The Changing Value of Draft Picks
With January here and a few big name free agents still on the market, one of the popular stories of the week is how the new compensation system is affecting free agent pricing. I wrote about it yesterday, as did Jeff Passan, and this morning, C. Trent Rosencrans documented some comments Kyle Lohse made on the radio. He seems to have a pretty good understanding of what’s going on:
“It’s not exactly an open, free market when you attach such things on a guy like myself, but yet a guy like a Zack Greinke or Anibal Sanchez got a get-out-of-jail-free card because they got traded midseason, so the rules don’t pertain to them,” Lohse said. “I’m obviously a little biased, but the rules could use some tweaking.”
Of course, Greinke and Sanchez also have the advantage of being younger and better than Lohse, but their lack of draft pick compensation almost certainly helped drive up their price a little bit, so Lohse isn’t wrong here. But, rather than spend another day talking about the rules themselves, I want to talk about the actual change that teams appear to be making this winter – significantly inflating the valuation of a draft pick.
It’s not like draft pick compensation for signing a free agent is a new thing, after all. Last year, the Angels, Tigers, and Phillies all surrendered first round picks in order to sign Type A free agents. The year before, the White Sox, Tigers, Rangers, Red Sox, Yankees, and Phillies did the same. In 2010, it was the Mariners, Braves, Tigers, and Red Sox. Detroit clearly had no problem surrendering draft picks to sign a free agent, as they did it three years in a row, but between those three years, you have nearly a third of the teams in baseball punting a first round pick to sign a free agent. This wasn’t just limited to a small group of teams who valued veterans more than prospects. Punting a first round pick to sign a premium free agent has always been part of the deal.
But, of course, punting that pick now comes at a greater cost, as the new draft rules also lower a team’s bonus pool, making it impossible for a team to simply reallocate their budget to later picks in order to mitigate the loss of the high pick to some extent. And, by decreasing the pool of protected picks from the top 15 to the top 10, the rules also make it more difficult to for losing teams to sign a free agent while retaining their first selection. There’s no question that the combination of these changes makes the cost of signing a qualifying offer free agent higher than signing a Type A free agent used to be.
But, at the end of the day, the cost is still a draft selection. And mid-to-late first round draft picks aren’t any more likely to turn into future big leaguers now than they were under the old system. The reward for having that kind of pick hasn’t changed significantly, and there is perhaps no better example of diminishing returns in baseball than the expected production of players by draft position. Let me borrow a chart from this well done post at Baseball Analysts on the subject to illustrate the effect.
The first pick in any given draft is insanely valuable. The next few picks are pretty great too. It falls off in a hurry, though. When Sky Andrecheck (now with the Indians, by the way) did this analysis back in 2009, he had the #1 pick producing an average of +20 WAR, but the #10 pick was just at +6 WAR, while the #30 pick was only +4 WAR, and then every pick after that leveled off pretty substantially, to the point where it’s not clear that a fourth round pick is that much more valuable than an eighth round pick.
It’s probably not a coincidence that the CBA left the top 10 picks protected, because those regularly turn out to be franchise building block kinds of players. The middle to end of the first round, though? It’s all kind of a crapshoot, with scattered stars here and there and way more busts than successes. You don’t want to just give those picks away for no reason, but at the same time, we have to acknowledge that these picks aren’t nearly as valuable as the earlier selections, and teams are generally doing well if they get a league average player with a mid-to-late first round pick.
Just as an example, let’s take the 11th pick in the draft since the turn of the century. This is the best possible pick that can be surrendered as compensation for signing a free agent. From 2000 to 2012, those picks have been:
2000: Dave Krynzel
2001: Kenny Baugh
2002: Jeremy Hermida
2003: Michael Aubrey
2004: Neil Walker
2005: Andrew McCutchen
2006: Max Scherzer
2007: Phillippe Aumont
2008: Justin Smoak
2009: Tyler Matzek
2010: Deck Mcguire
2011: George Springer
2012: Addison Russell
McCutchen is a huge score for the Pirates, and Walker has turned out nicely for them as well, even though he took a while to become a useful player. Scherzer is another win for this slot. Addison Russell had a great debut and has prospecters excited, so he’s probably got a decent amount of trade value at the moment. George Springer might turn into something, maybe. But after that? Justin Smoak hasn’t developed as expected. Tyler Matzek is still trying to figure things out in A-ball. Phillippe Aumont might have a future as a bulpen guy. Yeah, try to contain your enthusiasm.
A one-in-ten shot at getting the next Andrew McCutchen is an asset worth having, especially if some of the consolation prizes are pretty decent too. But, let’s keep in mind that the changes to the draft rules haven’t changed the fact that teams are still drafting kids who aren’t anywhere close to the big leagues, and the path from draft pick to productive big leaguer is long and treacherous. Having draft picks is a good thing. Teams should value them. But they should know what that value is, and be willing to trade it in for an established Major League player when the price is right.
Simply looking at a guy like Kyle Lohse and saying “he’s not worth a first round pick” is just incorrect. He’s not worth a first round pick and the kind of contract he was probably looking for at the start of the winter. But if a few more weeks pass and no one is stepping up with any kind of multi-year offer, even Scott Boras will eventually bring the price tag down. As Lohse notes, he’s going to sign eventually. The key is to find the right number that makes it worthwhile for a team to give up that pick.
What’s the right number for a 34-year-old innings eater with peripherals that suggest he’s an average-ish starter? Well, if we think he’s a +2 to +3 win pitcher, that makes him worth something between $10 and $15 million per year. In a lot of ways, he’s the pitching version of Torii Hunter, who signed for 2/26 without any compensation attached. So, depending on what price a team puts on their draft pick, maybe the right price for Lohse is 2/20, or 3/24, or something in that range. After all, there is real value in having a guy who can throw 200 average innings, especially for a team with a weak rotation who has aspirations of contending in 2013.
The same goes for Michael Bourn and Rafael Soriano. Yeah, it’s a speed-and-defense guy heading into his 30s, and its a reliever with some history of arm problems, and yeah, they’ll cost you your first round pick if you sign them. But pricing the cost of that pick into the offer is different than just not offering a contract to begin with. There are teams out there who could use Michael Bourn and Rafael Soriano more than they could use a first round draft choice in the #11-#25 range.
Even a rebuilding team should see these guys as potential options at the right price, as they could theoretically sacrifice their first round pick to sign them to a deflated contract, then market them as trade chips this summer, when there aren’t as many ways to acquire talent and the loss of a draft pick is no longer in play. While MLB might frown upon sign-and-trades during the same winter, they have no capability of stopping a team from trading a player in July that they signed in January. Teams regularly sign players to short term deals, then flip them for prospects at the deadline. This isn’t a new strategy, it’s just a new twist on it.
In fact, if I’m Scott Boras, that might be exactly what I start pushing for pretty soon. I only half-jokingly suggested that Roy Oswalt ask for a “must trade clause” last winter, allowing him to spend the start of 2012 with a non-contender in a pitcher’s park who would then flip him to a contender at the deadline, so that he could be part of a pennant race and not have to spend his whole season on a losing team. Perhaps Boras should aim for something similar for Soriano, even if it’s not actually spelled out in the contract. Landing Soriano or Lohse in a pitcher-friendly environment for a few months, then having them traded at the deadline, might be the best possible outcome for both players, and a team could potentially get more in return for them in June or July than the expected value of the pick they gave up to sign them in the first place.
The changes to the draft have made specific picks more valuable, and the cost of giving them up more harmful, but there still needs to be a realistic valuation of that selection in relationship to having a quality Major League player on your team in 2013, even if that player might not finish the season on your roster. Lohse, Bourn, and Soriano are worth giving up a first round pick for. The key is to just find the right price, and the right landing spot, in order to make sure that everyone comes out ahead.
Dave is the Managing Editor of FanGraphs.
Why in the world did Soriano turn down $13MM/1 to test free agency when he made pretty much the exact opposite decision accepting arb/essentially forcing a trade last time he had the option? If he thought 9MM/1 was a better bet than the market then why would he expect to fare better this time around w/ that much more mileage on his arm and more steep compensation attached?
My guess is that Boras was pretty certain that he would get the Tigers (or maybe someone else, but most likely the Tigers) to give him a multi-year deal at 10+ million per, except the Tigers aren’t cooperating.
So how long does the compensation attach for? Just out of curiosity, if a guy hit free agency when he was say 27 and really felt like the qualifying offer was dragging down his market could he say, sign a 1 year deal to play in Japan and then come back the next year without the compensation pick attached to him?
Even better for the really bad rebuilders with protected picks. I’m tempted to argue for the Twins to sign Bourn and use him to get Hicks some extra minor league time.
Nice chunk of risk attached, though, signing a guy like that to a 3- or 4-year deal in hopes of trading it later.
This was not supposed to end up here. I typed it at the bottom.
This is an interesting question. What if you’re in a situation where a guy got a qualifying offer, refused it (for whatever reason), then realized that he wasn’t worth the draftpick a team would have to give up. IE, even at vet-min, he doesn’t bring back enough value to offset the $7M or so a late first round pick is worth? (either he really misjudged the market, or got hurt while an FA, or whatever)
Is he just screwed? Is there a point where a team can sign him later in the season and not lose the pick?
If you are good enough to be worth 13mil/1year qualifying offer, Japan won’t pay you nearly enough to make it worth leaving MLB. Probably. The top Japanese salary over there is something like 5mil/year.
Once the season starts if they are still unsigned it might alter it, but the only rule I know of for sure is, if they don’t sign a major league deal comp doesn’t apply. So you could get a minor league invite to spring for a compensation player, without relinquishing the pick. I think any team with a protected pick, the Cubs or even better the Indians would gain value from the lost pick to return on trade. Look at the Wheeler/Beltran deal, and tell me Wheeler where he’s at in his progression/projection isn’t better then anything they take in round 2-5. Cleveland should be sneaky, and sign all of Boras’s remaining big three to one year deals where they won’t have comp next winter after a trade.
Previously, Type A/B compensation lasted until the following draft. After the draft (when compensation picks actually happen), players would be free to sign without compensation attached. I do not know if they changed that with the new CBA, but it would be my guess that it remains the de facto deadline.
Using the 11th pick is a little misleading, for whatever reason the 11th picks have as a whole done worse than 12-17th. David Schoenfeld just published a piece that averaged the WAR of each of the 10-25th picks.
The average WAR per round (using the drafts from 1990 to 2010)
11: 2.5
12: 9.0
13: 7.5
14: 5.5
15: 5.7
16: 6.0
17: 6.4
18: 0.8
19: 3.4
20: 11.9
There’s going to be some variation due to fairly small sample size. And that’s ignoring the likelihood of the prospect not reaching the majors at all, which would be an interesting analysis on it’s own. The value of the FA’s left really aren’t that much better, the real value you’re buying is short-term certainty.
With the new draft rules, you may get less value since signability picks may not be picked up in later rounds, due to bonus rules (a la Appel). But I think teams are still valuing these picks at around 4-6 WAR.
The numbers are skewed because of the previous way the draft worked. Teams like the Yankees taking top of the first round talent in the back of the first, and overpaying to get them, made it so that the bottom half of the first round outperformed its projected win expectancy. Top talent will sign late now far less then in the past, and I believe you will see a better curve on production out of picks going forwarded.
Yep Kyle, its all because of that comp pick that you arn’t getting Greinke money. It has nothing to do with your average season being roughly equal (by WAR) to Paul Maholm.
Would the CBA or Players Association prohibit a clause that paid a player a huge bonus if they played a single inning in August for the team they signed with? This kind of a “poison pill” clause could act as a “must trade” feature without actually spelling it out.
I think you could do it with a large player option, that becomes mutual if traded.
Isn’t just looking at WAR for the draft picks kind of missing the point? I think it would be more accurate to look at $/WAR for past draft picks after round ten and compare that to the $/WAR of the free agent.
What happens if a team like the Braves that has already lost their first round pick signed Lohse? Not that the have any reason, just curious.
They lose their 2nd.
without the massive swath of compensation picks(28 in 2012) second round picks will now be worth a lot more as well. In the 2012 draft the WS Champion Cardinals picked 5 times before the last place Astros picked twice, In 2011 winner of 91 games the Rays picked 10 times before the last place Pirates picked twice. While some compensation is owed to a team that loses a player like Pujols that isn’t going to help competitive balance.
Don’t forget about the “competitive balance” round after the compensation picks. That bumps the 2nd round back another 6-7 picks.
I think you’re also forgetting the value of draft picks as trade currency. For example Toronto previously stock piled draft picks from 2010-2012, they were able to turn a lot of those picks into Reyes, Buehrle, Dickey, Johnson, Bonifacio and Thole this winter. Without having the extra currency it’s unlikely those trades get done.
Data geekery: If you have access to the data, in the graph average WAR vs. draft pick #, the pattern strongly calls for both variables to be logged. If you point me to a data link, I’ll do it, but I didn’t see it at the original post.
I don’t really get why people compare pick #X in different years as if it means anything. A pick should be as valuable as the best player still available, not who has historically been taken in the same spot. Not to mention the various reasons when a player is chosen, for example Mark Appel at #8.
The idea is to get an idea of the average value of the draft pick. On average, the 11th best player in the draft (assuming they’re being drafted 11th), is worth X WAR. Therefore, not knowing far enough in advance which players will be available to draft and now they’ll look at the time, teams have to use their guess at the average value of a draft pick when making the decision to lose it or not.
Jason is speaking to a larger point of methodology which is both relevant and interesting, however, though it might be better if phrased somewhat differently. _Who_ gets picked in a given slot has many variables which have little or nothing to do with the absolute WAR that individual happens to return over their career. An org’s budget; a team’s positional needs’ a particular draft honcho’s area of expertise; luck; gross stupidity; the fact the the better guy had a bad day when the scouting director had eyeballs on him; &etc. Thus evaluating the value of _a_ pick slot by looking at _the_ person picked in that slot invites in all of those extranea into the analysis. That muddies the result rather than invalidates the conclusion of this post as refined in comments. That seems to me Jason’s larger objection. But can we do better?
Jason argues that a better procedure might be to retrospectively set the ‘peak value available per slot’ at the WAR of best performing individual subsequently taken. To me, that’s not a good bet, exactly because no one really knows _at the time_ who the best individual remaining is. Therefore, it’s impossible for a FO team the winter prior to a draft to handicap what the value of the slot their giving up is by using that methodology. This is the old ‘we shudda-cudda had X’ meme creeping in which really isn’t useful because _nobody_ knows who X is until post facto.
A different way to go, if cumbersome to go back and calculate, would be to evaluate the aggregate remaining WAR produced by the draft class from any particular slot on down in a particular draft. Exactly how one weights that on a per slot basis would make a nice game for the actuaries here, it’s not as simple as some percentage for the remaining draft slots even though that does include all the washouts and so perhaps more accurately weights the worth of a slot pick. The best way would be so aggregate tiers by sequence taken, if not necessarily by ‘draft rounds’ per ce, and then to get some weighted WAR per slot, both on a draft by draft basis (some are deeper than others) and overall. The refinements of outcome might not be much different than the 4-6 WAR mentioned by Klatz in comments above, but the conclusion would be more solid with less noise in it. Hmmmm . . . . well it’s up for grabs to anyone interested.
Even in a smart post like this one, writers keep failing to acknowledge this fact:
The reason draft pick compensation exists at this point in this form is PRECISELY to lower the salary offers a guy like Kyle Lohse gets. A secondary benefit is that it provides an incentive for guys like him to resign, and for MLB as a whole having players stay with the same team, is a good thing.
If Lohse and Adam LaRoche and Soriano aren’t getting big offers, that means the CBA is working as designed.
Yes. Good.
But also Monday morning has not dawned yet, so too early for quarterbacking..
So Blake, I agree completely, but would extend that. The function of compensation is to drive down the acquisition costs of free agent players so tagged. But it is also to _push_ down the number and value of offers from a smaller pool of competing orgs. Yes, MLB wants to pay the individuals less of the money pot. But an even more important goal is to keep ‘that fool owner’ and his GM from overbidding for mediocre talent, thus forcing _whoever_ signs the guy to overpay. Reining in ‘fool overpays’ and damping down feeding frenzies for middle and back end of the roster guys is the goal in general of the ownership from the CBA as presently designed. And lo and behold, that goal is being implemented quite effectively.
I don’t wish Bourn or Lohse, or Soriano or Adam L any particular ill will when I say there _should_ be no feeding frenzy for their services, even though these are all major league talents quite useful to the right org. But if the owners slot bids for these guys “That’s collusion!!!!” Sooo, here we are: if a _negotiation_ produces a procedure, that passes muster. This is exactly what the result was supposed to be.
“Reining in ‘fool overpays’ and damping down feeding frenzies for middle and back end of the roster guys is the goal in general of the ownership from the CBA as presently designed. ”
Right, but the salary cap/luxury tax threshhold draft pick penalties do that just fine.
This doesn’t just affect the first pick but the entire draft. Most teams signed their first round pick below slot so they could overpay for guys in the latter rounds. It allowed the teams to sign a couple of guys out of their college commitments or convince them to forgo another year. It’s a compounding affect if you forfiet your first pick. You basically then forfeit an overall stronger draft.
From the standpoint of the draftees and their advisors, that’s a problem. From the standpoint of MLB, that’s a feature not a bug. I’m just sayin’ . . . .
Looking at this as purely a “losing a single draft pick” and “what is the expected value of that pick” problem is the problem. This is the narrow thinking that Jim Callis used to claim that a record number of amateurs would not sign last year. Had the opposite effect. And teams quickly learned why as the process went along.
It’s not about a single pick or the loss of the slot money for that single pick. Losing a draft pick has a cascade effect on amateur player acquisition since there is also a cap on international amateur FA money. Losing the pick completely changes the way teams acquire players. There is no way to be confident that you got a decent big leaguer at worst in round 1, so you can take a flier on a tooled up guy in round 3. Suddenly that looks like a massive risk, especially since there is more incentive now for a legit guy to sign for slot, like a college junior who under the old system would have waited till next year and the likelihood of taking a step back was minimal while the reward for moving forward was so great. And on and on.
Also, think about it this way. If you’re a GM and you punt a first rounder, how does that go down in the scouting department? You can’t just tell them you’ll increase the budget for the draft and let them take a few over slot guys now. You would really run the risk of demoralizing them, which could have long term consequences. I was of the opinion that the Pirates blowing their pick on Appel last year was a fireable offense, and I maintain that stance.
People forget but the Pirates will get a compensation pick in 2013 for losing their pick. If they just lost the pick, yeah, I can see fireable, but maybe they made the judgement that Appel was a big enough talent to take a flier on in case they can get him to sign, but if not, they got the pick in 2013, just one back, and who knows, maybe they thought the talent was deeper in 2013 vs. 2012, and might be better than anyone else they might have picked in 2012 other than Appel.
I did not forget that and you clearly miss the point. In a highly restrictive environment you can’t afford to make a mistake like that. Get it wrong on a guy, guy get injured, it happens. That mistake was entirely avoidable. They lost a year of development of a top 8 player, who could get to Pittsburgh one year earlier. Again, cascade effect. Now in addition to their top two pitching prospects losing some shine and looking like they’ll take longer, they got squat out of the last year’s first round. Especially for a small market team like them, it’s a killer.
The circumstances surrounding the pick are even worse. They never even talked to Boras before they picked him. And if you’re right that they thought the 2013 talent was better, that’s even worse. A lot can change in a year, especially since in June there are still a ton of guys yet to really emerge on the road show circuit. The view on 2013 would have to be that the talent is overwhelming, and that is not the view in the scouting community.
“they got squat out of the last year’s first round”
And they get double out of this year’s first round.
“They never even talked to Boras before they picked him.”
Worth noting that the source on this, one S. Boras, has completely contradicted himself in his public statements on the question of whether the Pirates could’ve signed Appel.
I was just having this argument with someone about the Red Sox signing Adam LaRoche. They were adamant that LaRoche was just plain not worth a 2nd round pick (1st rounder is top-10 protected). I argued that assuming a confluence of unlikely things that saw LaRoche’s contract demands drop to something like 2/18, LaRoche is absolutely worth that, even if the Sox sign Napoli.
This is a general question: Are teams able to offer the qualifying offer (offer an offer? whatever) to the same player in subsequent years? For example, let’s say Rafael Soriano had accepted the 1/13 qualifying offer from the Yankees this offseason. Would the Yankees then be able to extend him the qualifying offer next offseason as well?
I’m pretty sure that the slight blip at about pick 450 is Albert Pujols. Wasn’t he a 400-something pick in the draft? That’ll bring up the average.
One thing I think Lohse misses in his quote is that the Greinke did NOT have a ‘get-out-of-jail-free’ card: in order to acquire Greinke’s services, the Angels had to give up:
– A shortstop who will start in the majors this year (Jean Segura),
– Two hard throwing righties who have a decent shot at making a major league roster at some point in their careers (Johnny Hellweg and Ariel Pena)
I would say that the Angels gave away much more mid-season via trade than they got back for him mid-season via the free agent rules — at last one player with almost 100% definite WAR value, versus a complete crap-shoot first-round pick.
The draft rules exist to prevent cash-heavy teams from utilizing no-strings-attached free agency to scoop up all good FA players. There needs to be an equalizing cost for this kind of behavior. There is in the regular season (you have to trade away players), so for the sake of parity, the same should be applied in the offseason to some extent.
Perhaps Lohse should just not kid himself and accept the fact that he is worth less than Greinke and Sanchez 🙂
Sorry, I would say that the Angels gave away much more mid-season via trade than they got back for him OFF-season via the free agent rules
I also resent his insinuation that Milwaukee is a jail! Spoken like a true Cardinal 🙂
I like the article and I agree that the market may be somewhat overacting to the new draft rules. However, I don’t think it is accurate to say that the “reward for having that kind of pick hasn’t changed significantly” since the chart that shows the value or production of players by draft position applies to years in which the draft rules were the old draft rules, meaning, years in which teams could simply find late 1st round talent falling to the 2nd – 10th rounds because of signability issues. Under the new draft system, you simply will not see players signing for bonuses in the $2-5mil range past the 1st round (as it has happened in the past) and it is very likely that more talent will either rise to the first round (very few highly regarded HS prospects will sign for less than $1.5 – 2mil) or will wait until they can move to the first round in future drafts.
Kinda seems like a way for teams to be able to afford to keep their star future FAs.
I understand about a team losing their pick for signing one type-A FA. But are there any additional penalties for signing multiple type-A FA in the same offseason?