The New Question at the Top of the Draft

The first round of the Major League draft is just a little over three weeks away, and the Houston Astros will select first for the second consecutive year. Right now, the consensus belief is that there are two college pitchers — Mark Appel of Stanford and Jonathan Gray of Oklahoma — who are a step ahead of the rest, though University of San Diego third baseman Kris Bryant is putting on quite the power display and could be an option if the Astros preferred to build around bats rather than arms. However, the decision for the Astros may not be made simply on talent alone.

Last year was the first draft under the new bonus structure, which assigns a fixed amount of dollars to each team based on where they pick in the draft, with some pretty severe penalties for exceeding those limits. Now, if a team is interested in paying over the slot value for a pick, they’ll have to borrow the money for that overpayment from another pick, making the draft as much a game of cost management as it is talent acquisition.

The Astros showed how the team with the top pick — and the largest overall bonus pool — could play that game last year. MLB assigned a $7.2 million bonus value to the #1 pick in the 2012 draft, a hefty total that was more than many teams were allotted for their entire draft. The Astros decided that they were better off spreading that $7.2 million throughout their draft rather than simply paying the slot bonus to the top pick, so this is how they allocated those funds, per Baseball America:

Round Player Bonus Slot Value Savings
1 Carlos Correa $4,800,000 $7,200,000 $2,400,000
Supplemental Lance McCullers $2,500,000 $1,258,700 -$1,241,300
2 Nolan Fontana $875,000 $844,100 -$30,900
3 Brady Rogers $495,200 $495,200 $0
4 Rio Ruiz $1,850,000 $360,200 -$1,489,800
5 Andrew Aplin $220,000 $269,700 $49,700
6 Brett Phillips $300,000 $201,900 -$98,100
7 Preston Tucker $100,000 $151,400 $51,400
8 Tyler Heineman $125,000 $140,700 $15,700
9 Daniel Minor $50,000 $131,100 $81,100
10 Joe Bircher $20,000 $125,000 $105,000
  Total $11,335,200 $11,178,000 -$157,200

By saving $2.4 million with the top pick, the Astros also were able to sign Lance McCullers and Rio Ruiz for well over the slot values of their supplemental and fourth round picks. In essence, the Astros traded down in the draft, as we noted that the new bonus structures might incentivize a team to do. In the NFL or the NBA, this is totally common, but it actually involves making a trade with another franchise who would rather consolidate their picks into one higher selection. In MLB, you can now essentially trade down with yourself, taking a player who will sign for less than the slot value at the top of the draft in order to give your scouting director more money to play with in the later rounds.

In talking with some folks in the game, the discrepancy in the size of the bonus pools was described as nearly as large an advantage as the ability to pick at the top of the draft itself. There’s certainly a perception that the ability to diversify a large bonus pool and be the only one capable of selecting hard-sign guys who fall later in the draft is an additional advantage, and it’s one that we may very well see another team take advantage of this year. Jim Callis notes that the Astros, Cubs, and Rockies all have total pools over $10 million this year, and one (or more) might decide that the Astros path from last year is a better plan than giving a majority of that pool to their top pick.

However, I think history suggests that there is some real risk to going with quantity over quality when you hold one of the top few selections in the draft. Here is a very informative graph showing the average career WAR by where they were picked in the first round.


This data is a few years old, but the point would hold even if we re-ran the study to include the most recent data. The historical gap between the top selection and the next few picks has been enormous, and when Sky Andrecheck — now employed by the Cleveland Indians — published a follow-up, he noted that this trend had been increasing in recent years. While the #1 overall pick isn’t always a superstar along the lines of Bryce Harper, you are far more likely to land that kind of player with the top pick that you are to land that kind of player at #2 or #3.

Last year, the Astros passed on the consensus top high school talent, Byron Buxton, in order to select Correa, who was rated a few spots lower by most draft analysts. It’s ridiculously early in their careers, but Buxton is destroying the Midwest League right now, hitting .349/.452/.597, good for a 189 wRC+ in full-season ball. The bat was supposed to be the questionable tool with him, with the physical talents to be a superstar all present, but questions lingered about how well he could hit good pitching after playing at a small rural high school in Georgia. As Mike Newman noted a few weeks ago, if Buxton can be an impact bat, then he’s the best prospect in baseball, because everything else was already expected to be elite.

This isn’t any kind of shot at Correa, who is doing just fine in full-season ball himself — and is a nine months younger than Buxton to boot — but Buxton looks like he may very well be the kind of player that you could regret passing on, even if Correa, McCullers, and Ruiz turn into quality contributors down the line. Hitting in low-A ball isn’t the same thing as hitting in the majors, so it’s certainly possible that the Astros made the right call in going for quantity of prospects rather than betting their future on one toolsy outfielder, but it should at least be noted that they had to make a talent trade-off to get additional funds for their later picks.

I don’t think the data supports an open-and-shut case for either option, especially if a team is not convinced that there is a transcendant talent to be selected first overall in that particular draft. Appel, Gray, and Bryant all sound like they have their own strengths and weaknesses, and if there is little difference between them, taking the one that will sign for the least money in order to bump up the talent pool of the later picks would be a good strategy. But before we assume that teams with large bonus pools will actually benefit from distributing that money throughout the draft, let’s remember that they can only do so by taking a lesser prospect at the point where the diminishing returns of moving down a few picks are the greatest.

Essentially, the draft is now engulfed in the stars-and-scrubs debate. Teams at the top of the draft have the option of selecting the cream of the crop and paying the going rate for that type of player, or taking a slightly lower rated player in order to improve their secondary picks. In some years, it’s probably right to take the top player on the board, and in some years, the difference might be so small as to make “trading down” the right call. The key is to figuring out which year you’re drafting in. The Astros decision last year suggests that they did not see a big enough gap between Buxton and Correa to justify spending most of their bonus pool on their first pick. This year, they might decide that Appel, Gray, or Bryant is worth the full slot bonus, and play it straight the rest of the draft.

Either way, I think it is worth noting that no matter what they decide, they’re making a trade-off. There isn’t a clear cut best path, at least not without knowing the future of unpredictable young players. So, while the teams with the largest draft pools also have the most flexibility, I’m not sure that I see the ability to trade down as a huge advantage in addition to picking the best players. It’s mostly an either/or. They can pick the best player, or they can have flexibility in the rest of the draft. Under this system, they can’t really do both.

Print This Post

Dave is a co-founder of and contributes to the Wall Street Journal.

37 Responses to “The New Question at the Top of the Draft”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. @AHume92 says:

    Doesn’t this setup run somewhat counter to what the draft should be about? The idea should be that the worst teams end up with the best players. Now (unlike the old system) a team selecting the best player available has to make a series of sacrifices with further picks. This will likely leave some upper-echelon players (in any given draft) available for taking by a mid-order team, who aren’t in a critical rebuild phase and can afford to punt other picks. The only solutions I see are going back to the old system, or (better) setting hard slots.

    Vote -1 Vote +1

    • Krog says:

      The draft exists mostly to limit the bargaining power of amateur players. The concerns over fairness and helping bad teams is secondary.

      +18 Vote -1 Vote +1

    • drewcorb says:

      The worst teams have the most money to play with because they have the highest-slotted picks. They have the advantage that they can be more flexible in their strategy for acquiring the best players.

      +7 Vote -1 Vote +1

  2. Terence says:

    I’m sure it would be difficult and time consuming, but what if we ordered players by signing bonus rather than where they were picked in the draft and then compared war totals. Front offices have been dealing with the issue of talent vs cost in the draft for quite some time, and the cost of signing a player might better reflect true talent and thus WAR production than ordinal position does. If this is true and the Astros signed three top-20 players instead of of a #1 last year, then there is really good reason to do it again.

    +9 Vote -1 Vote +1

    • Owen says:

      You did a much better job of articulating the point I was trying to make below.

      Vote -1 Vote +1

    • Dave Cameron says:

      It’d be worth doing, but in actuality, the argument would only strengthen the case to take the best player. If the best player occasionally fell to #3 or #5 or #10 or wherever because of signability concerns — you could probably make that case with Matt Wieters and Jared Weaver, for instance — than that would be artificially propping up the value of lower picks. Re-bucketing players by signing bonus would likely serve to make the gap between the picks larger, not smaller.

      Vote -1 Vote +1

      • Terence says:

        I have no doubt that taking the #1 player is better than taking the #2 player, or #5 or #10 or whatever (no matter how you measure pre-signing value). But my hypothesis is that taking the #2, #12, and #23 players (how Correa, McCullers, and Ruiz were paid) is more valuable than taking the #1, #41, and #129 (where they were drafted). I don’t have the information necessary to prove or disprove that notion. I could believe either side of the argument. I certainly believe the #10 and #20 prospect fall due to signability concerns more and more often than the #1 would.

        For what it is worth, if we plug those 2 sets into the equation from Mr. Andrecheck’s article we get expected WAR values of 23.8 and 24.7.

        Vote -1 Vote +1

  3. Owen says:

    Dave, interesting article but doesn’t some of this rest on the notion that historically the top talent has always been the 1st pick. In fact the idea of trading talent for dollars has always been present, albeit for different reasons.

    Vote -1 Vote +1

    • maguro says:

      Teams do occasionally make picks for reasons other than pure talent. But if factors other than talent were significant drivers of who got picked early in the draft, you would expect the win curve to be a lot less steep that it is.

      Vote -1 Vote +1

  4. Mike Green says:

    Career WAR isn’t helpful. You need to know pre-arb WAR vs. draft order.

    The graph probably will still look something like this, but it will be somewhat flatter. The thing is that occasionally there is a consensus best player- a Griffey or an A-Rod, and these guys generally do very well, but when there isn’t, I don’t know that the difference between the #1 pick and the #3 pick (for instance) is very large.

    Vote -1 Vote +1

    • Dave Cameron says:

      Follow the second link to Sky’s piece on Baseball Analysts, as he did just that, and it didn’t change anything in a material way.

      And, really, you need to account for more than the first six years, especially in the age of the long term extension, because the average tenure of a star player with his original team is much longer than six years. We cannot ignore the value that teams derive from the ability to sign franchise players to long term deals early in their carers.

      Vote -1 Vote +1

      • SKob says:

        Just remove Scott Boras clients like Ellsbury and you might actually have a representative sample. Too bad greedy agents make for strong outliers.

        Vote -1 Vote +1

  5. Mike Green says:

    Ack. It should be pre-free-agency WAR vs. draft order.

    Vote -1 Vote +1

  6. Ben says:

    Of course, the real way to solve this would be to just let the teams sort this out by themselves. Eliminate the stupid salary allotment ceiling and just let teams trade picks. Free market for the win!

    Vote -1 Vote +1

    • leeroy says:

      Although I’m a strong proponent of free market economics, this just wouldn’t make sense from the MLB’s standpoint. By eliminating the salary allotment they would drive prices of picks up towards fair market value. More money for amateurs, less money for teams. The price ceiling essentially forces players to take less money than FMV, since they lack other options. Might be interesting to see if an amateur ever opts to go to Japan for more money because of this allotment. If that became more commonplace MLB would have to rethink draft slotting.

      Vote -1 Vote +1

      • Ben says:

        I’m confused as to how driving prices of picks up towards fair market value would not make sense. I think it actually makes perfect sense. Maybe not for the owners, but certainly for the players.

        More money for amateurs, less for teams (I’m assuming by ‘teams’ you mean current professionals) only works if owners are maxing out what they can or will spend on their payroll or if there is a hard cap on total spending, but that’s not what currently exists. Payroll is not even in the same stratosphere as revenue. Payroll is entirely a function of projected revenue (will I make more money if I sign X player for x dollars?). That is all drafting would be as well.

        Also, we had a system not too long ago with no allotment caps that seemed to be working just fine. Teams like the Rays and A’s were exploiting the crap out of it. It’s unclear what, if any, impact trading would really have on that. Perhaps some players could try to demand a certain salary or claim they would only sign with a certain team, but that would just leave teams with a determination to make: trade a high pick to that teams, and get something of value, or call the bluff and force the kid back to college for another year (maybe more if it’s a HSer) unless he signs. We saw that situation unfold all the time under the previous system, and the only problems were ones that teams imposed upon themselves.

        Also, while prices might ‘approach’ FMV, there would still be a pretty sizable gap between FMV and actual price point because of completely uneven bargaining positions. The teams would hold significantly more leverage than the players.

        tl;dr: free market would handle this just fine. We have what we have because of what is basically legal collusion.

        Vote -1 Vote +1

        • Jason says:

          “maybe not for the owners but certainly for the players”

          Um, the players would obviously choose immediate free agency. The owners are the ones who have chosen to have a draft to maximize their profit. By ‘team’ he meant not current professionals, but the team ownership.

          Vote -1 Vote +1

    • Matt says:

      The problem with this idea is that there are two side who negotiate how to handle the draft:

      The Owners: who want to limit their investment as much as possible

      The Players: who want as much money as possible for themselves rather than giving it to future ‘potential’ players

      Buy implementing these ‘cap’ numbers – both sides seem to be pretty happy. Although, I don’t see why you can’t trade picks anyway (the salary allotment would have to be traded along with the pick).

      Vote -1 Vote +1

      • Ben says:

        Your premise is flawed, albeit perhaps not technically. When you say there are two sides who negotiate how to handle the draft, you are of course correct. But this is a problem, because in reality there are THREE sets of interests: owners, players, and those who are not yet players. While the system may work just fine for owners and players, it works quite poorly for future players.

        In reality, it also works quite poorly for players whether they realize it or not. They are not getting any more money because of this system, even if they think they are. Payroll doesn’t come even remotely close to revenue. The owners could easily afford both. In reality, players are just reducing total income for future players.

        Vote -1 Vote +1

  7. suicide squeeze says:

    Here’s the thing, though: You’re presupposing that the #1 pick in this draft has a chance to be a superstar on the level of Griffey, Arod, or (perhaps) Harper. Some years that may be the case, but I don’t think most people are going to make the argument that Appel or someone else is that awesome of a pick. Just because all historical #1 picks have averaged so and so WAR doesn’t mean that we can expect that much WAR from this year’s #1. Given that there isn’t someone of that level, shouldn’t the Astros spread out the money and get as many good lottery tickets as possible? If Harper was at the top of this draft, I have a feeling that the Astros wouldn’t think twice about taking him.

    Vote -1 Vote +1

  8. hoser says:

    For any given team, there may not be a tradeoff. If the Astros thought Correa was a better player than Buxton, they lost nothing. If it turns out to that Buxton is better, then that is just the inherent risk of talent evaluation, and the Astros just hedged their bet.

    Vote -1 Vote +1

    • Ben Hall says:


      While the piece is still a good thought experiment, I don’t think it accurately conveys the decision the Astros made last year. Everything I’ve read has said that the Astros either thought Correa was the best player, or that it was a toss-up between him, Buxton, and Appel. It is true that if Buxton is an impact bat then he’s a true franchise player, but that was a very big if at draft time last year. This was not a situation of choosing Correa over a player like Bryce Harper or Stephen Strasburg.

      Vote -1 Vote +1

      • E-Dub says:

        This is my thinking as well. As a SS with power potential, who is a full nine months younger than Buxton, Correa had a very good case for 1-1 status. I also agree that the players they were able to sign as a result (McCullers and Ruiz in particular), make this more interesting than the strict argument of picks in a vacuum.

        Vote -1 Vote +1

  9. Pr says:

    I know it’s unknown but isn’t the missing part of this article what the bonus demands are expected to be for the three players? How much would the Astros save by taking Gray? Same savings as Correa?

    Vote -1 Vote +1

  10. ttnorm says:

    The other thing is that “trading down” only makes sense if you can grab a $1M+ savings. Not many teams can do that (without completely punting on the higher pick). You can’t get a high level tough sign with less than that.

    Vote -1 Vote +1

  11. I’ve been saying this since I was one of the first to post a draft analysis publicly on average anything is really useless in analyzing the draft.

    You need to define what is good and what is not for your analysis to have good value in understanding the draft. Using average value is great if you have a bell shape curve to use, but in the draft, you have a very skewed data set because the vast majority of drafted players never even make the majors, let alone be a good player. The data needs to reflect this skewing.

    Still, the article’s point actually reflects what my study showed: if you want better odds of finding a good to great player, you need one of the top picks available, top 5 ideally.

    Vote -1 Vote +1

  12. soladoras says:

    What is the motivation for a player like Correa to be more “signable” and accept 1-2 million less than his slot value? From what I remember, he didn’t have a mid 1st round grade, to where a jump to #1 was a massive pay increase. He was already a top 3-4 overall prospect in that draft. Why not just ask for slot money?

    Vote -1 Vote +1

    • Terence says:

      It’s not about what the slot is worth, it’s about what are you worth. Carlos Correa got paid more money than anybody in the draft not named Byron Buxton. If you’re a top 3-4 prospect and you get paid #2 money, you just won.

      I’m sure the number that the Astros gave Correa’s agent was higher than what the Twins, Mariners, or Orioles would have been willing/able to pay him, and so he was glad to accept his best offer.

      Vote -1 Vote +1

  13. mike wants wins says:

    Just let teams trade actual picks……I know, that’s not the point here, but I could not help myself……..

    I’d actually like to see a limited FA for the worst 10 teams….something like the worst 10 teams get to sign any 2 amateurs they want, give them a pool of money for this, and let them either sign 1 or 2 players. Then go to the draft next (or, the worst 10 teams averaged for the last three years get a pool to sign, then a draft).

    That increases the free market some, while keeping the concept of the draft in tact (which has its flaws, frankly, but might be needed, not sure).

    Vote -1 Vote +1

  14. Paul Wilson says:

    Appel is going to play hard-ball and ask for more than slot. Scott Boras has not had the chance to play his hand out with regard to the strategy of the new draft, esp after last year’s dead-end negotiations with the Pirates. He’s got a strategy up his sleeve that will most likely be very effective and probably involves asking for more than slot until 2 min before the deadline.

    Vote -1 Vote +1

  15. GLS says:

    I wonder what would happen if they got rid of the draft altogether? Keep the signing bonus pools so the worst teams have the most money, etc., but let the teams recruit and sign who they want and let the players choose where they sign.

    Vote -1 Vote +1

  16. Ams says:

    For a few years, teams will be faced with the choice, but, after that, won’t the bonus expectations of the projected #1 pick adjust? Once teams show they are willing to select lower ranked players to spread the pool, shouldn’t they be able to get the best player available for a few hundred thousand above the slot $ for the second spot?

    Vote -1 Vote +1

  17. hobbes020 says:

    It’s an intresting concept, but I believe you are assuming way too much (as most are IMO) about selecting Correa over Buxton/Appel. Your argument seems to hinge on the fact that the Astros knowingly and willingly selected what they thought was an inferior prospect (even if only slightly) in order to save money.

    Regardless of what the ‘consensus’ was as to who the top player was its up to each individual team to decide their top guy, and their own order or preference. The combination of ‘consensus’ and signing below slot makes it easy for the jump in thought that they took a lesser play in their eyes. This may or may not be true, but to assume its true because of these other two points isnt logical IMO. The Astros could very well have Correa higher than everyone else on their boards AND they were able to sign him for less. If so they go to have their cake and eat it too.

    The other point that adds to this idea is that Buxton as you said is tearing it up and looks great so far (not to discount Correa). However, we can’t asign intent to pass on quality after the fact with hindsight. If Victor Roache turns into Ryan Braun in 5 years, the teams who passed on him didnt do so to save money. Its evaluation at the time, and I believe Buxton’s start is clouding the issue. Its rear-view.

    It makes for an interesting thought/narrative but I dont believe the basis that they intentionally ‘went cheap’ with the draft is a sure bet. It could have just fallen together perfectly in this instance.

    Vote -1 Vote +1

  18. Joebrady says:

    The problem, imho, is that the Houston strategy might lack linearity. Let’s say the #1 is worth $7.0M, and the #2 is worth $6.4M. If you made the #2 pick your #1 pick, would he still settle for $6.4M? If he signs for $6.6M, you saved .4M to spend later, but you lost .2M in overall value. You would almost need to have an agreement in place with the agent before you took this gamble.

    Then you’d need to know that the value can be recouped later. If McCullers was worth more than the $2.5M they gave him, maybe they make out. But wouldn’t that mean McCullers would’ve been a #11/#12 pick last year, except for signability issues?

    I think I’d rather be on the other side of the issue. I’d prefer someone (Bryant?) fall to the Red Sox on signability issues, to see if the RS can sign him for a little less than his true value, and make it up elsewhere.

    Vote -1 Vote +1

  19. Ian says:

    I think it might make sense depending on your farm system. The Astros aren’t one player away. They had a horrible system so they saw a way to add 3 decent prospects to that system. (Assuming they didn’t think Correa was the top talent). Something like that might help the Astros again this year. Would anyone really rip them if they took Stewart and used that savings to add a couple more quality prospects late so they can build a nucleus of young players up together? If they took Appel, by the time Correa makes the majors, Appel is probably in arb 2 year.

    A team like the Twins would be the opposite, they have a pretty decent farm system (through trades and player development) so for them, this year, they would be wise to go best player instead of trying to get a few more prospects and hope that player could jump into their Hicks/Arcia/Gibson/Sano wave

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>