Monday, 23 December 2013

Trading for David Price: Fun With Surplus Value

One of the hottest commodities on the potential trade market these days is the Tampa Bay Rays’ ace lefty David Price, and fans of many a franchise are understandably salivating at the prospect of seeing him don their team’s uniform. It’s not every season that a player of Price’s caliber becomes available while still in his salary arbitration years, which Price will be for the next two seasons. Trade speculation is obviously a favourite off-season pastime of fans everywhere, so there has been rampant speculation in all corners of the media and internet about where Price might end up and what it might take to get him there.

With the winter meetings now concluded without having provided much excitement for Jays fans, I wanted to do a bit of a pivot off some of the analysis I’ve seen recently that is either directly related to the Blue Jays’ quest for starting pitching, or at least very closely related to it. Much of the analysis regarding trades these days has to do with the notion of surplus value, which is calculated by finding a player’s value in $/WAR terms and subtracting their salary ( WAR * $/WAR – Salary ). The difference between the player’s on-field value and their salary is their surplus value. This is significant for obvious reasons. The less you pay for wins from one player, the more money you have to allocate to buying wins from other players. Trades can be analyzed by comparing the projected surplus values of the pieces involved to see which team ended up “winning” the deal from a value perspective. As mentioned, surplus value is at the heart of most transaction analysis, and for good reason.

It’s important to consider surplus value when considering trading close to the majors prospects and their six affordable years of team control for established major leaguers, even very good ones like David Price. For example, in a recent FanGraphs piece, Dave Cameron shows how six years of a top prospect like the Pirates’ Gregory Polanco, projected to provide six years of league average-ish production starting in 2014, provides about $120 million in surplus value against just $30 million in surplus value from Price over his final two arbitration-eligible years for a difference of about $90 million. You can quibble with any of the projections and $/WAR figures, but no matter how you slice it, a prospect like Polanco is likely to contribute far more surplus value over the remainder of his team-controlled years than Price is. Cameron concludes that even in a one-for-one swap, a prospect like Polanco who is ready to step in and contribute for league minimum salary is an overpay for David Price.

That's certainly one way of looking at a trade, but the market doesn't always behave in such a coldly logical manner. In a vacuum, a transaction analysis based strictly on projected surplus value makes perfect sense, but as we know, teams don't operate in a vacuum. Teams have a tendency to overpay in long-term surplus value for players they believe can help them in the short-term, usually when they consider themselves right on the verge of playoff contention. Intuitively, we know this to be true, even without the numbers to back it up. This is where I feel that a lot of transaction analysis based on a comparison of surplus value falls short, in that it compares players as if their values exist in a vacuum. It's a great analytical method, but doesn't necessarily paint the full picture. A team's perception of its own situation, its needs, and the supply of players to fill those needs in any given year all affect a player's value on the trade market in ways that aren't necessarily well accounted for by a strict surplus value analysis.

Michael Valancius over at D Rays Bay examined this reality by looking at 53 trades for starting pitchers between November 2011 and November 2013 in an attempt to get a handle on what the trade market for starting pitching tends to look like from a surplus value perspective. His findings are telling, and confirm what we already know. The results for the team acquiring the starting pitcher are varied, but skew heavily towards teams overpaying in surplus value for starting pitching. The team that did the best was the Cubs with Travis Wood, gaining $24.83 million in surplus value. At the other end of the spectrum was (*drum roll*) the Toronto Blue Jays! According to Valancius' work, the Jays gave up $75 million in surplus value in the Marlins trade that netted them Mark Buehrle and Josh Johnson. The R.A. Dickey trade was also fourth-worst in surplus value terms, in case you're keeping score. Overall, the average amount of surplus value a team gave up to acquire a starting pitcher was $11.39 million, and the examples of overpays are both more frequent and more extreme than the cases in which the team acquiring the pitcher actually gained surplus value.

Having established himself as a front of the rotation starter that can be projected to provide about 9 wins and $30-35 million in surplus value over the remainder of his team-controlled years, Price fits in pretty well with a lot of the names at the overpay end of the spectrum. By looking at some comparable trades in which teams with aspirations of contention acquire a front of the rotation starter, Valancius concludes that teams are willing to overpay in long-term surplus value for short-term wins from the starting rotation if they believe it will get them over the hump. This makes sense, as those marginal wins at the upper end of the win curve are both more valuable in terms of revenue and often more expensive to acquire. 

Based on his analysis, Valancius suggests that the Rays should expect to acquire $75-100 million of surplus value in a trade for two years of David Price. Getting up to $100 million in surplus value sounds a little bit optimistic to me, even in a market that just saw the Royals send the Rays $85 million of surplus value for two years of James Shields who, while a very good pitcher in his own right, is not David Price. Something towards the lower end of that range doesn't sound outrageous though.

So what does any of this have to do with the Blue Jays? Basically, with the Jays frequently being cited as one of the teams interested in Price, all of this got me thinking about what a trade package for David Price might look like if we assume the Blue Jays were to ship the Rays about $75 million or more in surplus value. I'm not saying this will happen or that these would be the exact names involved, and I'm especially not saying the the Jays SHOULD send such a package to the Rays for Price. It's just a fun exercise to see what types of prospects might be required to get a deal done. 

I'll be using the prospect surplus value figures found at Pirates Prospects, which are an updated version of Victor Wang's work from this Hardball Times article. Additionally, since Pirate Prospects only looked at the values of top 100 prospects, I'll be trying to use a quick and dirty estimate of what B and C grade prospects (as per John Sickels at Minor League Ball) might be worth given the inflated cost of a win since Wang's work in 2009. You can see a nice handy summary of Wang's work in this Behind the Boxscore post. For example, according to Wang's findings in 2009 a grade B pitcher could be expected to provide about $7.3 million in surplus value, or about a win and a half in 2009 terms ($4.88 million/win). If we figure a win in 2013 is worth something more like $7 million, then it's worth about 40% more now than in 2009, so I'll be bumping up Wang's calculations for B and C grade prospects by 40% to get an estimate of that value in today's terms. Similarly, the Pirates Prospect piece calculated their prospect surplus values at $5 million/win, which I think is low. I'll be bumping those up by 40% as well to reflect what I consider to be values reflective of today's market at more like $7 million/win. Again, this isn't meant to stand up to any rigorous analytical standards, just to get an idea of what a package for Price might have to look like based on a bit more data and analysis than the often half-baked trade proposals fans love to toss around in the off-season. It's for fun, so indulge me and play along if you like.

After my back of a napkin calculations, we end up with prospect valuations that look something like this:

Prospect Tier Average Surplus Value
Hitters #1-10 $59.08 million
Hitters #11-25 $46.70 million
Hitters #26-50 $25.37 million
Hitters #51-100 $14.42 million
Hitters Grade B $7.7 million
Hitters Grade C $0.84 million
Pitchers #1-10 $37.38 million
Pitchers #11-25 $26.45 million
Pitchers #26-50 $20.58 million
Pitchers #51-100 $10.85 million
Pitchers Grade B $10.22 million
Pitchers Grade C $2.45 million

Let's get the painful part out of the way first. There's no package for David Price that doesn't include at least one of Aaron Sanchez or Marcus Stroman, and probably both. By most lists, Sanchez and Stroman fall in the #26-50 pitcher tier, but some have them in the #51-100 tier. I see them more often in the higher #26-50 range, but to account for this discrepancy let's go with a surplus value somewhere in between the two tiers, but closer to the upper tier. Somewhat arbitrarily, let's go with say... $18 million. That gives us $36 million in surplus value that the Blue Jays would be shipping out.

As much as Anthopoulos has stated publicly that he's trying to avoid including both of them in any one trade, I'm pretty sure this just doesn't happen any other way. Most of the rest of the Blue Jays' high ceiling talent is so far away from the bigs that projecting their surplus value in any meaningful way is all but impossible, so while they may make interesting pieces in a trade, they don't really work well for the purposes of this exercise. Odds are both Sanchez and Stroman would have to go, especially if it's true that the Cubs are still holding out for both in exchange for the likes of Jeff Samardzija who has lots of upside but has yet to prove himself as a pitcher of Price's caliber.

Wait a minute though. Price himself projects to provide $30-35 million in surplus value, so really all we've done here by shipping both the Blue Jays' top pitching prospects to Tampa Bay is replace the surplus value they already had in hand with Price. From a surplus value perspective, this would be considered a fair trade. The Rays won't be looking for a fair trade though, and the market seems to dictate that it will take a significant overpay to acquire him, especially within the division. It's going to take a lot more to get a deal done, as we haven't even started making up the additional $75-100 million or so that Valancius figures the Rays should be seeking to acquire in a trade for Price.


It doesn't help that so much of the Blue Jays' talent in the upper minors is on the pitching side, as a glance at the table shows that on average hitting prospects provide more surplus value than pitchers. This may seem a bit surprising given how valuable pitching always seems to be, but it really shouldn't be such a shock. Remember, TINSTAAPP's a bitch, and even the best starter only takes the hill every fifth day. Really, the only potential impact hitter they have in the upper levels of their system is Anthony Gose, and he may not be quite as shiny a prospect as he was a couple of years ago.

While not technically eligible as a prospect, Gose also isn't necessarily a lock to make the major league roster, and exists somewhere in that weird limbo between being a top prospect and a legit major leaguer. As the Blue Jays' top close-to-the-majors hitting quasi-prospect, I think it's pretty likely that he would be involved in any trade for a David Price or someone similar. As mentioned, some of Gose's prospect shine has probably worn off given that he's only managed to post a 77 wRC+ in his 342 major league plate appearances. Projections have him at about replacement level going forward, but I think they might be selling him a bit short. Gose is never going to hit a ton, so will have to provide the bulk of his value with his defense and base running. So far the defensive metrics aren't real kind to him which is probably affecting his projected value, but they also probably aren't telling the whole story after just 108 games, especially with about half of his innings having been played in a corner outfield spot rather than his natural home in centre field. 

If we go off the assumption that, as I suspect, Gose can provide far more defensive value than the metrics have shown to date then we can start trying to project his surplus value. We can look to guys like Ben Revere, Leonys Martin and Juan Lagares as young players that manage to be roughly average big leaguers despite being significantly below average at the plate.

I plugged Gose into my handy little surplus value calculating spreadsheet at 2 WAR per season for his six team controlled years, which I think he'll be capable of producing with a full season of playing time in centre field. I started at $7 million/win and added 5% inflation every season because I doubt the price of a win is going remain stable or start trending downwards any time in the foreseeable future. I just made some quick guesses at his arbitration numbers that you may feel free to quibble with, but I doubt it will affect the overall picture that much. The results might surprise you.


Year Projected WAR Salary Value Surplus Value
Year 1 2 $0.5 million $14 million $13.5 million
Year 2 2 $0.5 million $14.7 million $14.2 million
Year 3 2 $0.5 million $15.4 million $14.9 million
Year 4 2 $2.5 million $16.2 million $13.7 million
Year 5 2 $4 million $17 million $13 million
Year 6 2 $6 million $17.9 million $11.9 million
Total 12 $14 million $95.2 million $81.2 million

So, even if he never amounts to more than a barely league average player who provides almost all of his value in the field and on the base paths, I don't think it's entirely unreasonable to project Gose providing a little more than $80 million dollars in surplus value over the course of his team-controlled years. You can take issue with some of my assumptions and numbers, but I think that everything is close enough to give us a picture of the kind of value Gose might provide. In this case, it's darn close to what is needed to make up the $75-100 million in surplus value that the Rays might be looking for if trading David Price.

If we figure Sanchez and Stroman at $36 million, and Gose at about $81 million, that gives us $117 million in surplus value heading the Rays' way. Subtract Price's $35 million (let's say) coming back, and that leaves a difference of $82 million that the Rays would be acquiring from the Blue Jays. There's lots of room to debate all of this, as the possible outcomes for all these players are wildly variable. Maybe Sanchez never gets his walks under control or Stroman's stuff doesn't play in the show. Maybe Gose never even hits enough to be more than a defensive fourth outfielder and pinch runner. Even Price's 2013 decline in velocity and K% should probably be a bit of a concern to any team that is acquiring him. The calculations for $/WAR aren't even close to settled science either, so there's plenty of room to debate those numbers as well. Regardless of any quibbles, I think a package of Sanchez, Stroman and Gose falls comfortably enough within Valancius' $75-100 million range to start a reasonable discussion.

Whether or not the Blue Jays should make such a trade is another conversation altogether. Personally, I don't think I'd make that deal, as it would leave the upper levels of the Jays' farm system all but entirely devoid of impact talent and create a sizable gap between the current major league roster and the next wave of prospects percolating in the low minors. Throw in the aforementioned decline in Price's velocity and K% and, while he compensated last season by pretty much refusing to throw a ball, I probably pass on such a deal. While I wouldn't be in favour of such a trade unless pretty much all free agent options have been exhausted, I think it's at least defensible in terms of what the market demands for high-end starting pitching, and possibly even less than the Rays would demand.

What do you think though? Would you trade Sanchez, Stroman, and Anthony Gose for two years of David Price? Would such a package even get a deal done? Perhaps you have different ideas for how to make up the necessary amount of surplus value to pry the ace lefty away from Tampa Bay. Feel free to join the conversation and leave a comment below with your own thoughts. I'd love to hear from you.

Follow Blue Jays Blackboard on Twitter @BJaysBlackboard
Like Blue Jays Blackboard on Facebook


  1. Not all WAR are created equal. There's been a lot of work looking at the marginal value of a win for teams that are projected to win 75, 85, 95 games, etc, with the wins being more valuable for "bubble" teams. But the value of open roster spots has largely been ignored.

    Price only takes up 24 months of a roster spot while a prospect like Gregory Polanco takes up 72 months of a roster spot. The average team gets 1.33 WAR per 25-man roster spot, per year. So you're also losing 4 years of an available roster spot, which on average, is worth 5.33 WAR, or about $35M. The first 1 or 2 WAR might cost $4M apiece while the marginal value of the 6th or 7th WAR might be worth $10M apiece. I just made up those values, but you get the idea This is something that is often overlooked.

  2. There's also the probability of each player providing their surplus value. Price is close to a guarantee, while the other three are question marks.

    Then there's the issue that Price provides his surplus value immediately. The Jays (arguably) need value immediately, not 4-5 years from now. The Jays are pretty good at drafting pitching prospects, so they could probably replace Sanchez and Stroman by the time Price leaves as a free agent. There's only so many roster spots, so I don't know that having, for argument's sake, 10 top 50 pitching prospects means you can just add up their projected surplus value (not sure if this is clear).

  3. I'm not sure I'd agree that Price is a guarantee, but as much as any pitcher who just demonstrated a decline in velocity and K% can be, sure. While the others are question marks, that can work both ways. Remember, the surplus value projections for Stroman and Sanchez are averages of similarly ranked prospects, so while it's possible they provide much less value, it's also possible that they provide much more.

    I agree with your point that Price provides his value immediately, which is why teams generally overpay in future surplus value for immediate wins. It's pretty much the premise of Valancius' piece at D Rays Bay. However, I think Stroman is pretty likely to start providing value in 2014, and depending on what happens with Colby, it's entirely possible that Gose does as well. Sanchez will probably start the following season, if not late in 2014. If we were talking about low minors prospects your point would be well taken, but Stroman, Sanchez and Gose are all players that can help the Jays within their current competitive window, not just 4 or 5 years down the road.

    Both you and the previous commenter make fair points about roster spots, which is one of the reasons I don't think surplus value is the be all end all of transaction analysis. It's not like the Jays have a surplus of starting pitchers though, and they could need a center fielder as soon as 2014, again depending on what happens with Colby and the 2014 season.

    Anyways, my intention wasn't to definitively declare such a package for Price an overpay based on the surplus value being exchanged, but rather to generate a bit of discussion by using some surplus value research I've seen to determine what a package for him might have to look like. I wouldn't love such a deal personally, but I also don't think it would be an insane overpay based on what the market demands for starting pitching.

    I appreciate both the comments. Precisely the kind of discussion I was hoping for :)