One of the criticisms we get most often is that we sometimes overreact to small sample sizes. It’s true that we do, and it’s true that we probably shouldn’t, given how hard analysts have worked over the years to caution people against that very act. I can tell you this much: Our intentions are always good. And I can also tell you this: Players like Daniel Murphy are why we can’t stop.

Everyone wants to be first to see the breakout, and in the 2015 playoffs, if you’ll remember, Murphy homered seven times in 14 games. We can only identify that as a breakout in hindsight, but that streak sent everyone to the video. There was a search for a reason, a search for understanding, and that’s when the world learned of Murphy’s work with Kevin Long. The Nationals subsequently took the chance on Daniel Murphy, Quality Hitter.

Look, you’ve read about Murphy already. You know what he did. But do you really know what he did? They don’t make many players like this. There don’t exist many seasons like that. It’s really quite extraordinary.

Read the rest of this entry »

Two weeks ago, when I was examining the New York Yankees, I noted that they were the only American League team to feature multiple relievers projected by our depth charts (which are powered by Steamer for now) to record at least 1.5 WAR in 2017. Only seven AL pitchers cleared that mark overall. Looking over at the National League, we find that even fewer pitchers hit that mark, though there is still a single team with two of them: the Los Angeles Dodgers.

One, as you can probably guess, is Kenley Jansen. The other, though, might come as a surprise. It’s Grant Dayton. Unless you’re a big Dodgers fan, the first time you probably met Dayton was in last year’s National League Division Series, when he coughed up the two runs, prompting LA manager Dave Roberts to call on Jansen in the seventh inning — and to bring Clayton Kershaw into the game in the ninth to close things out. So, in a certain sense, perhaps we should thank Dayton. If he escapes that seventh inning unscathed, we might not have been treated to Kershaw vs. Daniel Murphy with a runner in scoring position and the season on the line.

Read the rest of this entry »

Last Thursday, two free-agent hitters found homes, with Mark Trumbo returning to Baltimore and Luis Valbuena switching AL West cities, going from Houston to Anaheim. While the expectation was that Trumbo was going to sign something of an albatross contract — I named him the No. 1 Free Agent Landmine heading into the off-season — he ended up signing for a perfectly reasonable price; $37.5 million over three years. While the Orioles will need to resist the urge to put him in the outfield anymore, $12.5 million a year for what Trumbo can do at the plate is not some kind of franchise-killing overpay.

The Orioles did fine here, mostly; you could argue that they could have spent even less and gotten Chris Carter, a similar-enough player, but not making the most cost-efficient move doesn’t make this a disaster. Trumbo is a solid enough big leaguer, and $38 million in MLB these days just isn’t that much money.

I say all that up front to clarify that the rest of this post isn’t a criticism of the Orioles’ decision to retain Mark Trumbo. I just thought the juxtaposition of Trumbo and Valbuena signing on the same day was interesting because, well, look for yourself.

Over the last three years, Trumbo and Valbuena have both established themselves as useful players, mostly based on their ability to hit the ball out of the ballpark. Trumbo has a bit more power, but Valbuena draws more walks, and thus gets on base more often, so he’s actually been the better hitter of the two during that time. Oh, and Valbuena’s also a little faster, so he’s added a little extra baserunning value, expanding his lead even a bit more. Over these three years, Valbuena has been worth +23 runs relative to a league average hitter, while Trumbo has come in at +11.

Now, sometimes, multi-year comparisons aren’t all that helpful in figuring out why player valuations diverge, because the more recent data is the most important one. And of course, Trumbo is coming off the best year of his career, as he put up a 123 wRC+ last year. But in this case, looking at just the most recent year doesn’t change things much, because interestingly, Valbuena also put up a 123 wRC+ last year, the best mark he’s put up in his career. This isn’t a case of one guy trending up and the other trending down; both were good hitters last year, and they’ve both been above-average hitters the last three years.

There isn’t an age factor here either. Valbuena was born in November of 1985, Trumbo of January of 1986. They are both 31, at the point where we can expect both to start declining in value, but not old enough where a catastrophic drop-off is imminent.

There are, though, two differences between Valbuena and Trumbo; the thing I find fascinating about these contracts is how those two differences drive the valuations.

The first difference is that Valbuena has some defensive value, and if he ends up playing a lot of first base — with Albert Pujols questionable for the start of the year, that sounds likely — he might end up being a solid defensive first baseman. In nearly 4,000 innings at third base, Valbuena has a career +10 UZR (though it has been worse the last few years) and as a former middle infielder, he’s more athletic than most guys who end up playing first base. Valbuena probably isn’t going to be a gold glover at first, but as a guy with the flexibility to play both corners and potentially be an asset at first base, there’s some real defensive value here. Trumbo is a solid defender at first base, but that position is blocked in Baltimore, so he’s either a liability in the outfield or a designated hitter, and won’t be adding defensive value in either case.

So, Valbuena has been a better offensive player the last three years, matched Trumbo’s wRC+ last year in the best year of Trumbo’s career, and adds some defensive value as well. I’ve held off including it as of yet since it can often be the only thing people focus on, but it’s worth noting that Valbuena has been worth +6.3 WAR over the last three years, while Trumbo is at +2.0. By overall production the last few years, it isn’t even really close; Valbuena has been significantly better.

And yet, while Trumbo got 3/$37.5 million with the qualifying offer attached, Valbuena got 2/$15M, despite not being tied to draft pick compensation. The market looked at these same-aged players and preferred Trumbo, despite a lack of a massive offensive advantage and definite defensive limitations.

Why? Because there remains a significant difference in how teams value everyday guys versus platoon players, and fair or not, Trumbo is seen as a player you can stick in your lineup regardless of who is pitching, while Valbuena is viewed as a part-time player.

With Trumbo, you’re basically getting the same thing no matter who is pitching; his career wRC+ splits are 113/110, so you can put him in the lineup and expect mostly the same production everyday. Valbuena, like most left-handed hitters, runs a bit larger split, with a career 86/98 wRC+ split against lefties and righties, and an even more extreme 79/126 split over the last three years. Valbuena’s entire emergence as a quality hitter has been based on his ability to hit for power against right-handed pitching; against lefties, he still hits like the middle infielder he came up as.

A corner infielder who hits lefties like Valbuena hits lefties shouldn’t be starting against them, so the Angels are almost certainly going to platoon him, and the fact that they have to pay another player — and more importantly, dedicate another roster spot — to a guy to share his job dramatically discounts his value on the market. Full-time guys get paid on a different scale than part-time guys, and Valbuena is a seen as a part-time guy, so he gets less than Trumbo despite the performance advantages he’s displayed of late.

But I wonder if the discount being applied between the two groups is too heavy, because while it makes plenty of sense to platoon Valbuena and get a higher overall level of production, you don’t actually have to. The Angels could choose to play Valbuena everyday and save the roster spot for some other use, if they really see it being a significant negative to have to carry a right-handed first baseman to share that job. And if we just project Valbuena out to Trumbo-like playing time, it’s still not clear that Trumbo is significantly better.

Let’s just say the Angels decided to just play Valbuena mostly everyday, not platooning him more heavily than any other left-handed hitter is. In general, LHBs who get about 600 PA per season end up having the platoon advantage in about 70-75% of their plate appearances, we’ll use 73% just to split the difference. With exactly 600 PAs, that would mean Valbuena would get 438 PAs against RHPs and 162 against LHPs.

Let’s take a fairly extreme position, and say that his true talent platoon split right now is something like 90/110 wRC+ — remember, that’s almost double Valbuena’s career 12 point split — which is actually a pretty significant split as far as MLB players go.

If we give him the 73/27 distribution of PAs that most regular LHBs get, that would work out to something like this in a regular role.

That 105 wRC+ is an almost exact match for the average of the ZIPS and Steamer projections for Valbuena in 2016; Steamer is on the low side at 99 overall, while ZIPS is up at 113, so their blended average is 106. So that at least passes the smell test. But we don’t know what percentage of platoon advantage those systems were projecting, and since neither had him forecasted for 600 PAs, we should assume they’re probably letting him face a higher proportion of RHPs. So let’s make Valbuena a little worse than those overall projection numbers, and re-run the estimate to give us something a bit worse than what ZIPS and Steamer are projecting him for in part-time duty.

In this estimate, we’re being pretty harsh on Valbuena; his wRC+ against RHP goes down 21 points from what it was the last three years, while his vs LHP number only bounces up six points, and is still below his career average. This is probably pretty close to what Steamer is projecting Valbuena’s splits at, and it is definitely the more negative outlook, given his recent track record.

But even with that pessimism, Valbuena still projects as a league-average hitter while playing everyday. Trumbo projects for a 110 wRC+ as an everyday guy, in both ZIPS and Steamer, so the forecasts agree that, for next year, you’d rather have Trumbo’s bat than Valbuena’s, especially if you’re not willing to use a roster spot to platoon Valbuena with an RHB.

But the difference between a 110 and a 100 wRC+ over 600 plate appearances is about seven runs. That’s not nothing; that’s most of the way to one extra win. But then, there’s the baserunning, where Valbuena makes up some of that gap. And then there’s the defensive value, which isn’t trivial. If you give Valbuena some credit for being able to still play third, he probably makes back a few of those runs, and in the end, we’re looking at a gap of a couple of runs between full-time Trumbo and full-time Valbuena, if there’s any gap at all.

And full-time Valbuena is an inefficiency; with just a modicum of work, you can find a decent right-hander to face left-handed starters, and end up with a better overall rate of production. Yeah, it costs you a roster spot and that guy doesn’t play for free, so his cost has to be factored into the equation, but most teams carry a right-handed bench bat anyway, and if you have to pay a slight premium to get a guy who is worth starting occasionally, you still come out ahead overall.

The big story this winter has been the market correction on bat-only corner guys, but interestingly, the Valbuena signing points out that there’s even more room for those kinds of guys to come down in price in the future. Instead of paying even this reduced rate for the full-time slugger, signing a guy with a platoon player label at the discount currently being applied is an even cheaper way to get similar production.

Read the rest of this entry »

Rob Kaminsky got off to a rocky start in 2016. His delivery compromised by a balky back, he put up plenty of clunkers. Over his first 10 outings for Double-A Akron, the Cleveland pitching prospect allowed 30 runs in 45 innings of work. He also spent time on the disabled list. All in all, Kaminsky was a wreck.

Then things turned around. Buoyed by mechanical fine-tuning that accompanied his rehab sessions, the young southpaw was a stud from mid-June onward. In his last 17 starts, Kaminsky allowed two-or-fewer runs 14 times. Over that span, which included the Eastern League playoffs, he hurled 103 frames and saw just 30 runners cross the plate.

Despite his rebound — and having pitched the entire season as a 21-year-old in Double-A — Kaminsky is going in the wrong direction with regard to prospect rankings. He’s plummeted all the way to 26th on Baseball America’s Indians list after coming in at No. 9 a year ago. Our own prospect guru, Eric Longenhagen, has Kaminsky 14th, which is six spots lower than he was 12 months ago.

How did the 5-foot-11 left-hander’s season unfold, and what looms in his future? Here are answers to those questions, courtesy of Kaminsky himself, and Dave Wallace, who was his manager in Akron.

———

Wallace: “If I had to pick one story from this year — pick a specific player — [Kaminsky] would be it. He was absolutely our bulldog down the stretch, and in the second half in general. But in the first half, there were some games that weren’t pretty. He had some significant delivery adjustments that needed to be made. And they weren’t easy.”

Kaminsky: “I would have benched myself, with the first half I had. Everything that could have went wrong, went wrong. I was pitching terrible, and I was pitching hurt. But there were some days I actually felt alright and still got hit around, so it’s not something I can completely blame on the injury. I need to take some ownership of my performance.”

Read the rest of this entry »

After having typically appeared in the very famous pages of Baseball Think Factory, Dan Szymborski’s ZiPS projections have been released at FanGraphs the past few years. The exercise continues this offseason. Below are the projections for the Oakland Athletics. Szymborski can be found at ESPN and on Twitter at @DSzymborski.

Other Projections: Arizona / Atlanta / Baltimore / Boston / Chicago AL / Chicago NL / Cleveland / Detroit / Houston / Kansas City / Los Angeles AL / Los Angeles NL / Milwaukee / Minnesota / New York AL / Miami / Pittsburgh / St. Louis / San Diego / San Francisco / Seattle / Tampa Bay / Toronto / Washington.

Batters
Yesterday, in these pages, Jeff Sullivan wrote a brief post entitled “How Much Hope Do the Bad Teams Have?” In it, he examined the end-of-season outcomes for clubs that had received uninspiring preseason projections. It’s likely that Sullivan’s post has some relevance to this Oakland club as it’s currently constructed.

According to Dan Szymborski’s computer, only two field players, Marcus Semien (627 PA, 2.7 zWAR) and Stephen Vogt (467, 2.1), are expected to exceed the two-win threshold, typically the mark of an average player. By comparison, the Milwaukee Brewers — a club definitively in the midst of a rebuild — feature six field players forecast for two or more wins. Minnesota also has six. San Diego, five.

One, searching for optimism, might note that a number of Oakland’s starters — Rajai Davis (410, 1.1) and Jed Lowrie (399, 0.4), for instance — receive only modest plate-appearance projections, and thus better prorated figures. That said, the playing-time numbers are based on sums from previous seasons, too, which themselves have been modest due either to injury or role.

Read the rest of this entry »

What happened to Clayton Kershaw in Game 6 of the NLCS? According to a new study by Northwestern University, maybe it was jet lag.

Looking at 20 major-league seasons and 40,000 games’ worth of data, researchers found that jet lag perceptibly “impairs” player and team performance. The study is likely to be passed around many major-league front offices and strength-and-training departments. In a sport where every team is looking for hidden value at the margins, the value of better rest and recovery is just beginning to be explored, understood and focused upon — and is perhaps a considerable inefficiency in the game.

Dr. Ravi Allada, a circadian-rhythms expert, led the study:

“The negative effects of jet lag we found are subtle, but they are detectable and significant. And they happen on both offense and defense and for both home and away teams, often in surprising ways….

“For Game 6, the teams had returned to Chicago from LA, and this time the Cubs scored five runs off of Kershaw, including two home runs. While it’s speculation, our research would suggest that jet lag was a contributing factor in Kershaw’s performance.”

One of the homers in question:

Of course, Kershaw did pitch on extra rest that start, and Kyle Hendricks himself did just fine after traveling back east, but perhaps the rest could not save Kershaw from the clutches of jet lag.

Read the rest of this entry »

Read the rest of this entry »