I go over the various concepts. Essentially, I use a player’s Statcast data presented at BaseballSavant.com. Specifically, we’re looking at the Exit Velocity and Launch Angle of each of a player’s batted balls. Here’s how Cespedes grades out:

Here we see that my system has liked the quality of his balls in play a lot more than the results. However, my system does not take into consideration a player’s speed (and thus his ability to leg out infield hits) or how easy it is to shift against a player. According to Fangraphs’ speed score, Cespedes has graded out as an above-average runner every year except 2016, when he was likely dealing with his quad injury.

Cespedes went to the opposite field 24% of the time, the 57th least out of the 146 qualified players in 2016. Looking at his spray chart from brooksbaseball.net, he definitely pulls most of his grounders, but the opposite field isn’t exactly barren. Again going to Fangraphs, he was shifted against in 38% of plate appearances, compared to a league average of 16%.

It’s tempting to say Cespedes might be a better runner when he presumably enters 2017 healthy, but he’s also entering his age 31 season. So you wouldn’t want to expect something all that similar to when he was 29. Between the heavy shifting and probably being about an average runner going forward, it’s easy to discount the fact that my system likes his BABIP abilities more than the results.

Let’s look at his contact quality a bit closer, using graphics supplied by BaseballSavant.com:

The exit velocity formatting looks a little goofy and I’m not sure why, but only at extreme angles that we’re not spending any time dissecting. Cespedes has peaks in batted balls at 15 degrees and 0 degrees, not exactly prime angles for home runs. Going by angle, home runs peak at 27 and 28 degrees, and the large majority of homers occur between 18 and 44 degrees, inclusive.

You can actually see some sense of positive here: Cespedes put up strong power numbers while still not really maximizing his power. He hits very few balls at the most optimal angles for homers. By comparison, check out the same graphic for Jedd Gyorko in this article. Jedd’s second-biggest spike comes right around the 27/28 degrees mark that is best for homers. Cespedes might be just a slight adjustment away from a monster home run spree.

Looking at things granularity, and adjusting for speed and shifts, it seems like Cespedes’ results on contact were about what was expected given the contact quality. However, the projections expect quite a bit of regression. While Cespedes posted a 135 wRC+ in 2015 followed by a 134 wRC+ in 2016, the projections expect a 116 wRC+ in 2017. They see him retaining half of his walk rate gains from 2016, raising his strikeouts rate, and losing 30 points of ISO while carrying a similar BABIP to 2016.

Like we’ve done with similar Cardinals acquisition targets, we’ll look at his projected value. Using his current projection, pro-rating it to 600 PA, applying an average curve, and taking $8M as the cost of a win with 5% inflation going forward, here’s how Cespedes’ value grades out:

I would take the over on Cespedes’ projection. A half a win is a lot in terms of value, and that gets him up to $95.5M over 5 years. If you take his true talent level as his 2016 hitting line, and regress his defense and base running to average, that makes him worth about 4.2 WAR player. Apply the average decline from age 30 to 31, and he’d be worth about 3.5 WAR in 2017. That gets him to $108.3M over 5 years

That doesn’t count the fact that Cespedes did receive a qualifying offer, meaning to sign Cespedes the Cardinals would have to give up a draft pick valued at $10M, so you can adjust that down to a little less than $100M. Still, that most optimistic projection might not bring in the free agent with the most earning power.

Notice though that I didn’t say the best available player, but the most earning power. Despite the fact that Justin Turner has a more impressive immediate track record and a much better projection going forward, he’s not expected to earn as much Cespedes. That’s probably most attributable to the fact that Cespedes is most known for his home run power, which tends to get paid best in free agency. Let’s look at some contract comps for Cespedes:

From 2013 to 2015, Chris Davis was worth 13.4 WAR and scored a $161M deal before the 2016 season. Cespedes has been worth 13.2 WAR in his last three years, though with 38 less homers. Justin Upton’s last three seasons came in at just 10.4 WAR and 82 homers, 6 less than Cespedes over that time frame. His younger age helped, as he managed to get a $132.75M guarantee and an opt-out after the second year of the deal.

Curtis Granderson is the oddball among this group. He only took 245 PA in an injury shortened walk year, and when he did play he only amassed a 98 wRC+. The year before he had only a 116 wRC+ to go along with below-average defense. Still, he pulled in $60M despite compensation attached mostly due to the fact that he had back-to-back 40 homer seasons in the recent past.

The best comp for Cespedes might be himself. The fact that he was able to obtain $75M and an opt-out in a tougher market last year means he should beat that figure this year. In terms of WAR his 2015 may look much better, but regress the defense in both years back to his career marks and he’s been about the same player both years. Meanwhile, his 2012 where he posted a 102 wRC+ is farther away in rear-view mirror. In the article linked above, MLBtraderumors.com pegged Cespedes for $125M over five years. By comparison, Turner is projected also for 5 years, but just $85M.

A couple of weeks back, I looked at Dexter Fowler's value as a free agent. I came to the conclusion that he too would likely end up overpaid. However, Fowler projects just slightly worse at this point, and will likely go for several tens of millions of dollars less. Fowler looks like a better value relative to Cespedes.

I’m not sure why the projections are so low on Cespedes compared to his recent production. However, even taking the most optimistic of projections leads to a valuation that is on the low side of what he’s expected to get. The market just overvalues home run hitters right now, and he and his agent will look to take advantage of that. Look for the Cardinals to do their due diligence, but also to avoid overpaying for dingers, and spend their money elsewhere this off-season.