Updating the Prospects in the Paxton Trade
The headliner in the James Paxton deal is LHP Justus Sheffield. He’s been a top prospect for so long that it’s easy to find updated reports on him and understand where he falls in the prospects landscape. The short version is that he has an above-average-to-plus four-pitch mix, but his command ranges from average to below average, so he could still fit in a number of roles in the big leagues, ranging from multi-inning relief power arm to mid-rotation starter.
The more intriguing pieces of this deal are the other two names, RHP Erik Swanson and CF Dom Thompson-Williams. Neither were on the year-end version of THE BOARD, but both were on our radar; we intentionally didn’t comb through every 40 FV candidate in the in-season update since that’s what we focus on in the winter.
If we were doing the Mariners list today, both would be 40 FVs; they’re good examples of guys who sneak up on you during the season and in whom you have greater confidence moving up once the season ends. Swanson works 92-94 with a rising four-seamer, hitting 98 mph at times with some deception and life, and backs it up with a solid average slider and advanced feel for how to use both pitches in tandem. He could be something like a back-end starter who mostly uses two pitches, but he’s more likely to be the 5th-7th best starting option for a contender, and fits most comfortably as a David Phelps-type multi-inning fireman who can also do the job of long relief and spot-starting. There’s upside as a 50 FV here (4th starter or setup guy) but he’s more likely to be a 45 FV in the big leagues as a useful utility-type arm, so a 40+ or 40 FV would be appropriate.
Thompson-Williams is a sneaky athlete who’s a solid average runner with an average arm that some think can play a solid center field, but that most think is a fourth outfielder-type who can play all three spots. He has plus raw power and some feel to hit, so there’s low-end everyday upside if things continue to come together at the plate as they did in 2018. But he was 23 years old in High-A, so he’ll need to move quickly to be likely to reach that upside. More likely, Thompson-Williams is a useful bench option as a platoon at multiple spots or as a player who can provide some thump off the bench. Given his shorter track record and age, that’s a 40 FV for now with a chance to turn into a 45 FV with more performance, certainty, and proximity to the majors.
I’ve been asked a few times where these prospects fall on the dollar scale of our new prospect valuation metrics. Sheffield likely won’t rank exactly 54th on our next Top 100 in January, but the $29 million figure is about right. Swanson and Thompson-Williams combine for about $5 million more. Paxton is due in the $20-$25 million range for his next two years via arbitration while projected–using the same $9 million per WAR figure that generated the prospect values–to be worth somewhere around $60-$69 million in that span. So the Mariners receive around $35 million in prospect value, and send $35-$50 million of value back, depending on where Paxton ends up in his range. That’s within the margin for error, but is a bit lighter than expected for a Paxton package given the wide interest. That said, this trade appears to bring the Mariners out of the cellar of our first farm system rankings.
Kiley McDaniel has worked as an executive and scout, most recently for the Atlanta Braves, also for the New York Yankees, Baltimore Orioles and Pittsburgh Pirates. He's written for ESPN, Fox Sports and Baseball Prospectus. Follow him on twitter.
Thanks, Kiley. This helps put things into perspective, though it’s a bit maddening for Mariner fans.
The projection system on the site that employs you projects Paxton for 4 win in 2019, and will likely project him for right around 4 wins in 2020 as well. Last time I checked, 8×9=72. Subtract the $20M-$25M Paxton will be paid, and his surplus value is closer to $50M than the $35M of the prospect package.
Based upon the numbers available on this site, there is no way a writer for this site can logically conclude anything other than the Yankees got very good value by selling Sheffield to a team that valued him too highly.
That’s why this deal went down so early…the Yankees elite FO knew they just fleeced the morons in Seattle.
steamer projects 4.0 and he’s 30 right now, so 3.5 WAR in 2020 with a standard aging curve gets you to 7.5 over 2 seasons.
7.5*9 gets you to 60-something and I don’t think we have the precision in these calculations to be more specific than that.
Rounding 67.5 to 60 doesn’t make your calculations more accurate…
Given his injury history and the fact he’s only hit 4 WAR once… I don’t think it’s wrong to err on the side of caution. Listing a range is not wrong – eliminating likely outcomes is (like saying he’d be $60-$65 when median projection says $67.5).
His risk is also mitigated by the fact he is not guaranteed any money in 2020.
Well, Paxton is projected for 4.4 wins next year. So you’re looking at about 8.5-9 wins over two years (depending on how you think he’ll age). So that’s 75-80M in total value, minus $20-$25M. So that’s, what, approximately $55M in surplus value? Compared to $34M going back?
Edit: I see n_scheffel has made a similar point. My conclusion is similar…this is obviously a good deal for the Yankees (prospectively speaking).
Where are you getting 4.4? It’s at 4.0 here (3.8 for RA9, indicating either that Steamer thinks there’s a true talent gap, unless they’re just factoring in Andujar, haha).
That gives us 45 vs. 35, which is really just a function of their having a higher opinion of Sheffield than Eric and Kiley do.
Law had him at 12, Sickels at 32, Callis/Mayo at 31, which comes to an average of 35 and median of 32, when BA’s ranking of 47th is included.
Either number comes out to just over $40m on Craig’s chart. If you throw out Law, it’s still an average of 41st and a $40m valuation against a $46m surplus value, with the new valuation.
It’s dangerous to take a N of 1 on prospect rankings for valuation purposes – reasonable people can disagree on whether Sheffield is a 50 or a 55…and whether the new-look Paxton we’ve seen since 2016 has underlying issues causing his significant ERA-FIP gap, leaving his expected value closer to Astros’ Charlie Morton (a very good pitcher!) than Max Scherzer, on a rate basis.
I think people just got caught up in the returns for Sale and Quintana, who had been Top 15 pitchers by either WAR measure from 2014-2016, were two years younger, and offered 3 and 3.5 respective years of team control at a truly insane discount (seriously, look at Quintana’s contract).
This, it’s definitely close enough to be in the margin of error. Though it still seems light from recent, similar trades.
edit: spelling
I like this return better than the Cole trade, honestly…and Cole was younger, healthier and cheaper, with a longer track record of success.
For ceilings, I think I agree with you. This trade is better than the Cole trade. I think the Cole trade was better for floors.
I just both really dislike Musgrove’s game and never bought Moran’s power jump as a 24 year old repeating Fresno.
I also really like Swanson, and think he’s more likely to be Chad Green or Ivan Nova than Shane Greene.
I’d place all three as 45+ at the time of their respective trades, with Sheffield as a 50+.
So I don’t know where that 4.4 went. That’s what it was last night. I have noticed that after trades, the projections do change somewhat (happened with Yelich after his trade too). My guess is that it has a lot to do with how many homers he’s projected to give up in Yankee stadium vs. in Seattle.
Overall, I think that you need to do a ton of work to get them within the margin of error. IF Sheffield is truly a 55, and IF you don’t need to totally redo Craig’s rankings if you use different scales that are more generous with grading, and IF Paxton actually gets hit harder than his FIP would suggest, then sure, this looks like a reasonable trade for the Mariners to make. I admit I don’t agree with all (or any) of those things.
Using the numerical rankings rather than the FV, means you absolutely “don’t need to totally redo Craig’s rankings.”
Evaluators are pretty firmly split between valuing Sheffield as a low end 55 or as a high end 50 (with Law as an outlier).
Placement after 45 means BA and FG think he’s worth $30m. Placement between 20 and 45 means BP, Sickels and Mayo/Callis think he’s worth $40m. Law thinks he’s worth over $50m.
Averaging it out gets a $38m valuation ($36m without Law), plus whatever you want to label Swanson and Williams.
Anyway…if you think that Paxton’s issues with sequencing are entirely a fluke, and that he will set a career high in innings pitched at the age of 30, then Steamer’s FIP-WAR projection of 4.0 figures him for $46m in surplus value.
Frankly, the only way for it not to be within the margin of error is if you take BA and Eric/Kiley as sacrosanct and give no credence whatsoever to Jim Callis (one of the guys who made BA), Mayo, BP, Law and Sickels.
No, I don’t think you can totally divorce the measurement from the scale to which it is pegged. I just don’t. The whole point of Craig’s rankings (well, one of them) was that it was pegged the actual distribution of FV talents, as defined by Eric and Kiley.
I think, the main problem with I have with the present analysis, is this: If you’re willing to grant that there is some error around Paxton in one way, or around the value of Sheffield on the other, or anything else like that, you should be willing to grant that there is error in the other direction as well. Taking into account only the error that supports one’s priors is not something that I recommend.
I would also argue that with the level of error you’re arguing for here, the whole idea that we can peg value of trades with this framework sort of collapses. And I suspect that might be true, but it does make the whole point of this kind of meaningless.
But the distribution of FV talents as defined by Eric and Kiley is simply the industry standard based on standard deviations. Therefore any other Top 100 list should roughly map onto Craig’s valuations easily enough.
Especially since, y’know, Craig based his valuations off a different list.
My only argument here is that reasonable people can disagree on their evaluation of this trade – the existence of error bars proves my point as such.
And I would absolutely argue that an aggregated list should be used, rather than just FG’s or just BA’s or just BP’s, in order to gauge the market value of an individual prospect – the one guarantee we have about the lists of each of the 30 MLB teams is that none of them are identical.
If Cashman has Sheffield at a 50 and DiPoto thinks he’ll be a 60, then both sides are acting rationally, based on the info available to them.
Seems likely that as Jeff pointed to in his piece earlier today, the Mariners value Swanson noticeably higher than Kiley/Eric do based on his to-date performance in the minors. And while Paxton has avoided the major injuries, he has been injured often enough that he carries more risk than even a typical pitcher (already carrying high risk), so I suspect both of those elements lead to the discrepancy in valuations. That said, it feels as if the Mariners could have waited a bit, especially for Corbin to sign, and see if a team was desperate enough to offer more.
“[Swanson] more likely to be the 5th-7th best starting option for a contender,” which is why he’s the 2nd or 3rd best starting option for a team like Mariners. Likewise “Thompson-Williams is a useful bench option as a platoon at multiple spots or as a player who can provide some thump off the bench”… which makes him good enough to be a starter on the Mariners. They’re probably an 82 win team in 2019, which is all I really ask (since they’re .472 team over my lifetime).
I know age related to level is one of the key indicators of future big league success, but I wonder if the concept applies mostly to, say, the top 25% MLB producers. I’ve learned to read statements like “(Thompson-Williams will) need to move quickly to be likely to reach that upside” with a grain of salt these days. There are too many 25-28 year olds debuting and producing real value every season to make such statements with any type of certainty.
Yes…as a Mariner fan it’s tough to read this trade, by almost any measure, as anything other than underwhelming. And I like Sheffield. It’s tempting to think Dipoto and the rest of the scouting department found another Haniger…but the truth is that during Dipoto’s reign he’s only hit on these types of deals about 50% of the time. For every Haniger and Gonzales there is a Mallex Smith or Ryan Healy.
Unless there is some truth to the belief that the market is changing so much that anything other than a top line starter is not going to fetch top prospects in a return.
I agree with the general premise that Dipoto hasn’t had a huge amount of success but I’m inclined to believe bad luck potentially influences that more than anything else. For instance, if we assume injuries are more or less impossible to predict, then one or two bad breaks have had an out sized impact on his track record.
Drew Smyly and David Phelps returned essentially 0 WAR for the M’s due to injuries.
The thing is he wouldn’t have had to make so many risky trades to begin with if the Mariners had any idea how to develop their own players.
sheffield is overrated, swanson is really good as a depth piece though with his mix of improved stuff and control.
this entire framework of comparing WARs misses the point of this trade entirely though, in that Paxton is a premium weapon in the arms race for short series dominance. As long as teams have the payroll to absorb the risk, and yankees do, they just want the upside in a postseason series.
Mariners did not explore this arms race market at all and just sold Paxton for the first deal they liked.
I suspect Dipoto had other offers. It’s interesting to hear the perspective that GMs would have raised their offers later in the hot stove season. Any thoughts on why posters believe that to be true?
Not just this, but the Mariners may value more certainty when approaching the remainder of the offseason. In other words, if they wait for a Paxton trade and its uncertainty, how does that affect their decision making process in the beginning of the offseason?
uncertainty as in, big upside that his market would rise, plus small chance he gets hit by a meteor and nobody wants him? seems really difficult.
Sure, I suppose Paxton could get hurt in the 1H, but no one thinks this was an offer they couldn’t refuse. Why not wait for a better offer?
As it stands, Sheffield is not a Paxton replacement (he might be a bullpen piece), Swanson is a swingman, and the last guy is whatever. They needed to restock the farm system, and DiPoto just got undersold on his most valuable asset.
The only certainty this trade brings is that the Mariners will suck in the present as well as the future.
Dave Cameron and a lot of White Sox fans chided Rick Hahn for not taking a low offer on Jose Quintana in January 2017. Hahn waited 6 months and got a top-5 prospect in Jimenez and another top-100 prospect in Dylan Cease. Patience in negotiations has large rewards and especially for a guy like Paxton who Kiley McDaniel rated as the #48 highest trade value at this past ASB. (Quintana was #43 at the time of the trade with the Cubs.)
All it takes is a slightly optimistic prospect perspective (relative to the publicized talent evaluators) or a more pessimistic stance on Paxton and this turns into a win for the M’s. Not saying that will happen. Just pointing out the numbers are in the right ballpark. These monetary evaluations carry pretty significant error bars. People are treating them as if they sacrosanct, which they are not.
Paxton is a tantalizing pitcher. But when you hit 30 and the most innings pitched in a season is 160 there are some very obvious red flags. The M’s did well in selling now as the risk is just too high that Paxton turns into a pumpkin.
M’s fan here, James Paxton was my favorite Mariner, and he is an absolute stud. Scott Boras isn’t his agent for no reason, when healthy he’s a top 5 pitcher in the league, if not the sport. Having watched almost every pitch he’s ever thrown at the pro level, I don’t think he’s peaked either. He got a late start to pro ball, so the arm is extremely fresh relative to his age, and the injury history is full of yellow but no red flags. Paxton has legit 6 WAR upside, maybe more. However, as a flyball pitcher, I worry about him giving up home runs in New York, as good teams made hard contact last year when he wasn’t sharp with the fastball.
It wouldn’t surprise me at all to see him post 5+ WAR in 175+ innings, with a 3.50+ ERA.
Mariners trying to skip a step in the rebuild without the pieces necessary. I don’t see how this strategy gets them to a 90-win team in the next 4 years.
They need to be stockpiling teenagers now, not going for players who will accumulate service time immediately. Especially a fringe top 50 guy who has a very low chance to be a keystone piece when he hits arbitration and the Mariners might be competing again.