Volatility and The M’s Last Best Shot

marc w · January 3, 2017 at 5:34 pm · Filed Under Mariners 

The M’s won 86 games last year, and are once again forecast to be in the running for one of the AL Wild Card spots. They are also one of the AL’s older teams, and several players are now at an age when age-related decline comes quickly and decisively. This is the premise of this Thomas Jenkins piece at BTBS the other day, and it’s the focus of this great David Skiba post at LL a few weeks ago. It’s also something I wrote about in July. The ramifications are pretty big, I’d say – the M’s don’t have the farm system to make high-impact upgrades; they haven’t been players in the Chris Sale/Jose Quintana sweepstakes, and thus they struggle to close a deal at the trade deadline if they’re in contention. I like Tyler O’Neill, but the M’s don’t have a Lindor/Correa/Bryant/Betts-type about to hit the big leagues. They are within striking distance in 2017, but the downside risk is pretty huge.

This gets at an interesting divide that’s opened up about the M’s throughout the blogosphere and on twitter; two opposing camps who each point to a set of facts, make some logical arguments, and come to very different conclusions. The M’s are a good team that is projected to compete for the playoffs, and the M’s seem like they’ve fallen a step behind some of their rivals, and if the season started today, appear a player or two short in what may be their best shot at contention in a while. The key to this seeming discrepancy is the fact that their core, now supplemented with SS Jean Segura, is among the most volatile in baseball – something we see when we look at their projections for 2017.

About a month ago, there was an interesting conversation on twitter between several current/former LL folks about the Steamer projections for the M’s core of Nelson Cruz, Robinson Cano, Kyle Seager and Jean Segura. Logan Davis noted that the projections for this quartet shaved a grand total of about 11 WAR from their 2016 production; that is, their forecasted production was much, much lower than their recent history – a fact that narrowed or even eliminated the gap between the M’s and Angels, for example, and made the M’s look much worse than the Astros. That led to some follow ups about how often that happens. This seemingly simple question is really two separate ones: how often does a team lose 10 WAR from its star players (answer: all the time) and how often is a group of star-level players *forecasted* to lose 10 WAR between them (answer: not often)? The reason for this little historical diversion is two-fold. First, lol 2015 Nationals, but second, and more importantly, it highlights WHY many fans are so nervous about 2017, and why people are increasingly frustrated by a team that’s still projected to contend.

So, first, let’s take a look at teams that have lost 10 WAR between four players – four stars or regulars whose collective production drops by 10 full wins from one year to the next. My first thought was to look for good aging teams, and yes, that produced a ton of teams that have done this. 2009 Phillies? Of course. 2004 Mariners! Oooohhh yeah. You can find teams that meet this criteria by looking at teams that seemingly overachieved, like the 2005 Chicago White Sox, or, with a bit less emphasis on “achievement,” the 2007 Mariners. You can find teams who’ve lost 10 WAR between 4 players all over the place, because it’s remarkably common; baseball is unpredictable, etc. etc. But a few teams cry out for more explanation. I mentioned the White Sox, who lost a bunch of WAR the year after their improbable World Series win in 2005, but they hilariously (depending on your feelings for them) did it again the next year, with a different group of players. That is, they lost a bunch of production from 2005 to 2006 (Buehrle had a down year, and Scott Podsednik came back to earth, and Neal Cotts utterly collapsed, only to turn up years and years later with the Rangers), but then an entirely different group of stars face-planted from 2006 to 2007 (Jermaine Dye went from all-star to sub-replacement level, and Joe Crede completely collapsed at age 28. Still, though, the most unreal core-collapse in recent memory belongs to the 2014-15 Washington Nationals. In 2014, the Nats had four position players worth at least 4 WAR. A year later, that group shed a grand total of nearly 16 WAR, with their top 2 most productive players in 2014, Anthony Rendon and Jayson Werth, losing 11 WAR between them. By RA9-based WAR, their pitching staff was similarly hobbled, with Tanner Roark and Doug Fister losing another 8.5 WAR that year. No team could survive an implosion like this, and the Nats were no exception, despite the fact that Bryce Harper went from 1.4 WAR in 2014 all the way up to 9.5 in 2015. Their manager was fired, and their overpriced, deadline-deal closer tried to choke out the NL MVP in the midst of one of the best offensive seasons in recent memory. Someone should write a book about that team.

So teams lose production all the time – but do teams lose that many wins in projections? Almost never. Here, we need to find not only teams that overachieved, but players with really odd career trajectories – either guys who were too new for projections to get a handle on their true talent, guys at the end of their careers who fight off aging much better than did their historical comparison players, or guys with really odd career arcs. The M’s now have two aging stars in Nelson Cruz and Robinson Cano whose comps started to turn bad at around this age. Nelson Cruz’s top 3 comps through age 35 (and Cruz played half of 2016 at age 36, thanks to his July 1st birthday) were all essentially done at 36. Cano is just 34, but it’s not surprising that the projection systems don’t know what to make of him thanks to what we can only hope was an injury-plagued anomaly in 2015. His comps aren’t all that encouraging either, even including the hall of famers – Ryne Sandberg posted his last greater-than-average OPS+ batting line at age 33, and Bobby Doerr retired after his age 33 season. Vern Stephens was essentially cooked at 35, Scott Rolen by 36. There are some much more encouraging decline phases, from George Brett’s and Adrian Beltre’s, so it’s not all bad news, but you can see why any projection system that takes age into account is going to be somewhat suspect of Cano and Cruz.

And that brings us to Jean Segura, a guy who was a league average batter at age 23 before spending two full, injury-free seasons hitting like Luis Sardinas before busting out as a 5-win shortstop last year. Even giving more weight to 2016, any offensive projection (one using math, that is; you could argue that his work with hitting guru Bob Tewksbary overrides whatever the numbers say) is going to take him down several pegs. Pairing him with Cruz, owner of one of the strangest career arcs in a generation, and you have a perfect storm. The error bars around any projections for Cruz or Segura have to be really, really wide. If his 2016 was a career year – if he retains any trace of the Jean Segura that put up an OBP in the mid .280s *in Arizona* over the course of 1,100+ PAs – then the M’s need serious help somewhere to really make a run at things.

On the plus side, you know who else has a core group of stars who’re forecasted to give back a ton of value vis a vis 2016? The Cubs. Kris Bryant was worth 8.4 fWAR last year and is still improving, but comps for guys that good that young aren’t common, so he’ll get compared to regular old stars instead: his Steamer projection is 5.7 wins. Kyle Hendricks and Jon Lester, at least by RA9 WAR, should lose a ton of value, too, and Dexter Fowler (probably not a Cub in 2017, but whatever) had his best hitting AND fielding year last year, and thus is projected to lose over 1/2 of his value next year. A lot went right for the Cubs, and they have hard-to-figure players, making their projections look much less rosy than a typical Cubs fan might think. And yet the Cubs are still one of the very best teams in the league. Saying the M’s projections look like the Cubs is forced at best and dishonest at worst, so I’ll just say that the M’s and Cubs projections have at least one similarity.

All of this just brings us back where we started, though. We’ve illustrated why the projections are going to struggle, and thus the reasons why the M’s look remarkably volatile going into 2017. But volatility works both ways – every projection has them behind the Astros, but given the projections don’t know what to do with Cruz/Segura/Cano…how would they know? It’s always a fool’s errand to pencil in last year’s production from the best guys and then add in some (positive) regression for the guys who struggled and call it a contending ball club, but the M’s have reason to think that their stars will be better than many think in 2017. At the same time, the M’s could very easily collapse if the end comes early and ignominiously for Nelson Cruz and if Segura turns back into Zack Cozart, but without the glove. All I can say is that I really hope that the *potential* for the M’s to beat their projections by a lot isn’t a factor that’s led them to be so tentative this past month. I’d hate to think the M’s are confident that the team as currently constructed can compete for the AL West, thus kneecapping any sense of urgency to get another starter. They may not have ever been involved with Edwin Encarnacion (who apparently didn’t want to play on the west coast), and they may not convince Jose Bautista to join the over-30 squad in Seattle, but they need to get creative – and soon – to improve their run prevention. There’s not a lot of help on the farm, at least for 2017, and the club lacks pitching depth. That’s critical, because again, once Cruz and Iwakuma go, thus club starts to look pretty mediocre pretty fast.


5 Responses to “Volatility and The M’s Last Best Shot”

  1. ck on January 3rd, 2017 6:48 pm

    Marc, thank you for the ante to this Season’s pot / plot? Diaz, who started in AA, was worth 1.9 War in 2016. Who knows what Vogelbach, Haniger, Gamel, Zunino et cetera will produce in 2017. Projections can be used as a baseline but 162 games provide a battlefield where teams can easily be overcome by events. Just has Texas won a lot of one-run games last year, stuff will happen. Good news for the M’s is that the 40 man ‘floor’ is higher, the Manager and coaches should be better with the lessons learned from last year, and the new defense should prevent many more runs against. How many Wins that will mean is why we will have to watch, and enjoy each game as it unfolds. Fingers crossed to ward off injuries…

  2. maqman on January 4th, 2017 4:59 am

    While projection systems are interesting and have some value they are too flawed to base significant decisions on IMO. The M’s have potential flaws and so do most teams. Each player is a separate entity with an individual career arc. Luck and injury can demolish any projection for individuals and teams. We have to play the games to find the answers, that’s the fun part.

  3. The Ancient Mariner on January 4th, 2017 3:14 pm

    Given that Dexter Fowler has already signed with the Cardinals, I think you can remove that “probably” . . .

  4. bookbook on January 4th, 2017 8:25 pm

    Maqman, it’s either projections or prayer. Given that the same God cares for the fortunes of all 30 teams, and their players and fanbases, I’d favor using projections/analysis, warts and all.

  5. Westside guy on January 5th, 2017 1:34 pm

    With Segura, there’s the added wildcard that his kid was sick and then died in 2014. Could that have had some effect on his 2014 and 2015 stats?

    I’m not arguing that he’s necessarily the 5 win guy he was last season… but I do think there’s reason to believe he was worth rolling the die on. He did put up 3.5 WAR in 2013 as well.

Leave a Reply

You must be logged in to post a comment.