Jose Reyes Talented but Risky Bet For Marlins

Amid the elation of last night as Marlins fans everywhere rejoiced over the news that free agent shortstop Jose Reyes had signed a six-year, $106 million deal with the Marlins, I came to the realization that the Marlins had finally signed a big contract that was going to deserve poignant analysis. Prior to this deal, I have never had to come up with analysis on a deal as monumental to the team's very foundation such as this one.

Since I began blogging in 2009, the Marlins have barely made any waves in free agency; the biggest contract they had previously handed out was John Buck's 2011 deal, and that was for just $18 million for its entirety. Reyes will be making almost that much in one season, and he'll be doing so for five more years after that. This is a turning point in this organization's history, and its analysis should be managed seriously.

So let's get to the $106 million question: Did the Marlins sign Reyes to an appropriate deal?

Keep in mind what that question means. Yes, I am just excited as every other Marlins fan about this bit of news. Yes, this means the team was serious all along, unlike what so many detractors were saying as late as this past Friday with the news of the Heath Bell signing. It's very important for morale reasons, reasons outside of the actual nuts and bolts of the deal.

But when it comes to answering whether the move was a "good" or "bad" one from an analytic standpoint, I'm going to put on my objective glasses as best I can and view it like any other major free agent signing. Let's see if I can manage that.

Jose Reyes: The Past

This is the player the Marlins saw and liked enough to sign to a deal.

Reyes, 2009-2011	PA	AVG	OBP	SLG	wOBA	wRC+	fWAR*
2011	586	.337	.384	.493	.386	149	6.2
2010	603	.282	.321	.428	.329	104	2.9
2009	166	.279	.355	.395	.342	110	0.8
Three-Year	1355	.306	.352	.452	.355	124	9.9

*Denotes FanGraphs Wins Above Replacement

The 2009 and 2010 seasons were clear blemishes, what with the injury history and the fact that Reyes underperformed in his 2010 season. However, consider the following information:

- From 2009 to 2011, only two shortstops (Troy Tulowitzki and Hanley Ramirez) with over 1000 PA had a better wOBA and wRC+, which is park- and league-adjusted measure of wOBA, than Reyes's 124.

- From 2009 to 2011, only six shortstops (the above mentioned plus Derek Jeter, Alexei Ramirez, Yunel Escobar, and Elvis Andrus) had a better fWAR total than Reyes.

From mashing his best and worst seasons together, you got a player who was the best fifth best shortstop in baseball and the third best hitting shortstop in the game. Jose Reyes was good even when he was bad, and the overall .306/.352/.452 package is perfectly acceptable for a capable defender at shortstop. Even though he has missed significant time with injury, he managed to average 3.3 wins per season.

Now, what can we see in his peripherals?

Reyes, 2009-2011	K%	BB%	XB/H	ISO	BABIP
2011	7.0	7.3	0.464	.156	.353
2010	10.4	5.1	0.516	.145	.301
2009	10.8	11.4	0.414	.116	.307
Three-Year	9.1	6.8	0.480	.146	.325

None of these numbers come as any surprise for fans who have followed Reyes in the past. His perippherals were all very similar to his career totals as well, meaning there should be little fear of regression to Reyes's likely mean. Yes, his batting average on balls in play is just a tad higher than his career BABIP, but it should not fall significantly going forward, and I would not be surprised if we see more of the same of these three-year peripherals in 2012.

Each of the three different defensive metrics used by the three Wins Above Replacement (WAR) metrics often quoted here have Reyes worth between seven and nine runs below average defensively at shortstop since 2009. That boils down to just about three runs a season, and that is a conservative enough guess that I would not mind using it for his defensive performance.

Jose Reyes: The Projection

We may not have done a projection here on Fish Stripes before, and projection season for most systems has not yet started, but it is easy enough to make a close-to-accurate projection of the future simply by using a weighted average of the last three seasons a player played. Indeed, the last three years tells a good deal about we can expect of the player's following season, and I see no reason to believe that it would be any different for Reyes.

As we mentioned above, Reyes hit a solid .306/.352/.452 (.355 wOBA) since 2009, and it would not be surprising if that continued into 2012. The funny thing is that FanGraphs's Fan Projections and Bill James's projections also agree, tabbing Reyes for almost identical batting lines and wOBAs. Without looking at other sources, this offensive projection seems to make plenty of sense. Let's use the average of these three numbers and project a .355 wOBA for Reyes next season.

We mentioned before that wOBA can be easily converted to runs above average by doing some simple math against the league average wOBA. I took a weighted three-year average of the league's run environment and estimated that next season would look similar to 2010, when the league average wOBA was .321; if it is more like the run environment in 2011, the league average then would be around .316. Nevertheless, even without considering a park adjustment for the new stadium, Reyes would be projected to produce between 17 and 19 runs above average per 600 PA for the Marlins in 2012, depending on which run environment you use. Tack on an extra two runs from bonus value on the bases (not including stolen bases, which are included in wOBA), and you have a player who might project as just around 20 runs above average on offense.

We mentioned defensively that Reyes has average two to three runs worse than average on defense, and I am fairly comfortable going with that value for 2012 as well. Of course, being a shortstop has inherent value, as there are a select few players capable of playing the position in the major leagues. As a result, we add an adjustment for Reyes's position to give him credit for being a shortstop; the position adjustment for shortstops in most WAR metrics is around an added seven to eight runs per 600 PA. Therefore, the "defensive value" of a shortstop like Reyes adds up to around four to five runs above average in 600 PA.

We are almost there in building our Wins Above Replacement projection for Reyes in 2012. We have that he is a player worth 25 runs above average most likely, and to that we simply add a replacement level adjustment to give him value over a Triple-A scrub player who would theoretically replace him in a lineup. The adjustment is an added 20 runs per 600 PA, putting Reyes at 45 runs above replacement. Usually, we would then convert this to Wins Above Replacement using the old shorthand of "10 runs to one win" conversion, but the lower run environment means that it takes fewer runs to make a win these days. As a result, we would project Reyes to be worth around 4.7 Wins Above Replacement, or WAR, in 600 PA in 2012.

Jose Reyes: The Injury Risk

But of course, we need to take into account one thing we are not sure of: will Reyes make it to 600 PA? He has been around that mark for each of the last two seasons, so there is something to that estimate, but what if injuries take away more than the 30 or so games that he missed in each of 2010 and 2011? What might we then see from him?

Rather than try and predict what will happen with Reyes in terms of injury, I'd rather show you a spectrum of what to expect at various different PA levels. In the last two seasons, he has missed 65 possible games to injury. In the 258 games he did play, he averaged 4.6 PA per game, meaning that those 65 games cost Reyes 299 PA over the last two seasons. Let's use those figures and provide a playing time spectrum for his performance.

Games Played	PA	WAR
162	745	5.8
150	690	5.4
140	644	5.0
130	598	4.7
120	552	4.3
100	460	3.6
80	368	2.9
60	276	2.2

Rather than attempt to decide which one is the "correct" projection, you can choose yourself which of these numbers looks most appropriate. if you think Reyes will miss around 30 games again next season, you'd actually expect him to hit the 4.7-WAR mark we projected above. If you think he will miss more time, he is still projected to be better than a full-time average player all the way down to 60 games played. Having said that, it would not make me feel good if he only played 60 games, but it does comfort me to know that Reyes is that good. Oh, and if he plays out a full season, you may see a total close to the WAR he put up this season in only 126 games played. So Fish Stripes readers, you decide how much time Reyes will miss to injury.

Jose Reyes: The Slightly-More-Distant Future

Of course, the Marlins had to figure that they were getting a good player in 2012. The question will be whether the team will get a good one following this season and for the five years after that, and that question begins to get a bit murky. In order to look at how shortstops generally age, I looked into all shortstops who played between 1961 and 2005 and examined their Baseball-Reference WAR (rWAR) totals from their ages 23 to 28 seasons. The timeline for this was in order to exclude any players who have yet to have the opportunity to play six seasons after their age 28 year; excluding these players included taking out names such as Rafael Furcal (he finished his age 27 season in 2005), Yunel Escobar, Jhonny Peralta, Hanley Ramirez, Jose Reyes, and Jimmy Rollins from the sample. I cut off the sample to include only shortstops with greater than 15 rWAR in that time span.

The list can be read here. There were 21 shortstops on the list. The best player on the list was Alex Rodriguez, who amassed 45.8 (!!) rWAR through those six years. The worst was Dave Concepcion, who totaled 16.1 rWAR. The last guy to sneak into the sample was former Marlin Edger Renteria, who finished those seasons in 2005. As a group, that set of 21 players averaged 24.9 rWAR during that time span. They also averaged 25.4 rWAR in 3556 PA, which is how many PA Reyes accumulated in his ages 23 to 28 seasons. This is significant because Reyes totaled 25.3 rWAR in that same time span; this means that this population of shortstops is actually pretty representative of Reyes's skill.

How did they fare in their next seasons? Not nearly as well. All 21 shorstops did see play in the following years, but three players dropped out of sample earlier than age 34, their sixth season. The best player among this group was once again Rodriguez with 35.0 rWAR, while the worst was Jim Fregosi at a paltry 1.9 rWAR. As a group, they averaged 16.4 rWAR over that time span. They also averaged 3007 PA in that time span after averaging 3493 PA in their ages 23 to 28 seasons. That decline is quite significant, as it leads to a decline of 8.5 wins or 1.4 wins per season between those two time spans. Whereas players in their younger years averaged 4.1 rWAR per season, their older years had them averaging 2.7 rWAR per year.

How would that decline match up with Reyes? Well, he totaled 25.3 rWAR in his younger years, and if he suffered the same decline, he would end up averaging 2.8 rWAR per season over the life of the deal. If we start him at 4.4 rWAR (Baseball-Reference WAR has a slightly lower replacement level adjustment) for 2012 as per our projections, here is what his decline would look like as a linear drop in talent over the next six seasons.

Year	rWAR	Change
2012	4.4	--
2013	3.8	-0.6
2014	3.2	-0.6
2015	2.5	-0.7
2016	1.8	-0.7
2017	1.1	-0.7

This is approximately the sort of aging curve we may expect to see from Reyes. A combination of aging and injury may drop his eventual usefulness down to that of a part-time player by the end of the contract, so the Marlins are clearly at risk of committing a lot of money to a very bad player in 2017. But the bright side is that up until 2016, we would expect Reyes to still be an average player at age 33, which would not be surprising or all that disappointing I suppose.

Of course, Reyes was on the higher end of the scale of ballplayers we compared him to in the above links. It is perfectly possible that he ages better than the guys below him did. But even if he doesn't, Dave Cameron of FanGraphs believes that, due to salary inflation and the subsequent rise in what players should earn per WAR contributed, this may not be a huge problem for the team.

Here’s essentially what Reyes would need to produce to justify this contract as fair market price, based on $5 million per win and 5% annual inflation over the life of the contract:

Year	WAR	$/WAR	Value
2012	3.53	$5.00	$17.67
2013	3.37	$5.25	$17.67
2014	3.20	$5.51	$17.67
2015	3.05	$5.79	$17.67
2016	2.91	$6.08	$17.67
2017	2.77	$6.38	$17.67

At that price, the Marlins are essentially paying for a total of +19 WAR over the next six years. What does +19 WAR from a shortstop look like over a six year period?

If you bump up the average that I projected using historical references to adjust for the difference between rWAR and fWAR, you get that Reyes would total 18.1 fWAR or 3.1 fWAR per season based on those decline numbers. That would fall just short of Cameron's projection though, and that does not take into account the fact that he is using a $5 million per year inflation value that has yet to be established this season. It could very well be that high, or it could hold at $4.5 million as it supposedly did last season.

For fairness' sake, I'll calculate the same thing using $4.5 million per year and a five percent inflation rate.

Year	Salary ($M)	$/WAR ($M)	WAR
2012	17.67	4.5	3.9
2013	17.67	4.7	3.7
2014	17.67	4.9	3.6
2015	17.67	5.1	3.5
2016	17.67	5.4	3.3
2017	17.67	5.6	3.1

Compare this table to the above table with Reyes's projections. It turns out that the Marlins would only be getting their money's worth in years one through three of the deal, with the remaining steadily declining. This above configuration requires that the Marlins get 21 fWAR from Reyes, where we projected that he would get 18, meaning that as of right now, the Fish probably overpaid by something like $15 million. Ironically, that puts the club right around their initial six-year, $90 million offer.

But this is not surprising. While the Marlins likely jumped the gun a bit to try and ensure a signing, it is not all that concerning to me that they went over by about $2.5 million a year. Unlike in the case of Heath Bell, one would expect the Fish to have to pay some premium to acquire star-level talent, and this is essentially what happened. Plus, the improvement the Marlins are getting in this case is significantly greater than the improvement at a position like closer, so I am more willing to forgive the overpay. I'm impressed that the Marlins nailed the initial offer so well, and I am disappointed that the team did not wait out Reyes given that other teams seemed reluctant to go this far with him. But for the talent we are getting, it is difficult to complain at the moment, though it should get worse as the years pass.

So the final verdict as of today on the deal is that it is an overpay by $15 million based on projections of how shortstops age and where Reyes will be in 2012. Right now, Marlins fans may not care about this, but they will be upset when their new favorite player declines significantly for what the team is paying. There are real concerns about how well Reyes will age, but for now the team's fans may be content to simply worry about that later. I am happy the move was made, disappointed but not surprised by the price it cost. It is not a gross overage on a long-term deal, and there have been a lot worse, but it remains an overpay.