So how much is each draft pick really worth?
The following table combines all the factors discussed in the two previous posts to estimate draft pick value in both wins and 2014 dollars. The left side of the graph shows values for the model shown above, with net value calculated directly from the 1991-2005 draftees who stuck around the major leagues.
On the right side, I included values for a hybrid model which uses the same model on the left, but scales it to Sky Andrecheck’s original model. For example, since Sky’s model projects 80 percent of the team-controlled WAR for the fifth overall pick as the 1991-2005 model, the net WAR is also scaled to 80 percent. This is, of course, oversimplified, but for the most part, net value correlates very highly to total WAR.
For both data sets, a 10 percent annual discount was applied to future WAR on a year-to-year basis.
|Insert Title Here||1991-2005||Modified Sky Model|
|Draft Pick||Team||Net WAR||2014 $||Net WAR||2014 $|
|3||Chicago White Sox||3.1||23.6||2.5||19.3|
|9||Toronto Blue Jays||1.9||14.1||1.5||11.1|
|10||New York Mets||1.8||13.3||1.4||10.4|
|11||Toronto Blue Jays||1.7||12.7||1.3||9.9|
|13||San Diego Padres||1.5||11.6||1.2||9.0|
|14||San Francisco Giants||1.5||11.2||1.1||8.7|
|15||Los Angeles Angels||1.4||10.9||1.1||8.4|
|17||Kansas City Royals||1.4||10.4||1.0||8.0|
|20||Tampa Bay Rays||1.3||9.6||1.0||7.3|
|22||Los Angeles Dodgers||1.2||9.1||0.9||6.9|
|26||Boston Red Sox||1.1||8.4||0.8||6.3|
|27||St. Louis Cardinals||1.1||8.2||0.8||6.2|
|28||Kansas City Royals||1.1||8.1||0.8||6.1|
|33||Boston Red Sox||1.0||7.4||0.7||5.5|
|34||St. Louis Cardinals||1.0||7.3||0.7||5.5|
|40||Kansas City Royals||0.9||6.9||0.7||5.1|
Do these numbers look right?
Let’s look back at the 30th pick, which is roughly around the average pick forfeited for signing a player who turned down a qualifying offer. Matt Swartz projected this pick to be worth around $6.7 million. If we’re basing our valuation on more recent draft classes, it looks like the draft pick is a bit more valuable at $7.5 million. This accounts for the increase in player production, but this benefit is partially mitigated by the fact that I discounted future value, but accounted for the immediate cost of the signing bonus.
However, if we scale to Sky’s model and account for the immediate cost of the signing bonus, the pick actually appears slightly less valuable. If you believe that teams are getting better at identifying talented players earlier in the draft, stick to the left side of the chart. If a team is rebuilding, then you might even increase this value by 10-20 percent.
Since production is measured relative to the entire talent pool, there can’t be an increase in value for the entire draft, so if you’re looking at a pick in the 30s or later, I would tend toward the right half of the chart, based on Sky’s original research.
How Are Teams Valuing These Draft Picks?
Luckily, there was one draft-pick trade in the offseason that can help us see whether this model reflects the views of a pair of front offices.
At the end of July of 2013, the Orioles acquired Bud Norris from the Astros, hoping that he would help them compete for a playoff spot down the stretch. In return, the Astros received prospects L.J. Hoes (an outfielder) and Josh Hader (a left-handed pitcher). Neither was considered a top-100 prospect by Baseball America or FanGraphs’ Marc Hulet, although Hoes was rated as one of the Orioles’ top-10 minor leaguers.
While a player like Hoes has some value in the farm system, the fact that Steamer projected him for just 0.3 WAR in 441 plate appearances in 2014, along with the fact that he has produced negative WAR in just over 60 games for the Astros, hints that his value was minimal. Hader ranked just 19th in the Orioles’ system to start 2013, so while the Astros may have seen something special in him, his perceived value from the Orioles’ side was not likely high.
So, the Astros were trading Norris and $213,000 in international spending money for two fringe prospects and the 37th pick in the draft. Based on a few trades during the 2013 season, Jeff Sullivan estimated that teams are willing to spend about two dollars for the right to spend one dollar of international spending money. To keep things simple, we’ll value the international spending money at $500,000.
Before the trade, Norris had produced an average of 1.7 WAR in 170 inning pitched per season. He was having an excellent year before his trade, and given some age-related decline would have been expected to produce about four WAR in his time under contract in Baltimore. Plugging his expected production into our arbitration model, and applying a future discount, we get a net value of $14 million for Norris at the time of his trade.
When we take out the value of the international spending money, we see that Houston gave up about $13.5 million in assets to acquire Hoes, Hader and the competitive balance pick. Our model shows that that pick is worth about $7.1 million.
This tells us that the Astros felt that the two prospects were worth about the same amount as the draft pick they were receiving. On the other hand, they might not value the prospects as highly, but might not discount future wins as much, since they clearly won’t be competing for the next couple years.
That the Astros received two minor league players in the deal tells us that the draft pick was probably worth less than $13.5 million, but the fact that neither player looks like a difference-maker tells us that the pick was one of the primary pieces that was moved. It looks like $7 million is at least in the ballpark.
Are Free Agents Paying the Price for Draft Pick Compensation?
On average, teams are giving up an asset worth around $5-8 million to sign a free agent who turned down a qualifying offer. Therefore, if teams are acting rationally, their offers to these players should reflect this loss.
In 2013, free agents who turned down qualifying offers didn’t seem to suffer on the open market, perhaps with the exception of Kyle Lohse. Things have been much different this year, as Nelson Cruz and Ervin Santana ended up signing one-year deals before the season started. With Stephen Drew returning to the Red Sox, Kendrys Morales is the lone player who appears to be holding out until after the draft to avoid the penalty.
However, the players who did sign multi-year deals did pretty well for themselves. When Dave Cameron looked at the cost of a win in the 2014 offseason, he projected these seven players to earn an average of $6.0 million per WAR. Meanwhile, the 10 players who signed deals for three-plus years but had no draft pick attached were projected for just $4.8 million per WAR.
Let’s look at the teams that gave up draft picks to sign free agents. The Yankees (who gave up three picks) won 85 games in 2013 but failed to make the playoffs. The Yankees may have the greatest financial incentive of any team to make the postseason, which is why they spent nearly half a billion dollars in the offseason.
Like the Yankees, the Orioles and Rangers both finished with winning records in 2013 but failed to make the playoffs. The Braves made the playoffs in 2013, but after losing multiple starters to injury for the 2014 season felt they needed to improve to remain in contention.
The Mariners had a rough season in 2013, but projected to be close to .500 in 2014. More importantly, their roster was loaded with young talent and they had few significant payroll commitments in the near future, and their first-round pick was protected. The Mets? Good luck figuring that one out, although in their defense, their first-round pick was protected as well.
All the teams that gave up first-round draft picks were at a high leverage point in the win curve and would therefore be placing a premium on acquiring wins in the immediate future. While this could explain why these types of teams would be more willing to give up a draft pick, it doesn’t explain why they are paying an additional premium to sign these free agents.
For the Rangers, you could make a case that they had a fairly deep roster and could benefit by upgrading at a single position. However, the Yankees and Orioles both have several weak spots on their roster, and may have been better off spreading out their money elsewhere.
This is admittedly a very small sample, but it is worth monitoring how these players fare in free agency in future years and whether teams are indeed paying a premium (on top of the draft pick) for high-end talent, or if this is simply a one-year phenomenon.
What have we learned from this exercise?
First, it appears that teams are getting better at scouting and drafting amateur players, especially in the first 25 or so picks. Therefore, these picks might be more valuable than thought based on previous models.
As we already knew, the first few draft picks are extremely valuable. The gap in value between the first overall pick and the fifth pick is roughly $23 million. In 2013, only four games separated the Marlins and Twins, but the Marlins’ second pick is projected to be worth $29 million, as opposed to $19 million for the Twins’ fifth overall pick.
If you’re going to tank for a draft pick, make sure you tank big and get a pick near the top of the draft. Outside of the first five to seven picks, the difference in pick value is significantly diminished, and any financial incentive of losing to increase draft pick value is likely offset by attendance and other revenue factors.
Also, we probably won’t be seeing many teams give up picks in the low teens to sign free agents. The average value of the first five unprotected picks (usually 11-15) is around $12 million. This is much higher than the projected $6-8 million value of a pick near the end of the first round or the $4-6 million value for a second-rounder that many other teams would forfeit. Therefore, the team giving up the higher draft pick would need to value a free agent much higher than the rest of the market for it to make sense to pursue him.
Lastly, while these projected values may be slightly higher than previous ones, they may in fact be a bit conservative. As mentioned earlier, I didn’t factor in long-term extensions, which can represent significant extra value to the drafting team.
Also, the new collective bargaining agreement makes the draft slots more valuable than the pick themselves. Previously, if a team lost a draft pick it could try to make up for it by spending extra money in the later rounds. Now, penalties placed on spending in excess of the allocated draft day budget prevent this, while teams that stockpile draft picks have additional flexibility.
When it comes down to it, this is the real reason why Morales hasn’t signed and Drew didn’t sign until this week, and why Cruz and Santana settled for one-year deals. The draft pick penalty is nothing new, but how teams are valuing these same draft picks under the new CBA is.
Combined with the fact that teams are spending more money locking up their young players and fewer stars are hitting free agency, it’s easy to see how the increased emphasis on developing homegrown, cost-controlled talent is inflating draft pick values.
You can argue about the specific value of an individual draft pick, but one thing is certain: These factors have combined to make draft picks more valuable today than at any other time in baseball history. Unless the framework of the draft is reworked in the new CBA (the current one expires at the end of 2016), this trend is likely to only increase in future years.