This is a post about Lorenzo Cain, or more accurately, players like Lorenzo Cain. But, before that, I want to tell two short stories. They’ll be very short, I promise.
Story 1: Once upon a time, there was a really, really fast center fielder. He was really, really fast. No one knew if he would be able to hit. No one knew if he’d able to be a great center fielder, even though he was really fast. For a while, he didn’t hit, and he wasn’t a great center fielder, and it wasn’t okay. But then, he got better at hitting, and it was okay. He also got much better at routes and getting a jump on the ball, and it was okay. No, it was more than okay: it was great! He got really good, so good that even when the hitting didn’t go as planned, it was still great. One day, he was traded from a bad team to a good team. It took him a while to get his bearings, but after he figured it out, he became a monster. Not the bad kind, that scared his teammates, but the kind that terrorized his opponents. It was his best year ever. Luckily for him, that was also the year that his servitude ended, and he was free to get paid for his services whatever any team wanted to pay him. You’d think he’d get paid a king’s ransom, but then maybe some things happened with his agent and while he got a payday, it wasn’t a very big one. Unfortunately for everyone, that was just the first bad pebble in an avalanche of calamity. His hitting went back to being not-so-good. His fielding went back to being not-so-good. He used to be a monster, and now he wasn’t. Things only got worse; he got traded back to the team where he was a monster, but nothing happened. He didn’t even make it back to the majors last year.
Story 2: Once upon a time, there was a really good player. He was pretty much always really good — if there was a knock on him, it was that he tended to get injured and stay out of the action. He was really good when he first started playing, but then he got injured. He came back, and was really good again. And again. And then he got even better. He could hit. He could run. He could field. Eventually, he became a monster, and for a brief period, was one of the best players around. He then suffered a setback, but bounced back, because that’s what he does. As was said, he’s really good. Then, his period of servitude ended, and he went to the market to get a payday. But it was quiet. Tumbleweeds rolled. His former teammates were also there, but no one seemed to be too interested in paying them, either. A cold wind blew. He didn’t know why this was happening, but he had heard the first story, the one about the really, really fast center fielder before. It wasn’t a comforting thought. He shivered.
It’s now late January, and Lorenzo Cain is still unsigned. Lorenzo Cain, he of the 4.6 career fWAR/600. He of the well-above average Steamer projection. He of the top-40-since-2010 and top-25-since-2012 fWAR among position players. It’s not like Cain is coming off a down year that’s driving suitors away, either: he had 4.1 fWAR last year and played in 155 games. If there was a reason to be concerned about his 2016, where he put up his first below-average wRC+ (99) since 2013, his most recent season should have allayed those concerns. But, perhaps it hasn’t. It’s late January, and Lorenzo Cain is still unsigned.
At Fangraphs, Travis Sawchik’s headline says it all: someone should sign Lorenzo Cain. Indeed. (You should read it, if you haven’t already). Sawchik, unlike me, doesn’t spend forever getting to the point. He writes:
Part of the fear with Cain is how a speed-dependent player will age. The Michael Bourn, Carl Crawford, and Jacoby Ellsbury signings did not happen too long ago.
Sawchik is right — it’s a real concern. But just how real of a concern? Real enough that Dave Cameron pretty much acknowledged all of the above, and more, despite naming Cain his #2 “free agent bargain” of this offseason.
Like his former teammate a few spots up, the case against Cain is pretty easy to make: speed-based player who turns 32 in April, battled hamstring problems in 2016, and has played 150-plus games just once in his career. His 2017 season looks almost exactly like Jacoby Ellsbury’s 2014 campaign, which was his last as an above-average MLB player. Guys who rack up baserunning and fielding value on the wrong side of 30 are risky bets. There’s no question Cain is one lower-half injury away from losing his status as an impact player.
So, that question about just how much prospective teams should be concerned about Cain’s speed-and-defense profile and what it means for his future is what I tried to find out. Cain is 31, and his defense, by UZR, has eroded pretty steadily between 2013 and 2016, until a greater dip in 2017 (24 -> 19 -> 15 -> 14 -> 2 in UZR/150, and 30 -> 26 -> 21 -> 18 -> 5 by DRS/150).
I wanted to build a set of comparable players to Cain, but I didn’t want to be overly restrictive. Here’s what I settled on:
- A player that could play center field, and generally did so, especially early in his career. (Defined as at least 20 percent of his PAs
- An above-average player who accumulated fWAR/600 above 3.0 through his age 31 season, to mirror Cain.
- A player whose wRC+ in that period was not above 130, so that I was getting above-average players who weren’t accumulating essentially all of their value with the bat.
- A player who played at least some portion of his pre-age 32 career in the UZR/DRS era (2002 and on), to allow quick troubleshooting of earlier defensive data, if necessary.
This gave me a set of 35 players. Of these, I had to toss out 15, for the very simple reason that they weren’t yet 31 years old. I also had to toss out Lorenzo Cain (duh). That left 19, ranging from the elite (Carlos Beltran) to the who?.gif (Jason Michaels). Of these 19, five were in a bit of an awkward place: they were older than 31, but not much older — you’ll see why that matters in a minute.
(By the way, before I go on, it’s worth noting the other 15 players that won’t be touched on any more here: Kevin Kiermaier, Mookie Betts, Craig Gentry, AJ Pollock, Christian Yelich, Carlos Gomez, Odubel Herrera, Juan Lagares, Adam Eaton, Denard Span, Ender Inciarte, Marcell Ozuna, Randal Grichuk, and Jackie Bradley Jr. Not all of these are really speed-and-defense center field types, and some may be growing into their bats in a way that may kick them off this list eventually. But, when thinking about how these players age, you may want to think back to this analysis, or not.)
Because Lorenzo Cain is eligible for free agency, a key concern is not just how much it’ll take to sign him, but for how long his services will be contracted. As he’s already 31, and as players seem to start heavily declining in effectiveness around their mid-30s, it makes sense that a very long deal would be out of the running for essentially all bidders for Cain’s services.
In that same piece on his free agent bargains, Cameron included the following set of estimates for Cain’s 2018 offseason free agent contract:
Four years seems to have been some kind of conception of an industry standard. Tim Dierkes and the great gang at MLB Trade Rumors also predicted a four-year, $70 million pact for Cain, very consistent with the estimates above.
Four years would take Cain through his age-35 season. Given that that was the standard, the table below summarizes what I did. You’ll see it’s fairly self-explanatory.
Essentially, the table compares the Cain-types’ WAR, hitting outcomes, and fielding outcomes from their debut through age 31, and then from their age 32 through age 35 seasons. Out of interest, I also looked to see what these players produced at age 36 (i.e., “Should you give Lorenzo Cain a fifth year?”), and the age at which they first collapsed in their 30s, defined as their first season of 0.5 fWAR/600 or below.
There are a lot of takeaways here, so I’ll try to hit them quickly.
The Cain-comparables fare worse in their age 32-35 seasons than basic rules of thumb suggest they would
This group loses between 1.7 and 1.9 fWAR/600 when comparing the two age horizons. (The “master average” incorporates the group of five players that have not yet finished their age-35 season.) If we use the basic rule of thumb that a player loses 0.5 fWAR/season annually between 30 and 35, a hypothetical 4-win player would go 3.5 -> 3.0 -> 2.5 -> 2, an average of 2.75, or about 1.25 wins fewer than his prior performance. If we say that by age 35, the drop is one full win, we’re at 3.5 -> 3.0 -> 2.5 -> 1.5, we’re at an average of 2.625, or about 1.4 wins fewer than his prior performance. The drop is steeper than either of these, at 1.7 to 1.9. To put it in absolute terms: if you use Cain’s 4.6 fWAR/600 as a starting point, he’d end up at 2.7 fWAR/600 to 2.9 fWAR/600 over the length of a four-year contract. If you map the usual 0.5 fWAR annual decline on top it, that’s really a four-year production scale of 3.6 -> 3.1 -> 2.6 -> 2.1. It’s not something to result in the creation of a roster hole, necessarily...
...until you consider that Steamer and ZiPS both have Cain at 3.2 fWAR/600 next year. Using that as a starting point and mapping the usual decline, you’re now at 3.2 -> 2.7 -> 2.2 -> 1.7 over a four-year period. Given that that averages to about 2.5, and 2.5 adding back the 1.8 leads to a “through age 31 performance level” of 4.3 fWAR, this doesn’t seem unreasonable. But it does seem a bit scary. This gets even scarier if you’re of the mind that the Braves aren’t in a position to compete in 2018, anyway — if that’s the case, there’s really no point to signing Cain if he’s only really going to give you one year of above-average production before you’ll be looking to potentially upgrade his roster spot.
Most of the decline in the Cain-comparables in their 30s is due to defense
This makes sense. The wRC+ drop is only about six or seven points. The Def/600 drop, meanwhile, is somewhere in the order of 11 to 12 runs, or slightly more than one win. Given that the overall drop is under two wins, the bulk comes from defense, whether that’s actual defensive value retreating, or value being forfeited as teams move their aging players down the defensive spectrum into the corner outfield (or elsewhere). Also, given that going from a 106 wRC+ to a 100 wRC+ is the loss of about five runs as far as batting goes, the remaining drop in WAR is likely due to baserunning, which matters for the Cain-comparables more than for others.
Definitely do not go to a fifth year with Cain, or anyone similar to him
Of the 15 players that have aged past 35 at this point, only two gave average or better production: Mike Cameron and Curtis Granderson. Another two, Carlos Beltran and Torii Hunter, were average-ish. (Hunter, in particular, is a bit of an outlier: the only player in this group to improve across the board in his early 30s.) The remaining 11 were either below replacement level, or not in MLB by the time their age-36 season rolled around.
It’s likely that no one really needed an analysis to tell them this, but stay away from the Cain-comparables by the time they get to age 36.
However unlikely, a two-year deal for Cain might be ideal
If you look at the rightmost column, you get a sense of when the collapse strikes for the Cain-comparables. Two of the 19 players collapsed prior to, or during, their age-31 season. Another five did so in their age-32 season or age-33 season. Factoring in another three players whose collapse came in their age 34-season, and you’ve attained more than half of the sample... and this doesn’t even take into account that Jarrod Dyson, Jacoby Ellsbury, and Brett Gardner have not yet quite collapsed yet, though they may in the immediate future; Ellsbury is already below-average.
A two-year deal with a team option would be a severe discount for Cain relative to preseason expectations for the price of his services. But, if it could be achieved, it would mitigate a fair bit of risk. Not only would it mitigate against the decline risk associated with being forced to pay a below-average player during a theoretical contention window, but it would also hedge against the collapse risk better than a deal guaranteeing him money through his age 34 season.
What I’d pay for Cain
IWAG projects Cain for a point estimate of 4.3 WAR/600 in 2018, which is way more aggressive on him than Steamer or ZiPS. Factoring injury risk into account, it estimates that he will finish the season with a point estimate of 4.0 WAR.
To kludge together that estimate and the fact that his four-season WAR pace should reflect an average of 2.8 (i.e., his 4.6 fWAR/600 pace through age 31, less the 1.8ish differential from the table), that could suggest a four-season pace of something like 4.0 -> 3.2 -> 2.4 -> 1.6. That suggests the fourth year is not a great investment from a roster construction perspective. The first year, too, is of questionable value, though signing Cain does bring the Braves much closer to having his production actually matter next year.
To that end, I wonder if Cain would agree to something like $46 million over two years, with a team or mutual option for $22 million. This hedges the risk relatively well to the Braves: the upside is a lot of production assuming the decline is graceful, with relatively less risk exposure in terms of paying an average or collapsed player. The downside is that the budget the Braves have to play with gets dented by a sizable chunk in the 2018-2019 offseason, but again, assuming a graceful decline, Cain may still be an above-average player at that point, precluding the need to pay for a different solution in the outfield at that point in time.
Of course, this would probably only be a starting point, with the AAV (average annual value) beefed up to try to compel a signing. (For the record, the various Fangraphs crowdsourcing and MLBTR projections had Cain earning around $17 or $18 million annually; this would grant him around $23 million per season.) I have no idea how the evolving free agent market has really shifted the terms around the going price for Cain’s services. But, even knowing what we now know about how the Cain-comparables have aged and performed in their early-to-mid-30s, signing Cain could work out. The Braves just need to be careful not to commit for too long.