Jays Overly Aggressive with Arb Filing Strategy and Policy?
by MjwW on Jan 17, 2012 10:56 PM EST
Email
Print
Since we now have the figures that teams and arbitration eligible players have exchanged, it appears to me that the differences between the Jays and their two players are almost insignificant, and hardly seem worthy of going through a hearing (especially for Janssen, who will be a free agent after 2012 so it doesn't impact future hearings). Of course, it's quite possible either or both sides submitted different figures than what they were negotiating with as final offers, and that's why the gap is so small; but nevertheless, the Jays "file and trial" policy would seem to dictate that either they work out some sort of multiyear deal (which may be part of the hang-up), or end up going to a hearing.
I wondered if the Jays, relative to the rest of the filings, had smaller differences between player and team filings. This may reflect an overly aggressive policy (and perhaps needlessly so, given that arbitration hearings are adversial in nature and can engender hard feelings and resentments). Also, we saw a similarly small difference with Jason Frasor last year. I used MLBTR's Arb Tracker to compile a list of all filings from this year, eliminating the few cases where the team and player have already agreed after filing this afternoon. Below, I have then sorted by absolute difference (player filing less team filing) and relative difference (player filing divided by team filing):
Looking over, it appears that both the relative and absolute gaps are pretty small. Again, I don't want to make any inferences about the team strategy from a couple of cases, but there does seem to be a bit of pattern between these cases and Jason Frasor. I do feel comfortable saying that the Jays post-filing policy as currently articuled seems overly absolute and ill-suited for such a scenario (of course, it's possible that much like the "Blue Jays employees are always free to seek other opportunities" policy, there's some logical, unstated nuances).
There's one other question I had after looking at this. It would intuitively seem that the differences are more important the more arbitration eligible years remain for the player before free agency, since the outcome will be magnified by future increases though the arb process. For example, the difference between $2-million and $2.5 million might not be important if the player is going to be a free agent and have his future salaries determined by the free market, so team and player might just split the difference to avoid a hearing. But if it's the basis for future increases, both sides might dig in (for example, using the 40/60/80 rule, a player gets an increase of 50% the second time (60/40) and 33% the last time (80/60), so the final salary would be $5.3M vs. $6.67M, and the total difference would be over $2-million. This might play out through the filings. To check this, I plotted arb years remaining (omitting FAs who agreed to arb and Super 2s who have 4 total years) against the relative difference:
There is a resonably strong trend indicating that the teams and players have larger differences the more arb years are left. This helps explains why there's a gap with Janssen - in fact, knowing nothing else, we'd expect about a 23% difference for Janssen. But, it doesn't help explain Morrow.
Finally, I've plotted the relative difference against the midpoint of the filings to see whether there's a pattern in how large the stakes are in dollar terms that could explain why the differences for the two Jays are so small:
There's really no trend here to help us explain anything regarding the filing differences. So we're left with little to explain the Jays' approach to arb negotiations that result in filings, and again too small a sample in my opinion from which to draw inferences. The difference with Janssen seems more within the mainstream and explainable, the difference with Morrow seems rather insignificant - hopefully they're working on an extension or something.
27 comments
|
0 recs |
Do you like this story?
Comments
The more arb years, the bigger the difference
That makes sense given the unpredictability of a player that only has a few years of experience and the fact he is coming from a lower base (most likely closer to the MLB minimum). It’s much easier to project a player’s value the more years he has played and the more comps you have. Is Morrow worth $4.2MM? Well if you look at his numbers (3.53 xFIP, 3.4 WAR) last year, it’s still a bargain compared to the alternatives out there. Gavin Floyd’s 2009-2010 are somewhat similar to Morrow’s 2010-2011. Floyd made $5MM in 2011. Like you said I’m hopeful the Jays can work out a long-term deal with Morrow that brings cost certainty and security while taking care of a year or two of free agency.
The problem is
Morrow has underperformed his perips for the 2nd year running and even though all the advanced stats show Morrow as a great pitcher with great stats, that hasn’t transformed to a low era. In fact, it’s actually at a high 4.72. Obviously, we, as baseball bloggers pay less attention to era and understand that a lot of luck plays into having a high era or low era, the arbitration judges are not hardcore baseball fans and they just see the numbers. They see a 11-11 record with a 4.72 era and probably don’t see that great of a pitcher. From their standpoint and a casual baseball player’s perspective, he’s a back-end starter and isn’t worth 4.2 mil. The club will most likely win an arb case but could use those assumptions to secure a multi-year deal with Morrow while his era is still way over his fip, xfip and sierra numbers.
by Qd6 on Jan 18, 2012 5:24 AM EST up reply actions 2 recs
A-squared's comments about not signing Beede
When asked about his philosophy about negotiating with high draft picks, AA said the following:
The GM reiterated that his people do not panic in negotiations. They set a value on each player. They stick to it. When they ignore signability, they do not expect to sign every pick.
"Once emotions come into play, bad decisions are made," Anthopoulos said shortly after Monday’s signing deadline. "We just have to set values and talk about it in a rational state. We talk about it well ahead of time. We don’t deviate from those amounts."
I assume you're making a parallel
And I understand it, but there should also be room for a little reason.
I’ll be the first person to say, you need to objectively determine value, and stick to it, but also, there’s room within that framework for a little compromise. In this case, let’s say you think he’s worth $3.9M. You’re not going to non tender him if he wins the arb hearing at $4.2M, so is the $0.3M really a big deal? If you split the difference at $4.05M, it’s a fiference of 3% – that’s immaterial, and you have to consider the costs of a hearing (time, energy, and potential hard feelings). Not worth it all.
Similarly with Beede, their final offer was $2.4-2.5 million allegedly and Beede wanted matierally more, so that’s fine – walking away was the rght decision. But if it would have taken $2.6M-$2.7M, would they are settled there – I think so.
In Negotiating
One of the first things you learn, is that splitting the difference is actually a very ineffective way to negotiate. Also, I would assume that that is what AA does. He probably sets a number that he likes, and probably sets a number that is a hard cap. After the hard cap number, you’ve already said to yourself, it doesn’t make sense.
Also, in hindsight of the Beede negotiations. How much do you think he’s kicking himself for the way the new CBA turned out. He’ll get much less than he would have this year from us.
@VagabondBansal
Well
First, you’re talking about a general strategy of negotiating, at the outset of negotiations, and I agree, you don’t negotiate to split the difference (though, when both sides have established reasonable “goalposts” – intial positions – it’s remarkable how many times it ends up happening as you make trade-offs).
But I’m talking about splitting the difference in the context of a particular scenario, wit a particular fact pattern, such that sticking to the rigid numbers makes no sense for either side. To wit, we know the folliwing:
1) If Morrow were awarded $4.2M rather than $2.9M, the Jays still want him on the team in 2012 (paying less is an obvious preference, but it’s still objectively good value)
2) If Morrow were awarded $3.9M rather than $4.2M, he’s still going to play in 2012 (being paid more is an obvious preference, but it’s still better than any other altnerative)
3) The amount in question is fairly small, $0.3M difference is only 8% of the $3.9M base
4) The difference has no material impact on the Jays’ payroll next year
5) The difference is unlikely to materially affect Morrow’s 2013 salary if it went to arb, and very unlikely to affect any salaries beyond that
6) There is significant time and financial expense to prepare for, and hold the hearing
7) Arb hearings are adversial by nature, and potentially antagonistic, and can engender hard feelings.
GIven all the above, it this case (at the stage it’s at), it makes all the sense in the world for both sides to split the difference if they can’t agree on a long-term extension or whatever else may be going behind the scenes. It’s just not worth going to a hearing.
by MjwW on Jan 19, 2012 1:08 PM EST up reply actions 1 recs
I just saw this filing as further evidence an extension was going to happen
I certainly don’t think AA wants to go to arbies (who does, right?) so now the pressure is on to do something I’m sure he wanted to do in the first place. Morrow’s a good pitcher and controllable, I see him as a perfect candidate for an extension.
The flip side though
Is that Morrow is two years away from being a free agent…and he could really cash in. Think AJ Burnett – great stuff, hasn’t completely put it together. Burnett got $55M in 2005, Morrow (if healthy), even if he doesn’t break out, could get 75M in two years. It’s possible he wants to wait and go fort he big cash in
If that's true
Then I can definitely see him being traded this year or next. Morrow will only look better in terms of stats, driving his asking price up. So, if he’s already asking for more than AA has set his value at, then I don’t see the two sides ever meeting (money-wise). If that is indeed the case, AA will trade Morrow for more young, controllable talent before he loses him to free agency.
Yep
I’v said a few times before, it’s my opinion that either Morrow signs an extension into his FA years, or (barring major injuries that wreck his value), he’s probably not on the 2013 roster. Especially since an acquiring team would want a full season, and then they could get compensation if he left.
The only way I could see this not happen is if Morrow had a massive season (5+ fWAR and rWAR) and the Jays were close to contention. Then they may not be in a position to trade him, and would look to re-sign him as a FA or take the comp picks.
Please correct me if I'm wrong here
Take a guy like Kershaw. If he wins and is awarded 10M, but the Dodgers don’t believe he is worth that (silly), they are not under obligation to sign him and hence becomes a FA?
They would have to pay 1/6 the arbitration award
but yes, that is the case (though I am not sure if another team that signs him would still have controllable years at that point).
by Playoffs!!!!1 on Jan 19, 2012 9:36 AM EST up reply actions
So a guy like Lincecum, if he was let go, would still be owed 3.5M by giants in the award
But since difference was essentially that anyway between what was offered and awarded, it makes sense for them to resign at 21.5M. It makes sense to me now. Thanks. And whether those arb years carry over would be interesting to find out when another team signs him. And would signing team have to pay 21.5M and continue arb years or just sign a fresh FA contract? If anyone knows.
Theoretically
Yes, the signing team would have controllable years…because he would not have the 6 years service time required to subsequently file as a FA (if he signed a 1 or 2 year contract).
An example of this is Ryan Vogelsong, signed as FA last offseason to a 1 year deal, the deal is over, but he’s still arb eligible because he doesn’t have the 6 years service necessary to file for free agency havign been tendered a contract for 2012.
In practical terms, it wouldn’t matter because Kershaw would be a FA and would sign a long-term contract that would take him well past the three arb years such that when it expired, he would be a free agent.
David Ortiz
I find it curious that the Red Sox only offered $12.65MM compared to his $16.50.MM. He made $12.5MM last year and had a pretty awesome season last year. I would think the BoSox would go for something a little higher so its more possible. I would think after the season Ortiz had, its almost impossible for them to win by offering a 1.2% raise.
@VagabondBansal
I think it's about AA looking for a negotiating advantage
My thinking is that the figures submitted for arbitration are probably closer than the last figures exchanged in negotiations. I also think AA would prefer not going to arbitration, but if it gets to that, he doesn’t mind. What AA has done is create an artificial signing deadline to encourage players to sign on his tems. If players sign by this deadline, they get a one-year guaranteed contract. If they don’t, it’s a choice between one-year plus a club option or a non-guaranteed contract. Both of those choices are inferior to a one-year guaranteed contract from the viewpoint of the player.
To every complex problem, there is an answer that is simple, easy to understand and wrong
nice work, but
you need to be careful with those trendlines. just because there is a positive slope to the best fit line doesn’t mean that you’re necessarily seeing anything. Before reading too much into it, you should test the hypotheses to verify that we aren’t just seeing random noise as a result of a relatively small dataset.
"Look at me! I'm Tomokazu Ohka of the Montreal Expos!"
buried under work of late
the site is kind of a positive feedback loop — the more I am posting here, the more time I spend here. for the time being, it’s simpler to lurk and just post every so often
"Look at me! I'm Tomokazu Ohka of the Montreal Expos!"
Right
I actually did test them (not in individual models, but as variables in a model to predict the arb difference). I don’t recall the numbers, exactly, because the t-stat for to number of years reminaing was about 1.3-1.4 and the other was extrememly insignificant.
Those charts and lines of best fit were more for illustration than anything else, given the fairly low R^2 values – that’s why I made sure to put them there so that could be seen. Perhaps my characterization of the first as a “reasonably strong trend” was overly strong, but one factor that explains 12% of the variation (despite being able only to slot into 1 of 3 values), is worthy of mention, even if we might reject it in a significance test.
Thanks for the feedback.
yeah, it's kind of a weird situation here
There’s so much variance I’d almost think you should use a non-parametric test. Certainly, reading too much into the two data points with zero years remaining may be a mistake — particularly considering that the article also posits that one of those datapoints (Janssen) may be indicative of an (overly)aggressive team-wide policy that would reduce the relative difference between the two awards.
Have you thought about testing whether position (particularly, starting pitchers) would have larger differences in their awards?
"Look at me! I'm Tomokazu Ohka of the Montreal Expos!"
Yeah
To be honest, I didn’t spend a ton of time looking for trends. This just kinda evolved out of a benign thought that the differences between the Jays and their players seemed pretty small, and when I saw the MLBTR reporting on others it seemed to be the case, and I wondered if the policy/team strategy could be responisble. The data was easy to gather so I just wanted to confirm, and once I had, it occurred to me that the players with more years left seemed to have larger differences, which (to me) could intuitively be explained. I never really intended for this to be a rigourous research project, I just thought the question/data might be interesting to others. As you say (and I tried to make clear in what I wrote), a lot of caution is on order before drawing any conclusions.
As to the point about a non-parametric test, that’s unfortunately not an area I’m particularly well-versed in, and I’m just using Excel here so not sure exactly what would be possible. For what it’s worth, I ran a quick two factor model using arb years left (X1) and midpoint filing (X2) (idea being that larger filings might cause lower % differences – that is, larger abosolute differences but smaller relative differences). I got Y = 0.29 + 0.085X1 – 0.007X2. T-stats of 3.20, 1.62, -0.91, respectively. So the arb years left comes close to significance, though as you say OLS might not be best here.
I just quickly seperated out the 8 guys who were primarily starters up to this point in their careers (and player/team difference, ):
Clayton Kershaw, 53.8
Brad Bergesen, 50%
Jeremy Guthrie, 41.4%
Shaun Marcum, 28.9%
Matt Garza, 28.6%
Tim Lincecum, 26.5%
Jeff Niemann, 18.2%
Brandon Morrow, 7.7%
This gives a simple average of 31.9% difference and weighted average difference of 31.0%. Which is really close to the overall difference. Without running any tests, considering the small nature of the sample I’d be comfortable saying there’s no difference, at least in this data set.
The obvious answer would be to expand the dataset back a couple years, but frankly, as I alluded to above, it’s not really something that I think is worth looking into that seriously (especially since it won’t really help answer the initial question, as the AA Hays would still have too few data points from which to draw an inference).
interesting
i don’t know if you can do rank-order (nonparametric) testing in excel, though I’d suspect that you likely can with the statistical analysis toolpack. You might have needed to do what you’ve done so far (I’m not sure what excel can and can’t do), but if you don’t have it and you’re interested in stats, you should activate it (it is free and easy to do).
Thanks for taking the time to run those numbers on the pitchers. One thing that I think might be interesting to take a look at (though, unfortunately the samples would be incredibly small even going back several years) is players who choose to accept arbitration — incidentally, I don’t think you included that in your initial model (there doesn’t seem to be a datapoint for David Ortiz).
"Look at me! I'm Tomokazu Ohka of the Montreal Expos!"
Ortiz is in the initial list. I left Ortiz and other Super 2s who were first time eligible out of chart on arb year, as well as the regression above (it’s actually mentioned in my write-up right above the right chart). When I did this the other night, I know I had a reason for doing this that seemed right to me (obviously), but I can’t quite recall exactly what it was (so maybe it wasn’t so good).
Before I go further I should mention one thing I omitted in my previous response – the regression I ran only had r^2 = 0.15 and adjusted-R^2 = 0.085. So basically, it didn’t help us predict the relative difference of an inidvidual player much at all, compared to just an average of all relative differences. In response to your point above about Ortiz, I decided to just toss Ortiz and the 4 Super 2s in and see what happened. I got this:
So basically it just looks random, and now we’ve essentially got two individual variables that don’t tell us anything. I reruna the regression, as expected, you get R^2 = 0.03, adjusted R^2 = -0.02, and insignificant t-stats on the variables.
As for the rank-order thing, I’ve got Excel’s statistical tool pack (would be a nightmare to run regressions without it), I don’t see anything rank-order test. There’s something to do with rank-percentile, which sounds like it could basically be the same thing, but as I said, I’m don’t have experience working this type of testing (at least, that I recall) so I’d have to look into it more before I’d feel comfortable putting results out. I’m working on a few other baseball research projects right now when I have time, but I’ll keep this in mind
Oh, screw MLBTR
They decide to include servie time in their arb tracker today, after I had to gather that for each player indiviually through B-R the other night
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Bluebird Banter
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
Baseball Nation Recent Stories
In Association With
Big League Stew - MLB - Yahoo! Sports
- Giancarlo Stanton misses cutoff man, and Logan Morrison is thankful he did (Video)
- Snoop Dogg delivers memorable first pitch at U.S. Cellular Field
- The Juice: Mesoraco’s grand slam helps Reds complete four-game sweep of Braves
- Please hold: Indians’ Joe Smith jumps the gun on relief appearance
- Local radio host honors Cubs’ Tony Campana with parody song
