Contextualizing The Qualifying Offer System

At base, the qualifying offer system provides a mechanism that allows teams to allocate resources elsewhere while still obtaining the services of certain desirable, established ballplayers. Those players, in turn, sacrifice a portion of the contractual guarantees they would otherwise obtain in an unqualified market. In effect, the system taxes the prospective earnings of certain players who are (or soon will be) free agents.

For more background on the function of the qualifying offer system, see Avoiding The Qualifying Offer, by MLBTR's Tim Dierkes.

The impact of the qualifying offer system has come under much scrutiny over its first two offseasons of implementation. Some have argued that it unfairly penalizes the above-average but non-superstar players that are made a QO (which, if declined, requires another club signing that player to give up its best non-protected draft pick and the accompanying bonus pool money). Others claim that it allows such players a fair chance to sign a substantial contract, pointing to the offer's value (last year, $13.3MM; this year, $14.1MM). A range of arguments also claims that the system perversely favors larger-market clubs.

But before considering such criticisms (and potential reformulations of the system), it is worthwhile to put the system in its broader context, and to consider carefully how it serves (or disserves) its various actual or potential purposes. While it seems exceedingly unlikely that changes will be made before the current CBA is expired and replaced in December of 2016, the topic deserves consideration and debate leading up to that point. In approaching the issue, it is worth looking carefully at both the money at play and where the various risks, benefits, and incentives fall.

A. Overall Market Impact

How much money — in both real and baseball terms — is at stake here? Dave Cameron of Fangraphs posited recently that a 3x valuation of a draft pick's slot value is a good approximation for that pick's value. In other words, a team considering gaining or losing a draft choice would factor that amount in when assessing the potential impact of signing a player (or allowing that player to sign elsewhere). For the 2013 amateur draft, slot values rose by 8.2%, according to Jim Callis of Baseball America. I will assume both a 3x slot value in reaching a market rate, as well as a like 8.2% increase for the upcoming 2014 draft. (Note: because slots shift with every move impacting the draft, the resulting numbers will not be perfectly precise, but should nevertheless easily qualify as accurate for our purposes.)

In 2012-13, six players declined qualifying offers and changed teams. According to River Ave. Blues, teams sacrificed the 17th, 22nd, 28th, 29th, 42nd, and 70th picks, while other clubs picked up the 28th through 33rd selections. 

In 2013-14, to date, nine players have declined qualifying offers and changed teams. (Two QO-declining players have yet to sign.) Again, according to River Ave. Blues, teams have sacrificed the 17th, 18th, 21st, 26th, 44th, 48th, and 54th choices, while other teams gained the 28th through 34th pick. In addition, the Yankees both earned and sacrificed two supplemental first round choices. Because they finished with the worst record among clubs to have earned a supplemental pick, they would have stood to gain the first two of those picks.

Add up the slot values, and apply the 3x multiplier, and these are the results: In 2012-13, teams sacrificed a total value of $27.51MM ($4.585MM per player) and gained $29.33MM ($4.889MM per player). In 2013-14, to date, teams have sacrified a total value of $48.90MM ($5.433MM per player) and gained $48.85MM ($5.428MM per player). The 8.2% assumed increase in draft slot bonuses fairly well accounts for the rise in per-player figures, with the increase in overall impact coming otherwise from the increased use of qualifying offers.

QO pick value

It is worth putting these numbers up against the overall free agent spend to gauge their magnitude. I recently broke down free agent spending trends, including numbers for the 2012-13 period and the 2013-14 period to date. In the 2012-13 free agent period, clubs committed just over $1.463B in total via free agency, while this year's spending has now topped $2B (it was at $1.88B as of that post). Thus, the net draft value transferred through the qualifying offer system has been around 2% to 2.5% of the total open-market spend over the last two years. 

That, surely, is a relatively minimal savings for MLB teams on the whole. When considering the full gamut of ways in which teams invest money into players — including free agency, extensions, the draft, and international signings — the money saved (i.e., allowed to be re-allocated) makes up a meager portion of the aggregate. Of course, it is worth wondering the extent to which the prospect of a future qualifying offer also transfers savings and leverage to teams negotiating extensions with players close to hitting free agency. This is impossible to calculate, but nevertheless does also transfer value to individual teams and savings to the league as a whole, as against current MLB players.

B. Impact On Individual Teams

On an individualized basis, teams assessing whether to make a qualifying offer, or whether to sign a player bound by draft compensation, have ample flexibility to value the choices at stake and decide whether and how to factor them into their decision. 

Generally, a team making a qualifying offer to a player obviously reaps a substantial, albeit variable, benefit (in addition to having the right to choose whether or not to make the offer). It may ultimately retain that player at a one-year rate which, in theory, is at or below that player's market value. Or, it may instead receive a supplemental draft choice that lands between the first and second rounds of the draft. (It has been suggested that a team can sign a player that declined a QO at a lower rate than can other clubs, but that is not entirely true: once the offer has been declined, a re-signing effectively entails the sacrifice of a draft choice that, as explained below, comes with roughly the same value as the choices given up by other teams.)

A team signing a player that declined a qualifying offer, on the other hand, theoretically ends up in a neutral position, because it can discount its offer by the value it places on the pick that it sacrifices. While draft picks move as they are added and subtracted, teams have a very clear idea of what they are giving up in future value, and can apply their own valuations (including the tradeoff of present and future, and the actual slate of available draftees) to reach a discount.

As noted above, teams that are considering extensions of current players — especially, those close to reaching free agency — can also utilize the threat of a QO (implicit or otherwise) to achieve leverage. That may mean little in negotiations with superstars, but could have a sizeable impact on talks with "merely" above-average players. The possibility of a qualifying offer adds to and even enhances the risks of injury or performance decline already present to an above-average free agent-to-be, which could result in savings to current teams. This factor is, perhaps, enhanced in the situations of older or defensively-limited players who could be looking at short-term contracts after their walk years, even if they put up good numbers.

C. Impact On Players Subject (Or Potentially Subject) To Qualifying Offer

Meanwhile, for players, those issued a QO are subject to the whims of their team and other circumstances outside of their control. The take-it-or-leave-it offer is presented at the team's initiative, before the market even opens. (This year, for instance, Ubaldo Jimenez came with draft compensation; Matt Garza and A.J. Burnett did not.)

Further, the cost of the small reduction in overall spending falls squarely on the shoulders of the small number of players who are made a qualifying offer: their market value is decreased by whatever amount prospective teams choose to discount for the lost pick. And that burden is not shared by any other stake-holders (other players, the league, teams).

One line of sentiment says that some players over-value themselves by rejecting the QO. (No player has yet accepted one.) Even if the offer is an arguably fair price for one year of a certain player's services, however, that does not end the matter. It is quite difficult to reach free agency through service time, let alone to do so at an optimal point in time. Players that do naturally seek to maximize their overall guarantee, often choosing a longer contract with a lower average annual value to avoid the risks inherent in suiting up for another season before securing a future commitment. As Ubaldo Jimenez recently explained it: "I knew that I would have a better offer than the qualifying offer. Or at the very least, I wasn't as much worried about the annual salary, I was more concerned with having the long-term security."

For those that decline the qualifying offer and test the market, the resulting tie to draft compensation can be a significant drain. Though it is often suggested that the downside of the QO system lands solely on the more marginal players that receive an offer, that may be true only in relative terms. We do not know whether, say, Robinson Cano would have garnered a $245MM (rather than a $240MM) guarantee from the Mariners without an offer. But it is likely that the impact of a lost pick was factored in on some level.  While the effect is a relative drop in the bucket for top-dollar free agents, it appears nonetheless.

(This is not to suggest that the team would have any focus on the loss of the draft pick in negotiating with a star-level player; rather, the value of the lost pick would be baked into the valuations that the team uses in setting internal parameters before entering negotiations. As Blue Jays GM Alex Anthopoulos has explained, for example, "there's still value" even for picks after the first round, "and you still build that into an offer." Of course, teams may still be willing to go past those valuations to land premier free agents, and/or effectively adjust their internal valuation of a pick downward based on their present roster construction and expected point on the win curve.)

More importantly, though, it is clear that non-superstars take the largest proportional impact. That is made obvious by Nelson Cruz, who recently became the first QO-declining player to sign a deal worth less than the qualifying offer itself, with his one-year, $8MM contract falling over $6MM shy of the $14.1MM he could have taken before the start of free agency. (Previously, the least guaranteed money in a deal to a player who declined a qualifying offer was Adam LaRoche, who signed for two years and $24MM last year after declining the one-year, $13.3MM QO.)  

Indeed, the greatest risk for a player may lie not in the possibility that they will see a reduction in their earnings to compensate for the value of a lost pick, but in the impact on that player's overall market. Some teams may decide that they will not sacrifice their draft pick in a given year at all, for any number of reasons. Others may decide that, whatever the theoretical value of the pick, organizational needs dictate that they not cede the choice unless they are signing a player of a certain, higher order. Some clubs may simply put a greater internal value on the draft choice than that suggested by Cameron.

Whatever the case, teams clearly do not take lost picks lightly. "You hear people say, 'Well, what if the [drafted] player doesn't make it,'" Brewers GM Dough Melvin has said"That's not the sole purpose of a draft pick. You can use those picks for trades. … I'm glad we have Kyle [Lohse], but don't tell me that about overrated draft picks. Their asset value is huge." The net effect, in all likelihood, is often to take away some teams that would be potential suitors and, perhaps, to limit the willingness of some others to offer multiple years.

This could explain, in part, how a player like Ervin Santana ended up taking a one-year pact (at the precise value of the qualifying offer) just to get in camp before the end of the spring. This kind of situation is not unprecedented even without draft pick implications — for instance, Edwin Jackson signed a $11MM pillow contract with the Nationals the year before the QO system went into effect, and did not cost the Nats a pick (he was a Type B free agent under the old system). Like Jackson, Santana had multi-year offers, apparently at about the price achieved last year by Kyle Lohse. But Santana was unwilling to sign for three years at a total value roughly equivalent to two years of qualifying offers, as was Lohse, and was apparently unable to translate relatively strong demand into a fourth year.

The point here is not that the QO in and of itself prevented Santana from achieving the four or five-year deal he was hypothetically entitled to. Rather, it is that, for the reasons just noted above, the qualifying offer played a role in dampening his market and reducing his leverage to pry away more years at a strong average annual value. And that, in turn, makes for an even greater potential increase in the proportion of the burden borne by a limited class of players. 

Then, there are the soon-to-be free agents facing the possibility of a market hampered by the burden of draft compensation. As noted above, their leverage is surely reduced by the prospect of carrying that added cost, especially if always-shifting demand turns out to be less than robust. And the qualifying offer introduces risk to the player through the unpredictability of its effects, especially since the system drastically skews the market (through the protection of some draft picks and the fact that teams signing multiple compensation free agents sacrifice increasingly less valuable choices.) Indeed, players like Justin Masterson and Chase Headley have reportedly seen their current teams insert the qualifying offer into extension talks.

D. Conclusion

With this understanding of the broad parameters of the function of the qualifying offer system, the question becomes one of purposes. What does the system hope to accomplish, and hope to avoid? And how well has it done? I will suggest an approach to those questions in a second post.


Full Story | 93 Comments | Categories: MLBTR Originals

Leave a Reply

93 Comments on "Contextualizing The Qualifying Offer System"


LazerTown
1 year 5 months ago

Morales and Cruz would not have gotten anywhere near $14M aav though without the offer. If you take a pretty optimistic WAR projection for Morales and stick $6M per WAR, it pops out $9MM salary, which I think is a lot closer to what I think he is worth on the market. Cruz is maybe worth $11MM, both on short term deals. They both overvalued their markets IMO. March can be a tough time to get a big contract, just because teams already have their teams set, and have convinced themselves enough that things will work out. Red Sox have warmed up to the fact of Bogaerts, and DH is even harder for Morales, just because so many players can fill that position. At least Drew has position scarcity, Morales and Drew don’t, which makes the fact of their only average offense hard to get the contract.

Rally Weimaraner
1 year 5 months ago

Good piece! My biggest issue with the QO system is not that it harms players’ earnings, 14+ MM a year is a fine salary, but that it unfairly benefits large market teams. Large market teams like Yankees, Red Sox and Dodgers can afford to make QO’s to far more players than small market teams, for example the Pirates. The Pirate couldn’t risk spending 14 MM on 1 year of AJ Burnett but the Red Sox are perfectly happy to risk a player like Drew accepting a 1/14 deal.

daveineg
1 year 5 months ago

The non superstars have had inflated salaries relative to their production for far too long. The Pirates made a mistake not tendering a QO to Burnett. They could have afforded taking that chance on a one year deal because they could have traded Burnett halfway into the season.

Rally Weimaraner
1 year 5 months ago

As I stated above I have no issue with the QO system decreasing salaries for non-star players but to teams like Pittsburg or the Rays there is far more risk involved in offer a player a 1/14 deal than their is for larger market teams. Spending 14 MM a year on a player just doesn’t carry the same risk for large market teams as it is for smaller market teams thus allowing large market club the ability to offer QO to more FA and acquire more draft picks.

1 year 5 months ago

I don’t think Burnett is a fair example given his “Pirates or retire” sayings and the fact that the Yankees had been paying a bulk of his salary.

The Yankees and the Red Sox are benefiting from this right now because they’re the ones with big money players hitting the market. We’ll see more of the smaller market teams involved once their younger players who signed extensions early on hit the open market.

Rally Weimaraner
1 year 5 months ago

Fair point regarding AJ but take the case of Bartolo Colon and Hiroki Kuroda. Both put up very similar numbers, Colon 3.9 fWAR age 40 Kuroda 3.8 fWAR age 38. Kuroda was offer a QO by the Yankees, Colon was not offered one by the A’s. Not a perfect comparison but it does eliminate the “Pirates or Retire” element

1 year 5 months ago

Kuroda made 15 million last year. Colon made 6 after his incentives were met. So you’ve eliminated the “Pirates or Retire” element, but citing Colon’s numbers doesn’t take away from the A’s reasonable reluctance to give him an 8 million dollar raise.

Rally Weimaraner
1 year 5 months ago

That is precisely my point. NYY wouldn’t bat an eye at giving a player a substantial raise like that. NYY gave Kuroda a 5 MM dollar raise in 2013 no problem, Boston offered Drew 4.5 MM raise but to teams like the A’s and Pirates giving players a major raise is just not an option.

1 year 5 months ago

The Yankees did let Colon go once over what seems to be no problem to you.

The problem with Kuroda and Drew is that they were both wanted back by their teams. Did the A’s really want Colon back? I’m not sure the answer to that is yes.

Rally Weimaraner
1 year 5 months ago

If Boston wanted Drew back why is he still a FA?

1 year 5 months ago

I used the past tense for a reason. I don’t think they want him back anymore now that both Sizemore and Middlebrooks are exceeding expectations, but Drew had plenty of opportunities to go back that Boras likely botched.

Rally Weimaraner
1 year 5 months ago

Fair point regarding AJ but take the case of Bartolo Colon and Hiroki Kuroda. Both put up very similar numbers, Colon 3.9 fWAR age 40 Kuroda 3.8 fWAR age 38. Kuroda was offer a QO by the Yankees, Colon was not offered one by the A’s. Not a perfect comparison but it does eliminate the “Pirates or Retire” element

Croagnut
1 year 5 months ago

What makes Pittsburgh a “small market” and Boston a “large market”. Population: Pitts: 20th, Boston 21st, Pittsburgh has 7 Fortune 500, Boston has 2.

Pittsburgh isn’t hurting in football or hockey. Yet, you include Boston with NY and LA, the two largest cities in the country. How about Pittsburgh is a football town, and Boston is a baseball town.

Rally Weimaraner
1 year 5 months ago

Boston’s 2013 team payroll: 153 MM
Pittsburg 2013 team payroll: 66 MM

Team spending not population determines large v small market teams. Pittsburgh may be a rich city but the Pirates are not a rich MLB team.

Croagnut
1 year 5 months ago

That means LAs market has blown up from 95 mil to 221mil in 4 years. Don’t think so.

rct
1 year 5 months ago

I disagree with this, as it conflates cause and effect.

imo, revenue is what determines small market vs. large market, not payroll. Would you consider the Mets, who play in maybe the biggest market in all of sports, a small or even mid-market team? Their payroll is b/t $80-90MM, so by your standard they would be. Even as a fan of theirs, I’d say they’re large market, based on revenues/tv money/etc.

Pennsy
1 year 5 months ago

Boston metro area has a GDP two or three times larger than that of Pittsburgh’s. And many of the markets ahead of Boston by this measure (NYC, Chicago, LA, Bay Area) support two clubs. This is what makes Boston a “large-market club,” thanks for asking.

Croagnut
1 year 5 months ago

“3 times larger” no way. But, you make a legit point.

1 year 5 months ago

People who live in suburbs don’t watch baseball?
Boston is the 10th largest metro area, with a population of 4,591,112
Pittsburgh is 23rd, with 2,359,746. Boston’s market is almost double that of Pittsburgh.

pft2
1 year 5 months ago

Bostons market includes most of NE. Pittsburgh has a lot of competition nearby in geographical terms,

daveineg
1 year 5 months ago

I like the QO system. One of the biggest problems in baseball has been the overpayment to “non superstar” players once they hit free agency years. The reason this happens is that only the top 8-10 franchises can realistically bid in the open market for actual superstar players. This is a structural problem in a sport that unlike the NFL doesn’t equally share media revenue.

As a result the good, but “non superstar” players have had contracts inflated over the years by having more teams vying for their services than the actual stars of the game as many of these teams are completely shut out of the superstar market. That actually deepens the talent gap between haves and havenots because the fail rate for the lesser players is higher and low revenue teams get stuck with bad contracts more often.

It’s theoretically a better game when there aren’t players who no longer play to acceptable standards taking up roster spots solely because of the amounts remaining on their deals signed years earlier. In the NFL if a player can no longer play, he’s gone, presumably replaced by a player who can play. That makes for a better product on the field. Now baseball will likely never revert to non guaranteed deals, but the QO system at least attempts to reign in spending on guys who’s impact on the game was not all that strong to begin with.

Herb G
1 year 5 months ago

There is absolutely nothing wrong with the QO system that a simple tweak won’t fix. It is absolutely senseless to penalize a team for signing a free agent merely because he declined a QO. If a QO was not made to that same free agent for whatever reason, there would be no penalty. The signing team has nothing to do with whether the FA’s prior team does or does not make a QO. I can understand the logic of compensating the team losing the FA after they have made him a QO. That team sould receive a supplemental draft pick in a supplemental round between the first and second rounds. If they lose a second QO FA, they should get a pick in a second supplemental round between rounds 2 & 3, and so on. The picking sequence in the supplemental rounds should be the same as the original selection sequence. The signing team should, in no way, be penalized.

Croagnut
1 year 5 months ago

Your tweak would be a boon for big market team and big spending. The exact thing the QO system is trying to slow down. It would create a huge disparity for teams that can’t afford to offer QO’s or sign top FAs.

Big market teams would out bid for the best FAs and, also, have a stacked minor league system. The current QO system is trying to say “one or the other”.

Herb G
1 year 5 months ago

No! Actually, it is a boon to small market teams who desperately need the draft to continually build their teams. Big market teams (Yanks, Dodgers) spend like water under any system, make the QOs to their departing big ticket FAs, and thus often recoup the draft picks they lose in signing high priced FAs. They are indiferent to the draft, relying primarily on the FA market to build their rosters. It’s the small market teams that have to think long and hard as to whether to extend their budgets to sign an elite free agent at the cost of a precious draft pick.

You must realize that the big market teams almost always out bid smaller market teams under the current system, except when the smaller market team is willing to seriously overpay. (e.g. Cano to Seattle) Or are you a Yankee fan?

daveineg
1 year 5 months ago

A big problem in baseball are “dead money” contracts. The QO system at least is an attempt to reduce that impact. The Cubs being pitiful for 3-4 years and basically not even trying to compete for that long is not good for the game. The need to rebuild them from the ground up was essentially the result of them being buried by dead money for deals negotiated by the Hendry regime and while they are “big market” in a sense they are limited by owners heavily in debt, and a park that doesn’t generate much revenue.

Norm Chouinard
1 year 5 months ago

One thing not mentioned is the incentive for a player to remain with his current club. I don’t think that there is a usable metric to put a dollar value but I think there is long term revenue value there for the clubs to keep their fans watching games live or at home.

TheNextEpstein
1 year 5 months ago

I actually like this system, it forces more players to entertain extensions with their current teams rather than allowing a pure free market for the bigger spenders to spend on all the premium talent and inflate the market. This allows smaller market teams to better compete on the field and help keep their star players in town, at a fair value, to sell more tickets for longer. Stars spread throughout more teams is better for the game rather than having a few teams with all the stars.

Pennsy
1 year 5 months ago

It helps mid-market teams but the benefit for the smallest market teams is practically negligible. When they do get these quality talents and they do end up heading toward Free Agency, the teams at the bottom still need to entertain moving them in a trade for a greater prospect return than one draft pick. So this system is meant to reward a small-market team that develops a player and holds onto them until Free Agency- but mostly you get a team like New York or Boston that are extending offers to players they’ve previously acquired (or maintained control of) through FA. Boston got Napoli and Drew on one-year deals just last year.

It seems to me the system is actually much more beneficial to large-market teams that are picking up quality veterans on one and two-year contracts, as they can then attempt to either retain their services at a market discount or get draft pick compensation for their departure. It would be interesting to see the varying types of players that get Qualifying Offers, on the range from BJ Upton or Michael Bourn (organizational fixture) to Stephen Drew or Shin-Soo Choo (recent acquisition)

1 year 5 months ago

All teams need to entertain moving players heading toward free agency for a greater prospect return than one draft pick. The level of return isn’t up to them.

Pennsy
1 year 5 months ago

All teams make this calculation but it’s Tampa Bay, Pittsburgh and so on that most heavily rely on return from developed players for organizational health. For sake of keeping their clubs competitive and cost-effective they often need to get more than a supplemental round draft pick for a player they’ve developed. This is why everybody was so shocked when the Rays actually held onto Price.

1 year 5 months ago

Tampa Bay has frequently let players go in free agency to get picks.

B.J. Upton, Carl Crawford, Carlos Pena, and Rafael Soriano got them nice draft picks. They might have traded Kazmir, Shields, and Garza, but citing them to prove your point does not really work.

TheNextEpstein
1 year 5 months ago

Pittsburgh uses the draft and extension strategy. McCutchen is a great example of this. They haven’t had to extend anyone else yet, but will as more of their youngsters start reaching arbitration. The Pirates make money anyone saying they can’t afford to pay their players more is lying.

Samhors
1 year 5 months ago

I think after this years fiasco with players taking equal or less money as the QO, many players will realistically look at if they deserve more money. Like said in the article, many players over valuate themselves. I wonder if an agent will tell someone in the future that they probably won’t get more money on the open market.

Samhors
1 year 5 months ago

I think after this years fiasco with players taking equal or less money as the QO, many players will realistically look at if they deserve more money. Like said in the article, many players over valuate themselves. I wonder if an agent will tell someone in the future that they probably won’t get more money on the open market.

Aaron Johnson
1 year 5 months ago

Players still want, and deserve the opportunity to sign multi year deals.

Samhors
1 year 5 months ago

Sure, but they could still sign a multi year extension with their current club.

Aaron Johnson
1 year 5 months ago

Depends on which situation, and either way, forcing a player to narrow his market from 30 teams to 1 team gives the club a monopoly on the salary of that player.

Aaron Johnson
1 year 5 months ago

I don’t think the Red Sox were interested in signing Drew to a long-term deal, if not at a severely discounted price. They didn’t want to block Boegarts, so Drew should have a fair shot to test the open market. His pick compensation then killed his market. He was in a lose-lose unless he wanted to sign the QO. But then again, doesn’t he deserve the opportunity to not play year-to-year?

Robert Eichhorn
1 year 5 months ago

The QO doesn’t create enough consequences for teams that offer it. They can essentially offer it, knowing that the player will decline and then not negotiate in good faith. Free agents want years of security. Teams need to be obligated to more than one year/14.1 million if they’re going to make that offer.

Aaron Johnson
1 year 5 months ago

Interesting idea….how about a tiered QO system based on contract length? For example:
1 year QO $14 mil- 3rd rnd pick compensation
2 year 27 mil- 2nd rnd pick compensation
3 year 40 mil-1st rnd pick compensation

Mario Rubio
1 year 4 months ago

There isn’t a risk to the offering team? The Red Sox were a simple “yes” away from making Stephen Drew the 5th highest paid SS in baseball. His career OPS+ is 98, and he managed a 110 last year, his first time above the league average since his age 27 season.

Drew is not an All-Star, he’s never lead the league in any category (or even really entered such a discussion) and he’s not likely to become a different player at age 31 than he has been his entire career. But we’re supposed to think he’s entitled not only to a $14M offer, but multiple years?

With League Average upside for Drew, why wouldn’t a team take a shot with a much younger SS with long term control.

Also, with so much talk about long-term security, we should also realize, had Drew accepted the Red Sox offer, his career earnings after the 2014 season would have neared $45M. This is not a guy who’s been making league minimum his entire career and is by no means hurting financially (at least has no reason to be).

Of the 13 players who were offered a Qualifying Offer, Only Cruz could make an argument that he had missed out on a opportunity previously not available to him. Many had signed extensions of some sort during their career or had used the arbitration system to find significant paydays. Others were fully capable of cashing in on their new opportunities and signed deals that would provide security to anyone.

Morales, Drew and Lohse are the poster children for the evils of this Qualifying Offer system, but the truth is, they are/were players with questions and the big payday wasn’t going to be there regardless. Morales is a guy who was never the same since breaking his leg and is essentially a DH only. Lohse already signed that big deal when he got 4 yrs and $40M from the Cardinals. At 34, he wasn’t going to get a 5 year deal, not for a guy who’s never been an All-Star nor really been a Cy Young candidate. Drew I already discussed earlier.

So who’s the Stephen Drew of 2014? JJ Hardy seems to be lined up for a similar off-season. The Orioles haven’t reached out to him for a long-term deal and Machado is going to play SS at some point. A multi-year deal is likely for defensively gifted SS with serious pop in his bat, but will the QO of more than $14M be too much to pass up considering he’s coming off a $7M 2014 salary? Are the Orioles prepared to pay him that money for one more year when their preference might be for him to simply go away. The Giants may be painted into a similar corner with Pablo Sandoval.

I think there’s plenty of risk to go around, Drew and Morales just made poor decisions and their agents should be the ones taking the brunt of the criticism.

goat
1 year 5 months ago

One change needed is teams who acquire a player part way through a season are not entitled to any compensation when/if that player walks at the end of the year. When you trade multiple good prospects for a player, and that player walks you deserve some compensation for the player leaving. You should be able to make a QO to a player and gain some draft compensation.

Pete
1 year 5 months ago

Jeff – what about this tweak? Expand the QO to be a choice – more money but shorter term (say, 1/14 as is the case this year) or less money and longer term (say, 2/10 per year)? Leave it to the player to decide whether he prefers more money, more security, or take his chances in the open market. In other words, the QO itself doesn’t have to be a single option.

Aaron Johnson
1 year 5 months ago

Jeff-It seems that the QO system has stratified the players that sign as free agents into their comparative team level. By that I mean, the middle tier QO guys (Santana, Cruz, Jiminez, Lohse) are going to middle market teams after the cream of the crop guys (Cano and Choo) end up going to the bigger market teams without much regard to the comp pick. All in all, the small market teams seem to be completely out of the equation. I don’t really know that I’m trying to say that there needs to be a system in play that “forces” small market teams to be involved in free agency, but it seems like the current system isn’t much different and may even be worse for parity than the old system. Do you agree?

pft2
1 year 5 months ago

Small market teams get revenue sharing dollars. Many of them spend far less than the MLB average of 47% payroll/revenue, as low as 15% in some extreme cases . Limit teams ability to collect revenue sharing dollars to those teams who spend 40% of revenue (including revenue sharing dollars) on payroll. So if the Marlins have a payroll commitment that would be only 20% of payroll with revenue sharing dollars, they can kiss that 30 million dollars in revenue sharing dollars goodbye.

Another idea is to take all the revenue sharing dollars and distribute it small market teams based on performance and not revenue. Give them an incentive to win because right now there does not seem to be much of an incentive. Teams are content to watch valuations and TV revenues go up and pocket revenue sharing dollars w/o much care of ticket revenues. They can get taxpayers to build them new stadiums and they have a legal monopoly where their market is protected from competition.

Mike1L
1 year 5 months ago

I’d like to suggest a alternate argument. I’d put an end to the loss of draft picks and slot money. I would argue that the more draft picks and slot money you take from the high revenue teams, the less likely it is that they will be able to draft and develop the cheap younger talent that allows them to not splurge in the FA market to fill every position. And the “protected pick” idea in some respects makes it worse, because each year there are high revenue teams who have a poor season, and end up with a protected pick that they can use as a bidding tool. I think we are focused on driving down the price of a handful of players, without acknowledging the impact on the rest.

pft2
1 year 5 months ago

Why should a player be taxed as a free agent after they have paid a hefty tax by playing 6 years at below market value already?

Nobody really argues with compensation, which is giving a team losing the player a compensation pick. Its the tax, which is loss of slot money and pick on the team that signs the player which is the problem,

Why this is a problem now and not as much when teams lost a pick signing a Type A free agent (except for lower paid RP’ers) is an interesting question. It could be teams have independently arrived at the same time on a higher valuation for the pick, or this could be slick attempt by a habitual offender at a limited form of collusion using the draft picks as a cover.

Clark should play the collusion card and force the issue before 2016. End the tax as the tax rate (determined by teams arbitrary and self serving valuation of picks) has become too punitive.

discollama
1 year 4 months ago

The guys hurt the most have been below average performers over the course of their careers. Perhaps the compensation is waking teams up to the fact that these guys don’t deserve 3+ year contracts or mega deals like they were looking for.

stemog
1 year 4 months ago

Why is this site set back to March 13? Weren’t the past 5 weeks real?

WrigleyTerror37
1 year 4 months ago

is this real life?

discollama
1 year 4 months ago

“As Ubaldo Jimenez recently explained it: “I knew that I would have a better offer than the qualifying offer. Or at the very least, I wasn’t as much worried about the annual salary, I was more concerned with having the long-term security.””

I love this quote, but mostly because Ubaldo’s WAR track record (3.76 fWAR per season) shows that he’s worth more than the QO in overall value. Guys like Drew (~1.8 fWAR per season), Cruz (1.65 fWAR per season), Morales (1.8 fWAR per season), and Santana (2.18 fWAR per season) are older, have injury concerns, or have had major inconsistencies in their pasts, these guys have historically played at less than league average (2.0 fWAR) over the course of their careers. So, paying them for one year of 2 fWAR production isn’t exactly unfair and generally speaking offers them a chance to prove that they are better than their histories have shown. I mean the only guy who was above average that was ‘hurt’ by the QO was Santana and he was only barely above average, just because in the past these types would get 3 year offers off the bat doesn’t meant that they should have been getting them, or that they are entitled, as you so eloquently put it, to such deals. The players may want security, but if they haven’t been able to prove that they’re capable of consistent above average performance, then why should they get it?