Otto von Ballpark

Rooker has some exciting skills, but also some serious weaknesses. Let's not forget he had a 34.7% K rate at AAA in 2019. Yes, some hitters can still do some damage even with a high K rate in the modern game -- but others can get carved up too. I had expected Rooker to get an opportunity to sink or swim, but I guess if he looked bad enough at the end of the spring, maybe all signs were pointing to "sink" and there wasn't much point to it. Garlick doesn't do much for me either, but I suppose he could wind up an average-ish stop gap / placeholder. Hopefully we can use LF to get Arraez's bat in the lineup (maybe even at the top of the lineup), while our starting infielders are healthy and productive too.

Not saying anything like that at all. Just that while there is definitely value in Happ-like signings, there are still limits to that value come October and it's important to keep that in mind too. For the Twins, that doesn't mean they have to sign elite FA instead of guys like Happ -- they may just need to calibrate their trade deadline and prospect promotion strategies a little differently.

At the same time, there are still lingering questions about how bargain shopping can influence postseason results. A guy like Happ is useful to almost every team to get through the regular season -- but the Yankees were pretty deliberate to minimize his use/exposure in the postseason. Can/will the Twins do that?

I'm not so sure that's a better methodology. Looks like this is the direct link for it: https://www.baseballprospectus.com/prospects/article/6516/future-shock-positional-differences/ That link and the specific .819 figure are outdated -- it was based on stats covering 2004-2006. League non-pitcher OPS was .773 over those 3 years; it's only .753 the last 3 years (2018-2020). Adjusting your .819 figure down by the same amount puts it at .799 -- with little practical difference from Kepler's .786 OPS in the same 3 years. On that basis alone, I'm not sure this is a meaningful criticism of Kepler, but there are more issues with the figure too: 1. The article says it includes the 30 guys with the most starts at each position. Assuming that's top 30 overall over the 3 year period, the "average RF" numbers include guys like Larry Walker -- but he only started 147 games total over the 3 year period (indeed, he was retired and didn't even play the last year). If Kepler can play 150 games a year, is it right to compare him to guys who can only handle 50-80 games a year? And to ignore the 80-100 games those teams have to play other guys? 2. With that low threshold of starts, it's also likely including guys who are part-time at the position, but full-time overall (filling out the rest of their at-bats at 1B and/or DH). And based on the primitive nature of most online splits circa 2007, I'm guessing they included a batter's total stats, not just those accumulated at the particular defensive position. If Kepler can handle 150 games a year defensively at the position, is it right to compare him to guys who can only handle as few as 50 games in the outfield but accumulate up to 100 games worth of stats at easier positions? And again, ignoring the 100 games those other teams have to fill at the position? 3. Raw OPS isn't park corrected. In league-wide stats, this is a wash, but if you are looking at a subset of players like this, it could skew the sample a bit. (Indeed, I notice the 2004-2006 RF list includes two Colorado Rockies batters, Burnitz and Hawpe, plus 20% of Larry Walker's PAs those years were with Colorado too.)

This makes all the sense in the world for Dobnak. His signing bonus back in 2017 was just $500, and he's still 2 full seasons away from arbitration. The $9.25 mil guaranteed in this deal has an incredible marginal utility for him!

Thielbar has an option year remaining too, so he wouldn't have be waived (unless they needed the 40-man spot).

You can't option injured players, so it's just a question of whether Colina is healthy enough to option (unlikely at this point) or he goes on the MLB 10-day or 60-day IL. I expect Colina will go to the MLB 60-day IL to open a 40-man spot for Law. Then if Colina is ready to pitch before the 60th day of the MLB season (but after the minor league season starts in early May), he can simply be sent to AAA on a rehab assignment until he is healthy enough to option.

Happ to the IL wouldn't make much sense, as he wouldn't be able to ramp up in a rehab assignment -- just simulated games. They've already announced Dobnak is starting the year in the bullpen, so unless some new injury/condition develops with Happ, he's not going to the IL. The Twins have a large staff with enough flexibility that they don't need Happ to throw more than 75 pitches his first time out anyway.

Bump! Watch the Twins-Red Sox on MLB.TV now (via the Red Sox broadcast). The free access works for archived games too, so you can log in and re-watch later.

Dobnak's defining stats this year seem to be a super-low WHIP (0.44) and more K's (18) than innings pitched (13.2). Let's take a look at the regular season SP results for some similar recent spring performers. To explain, I went to MLB.com and looked at their spring training stat registers for the past few years. I set it to all pitchers (not just "qualifiers") and sorted by lowest WHIP, then eyeballed the list to find SP who pitched a decent number of innings and then were primarily SP during the season. 2019: Sonny Gray -- great season! Eric Lauer -- meh season Marcus Stroman -- great season! Trevor Richards -- meh season 2018: Jacob deGrom -- great season! Lance Lynn -- meh season Luke Weaver -- bad season Jakob Junis -- meh season 2017: Dylan Unsworth -- did not pitch in the majors? Masahiro Tanaka -- meh season Felix Hernandez -- meh season Clayton Kershaw -- great season! This is by no means scientific, and league WHIP has dropped and K's have spiked the past couple years, so it gets harder to find comps the further you go back. But it seems pretty clear that the sample is too small to warrant reading too much into Dobnak's spring line so far.

I'm not sure where you are looking, but MLB corner OF as a whole had a .761 OPS in 2020. Basically a 105 wRC+ every year lately: https://www.fangraphs.com/leaders/splits-leaderboards?splitArr=38,40&splitArrPitch=&position=B&autoPt=false&splitTeams=false&statType=mlb&statgroup=2&startDate=2016-03-01&endDate=2020-11-01&players=&filter=&groupBy=season&sort=-1,1

MLB.TV is having a free preview the next 4 days (March 27-30), and no blackouts for spring training games this year. So you can create or sign in with a free MLB.com account and watch the Twins game starting now! https://www.mlb.com/live-stream-games On Tuesday March 30, the "free MLB.TV through T-Mobile" sign-up begins too -- I'll post more info on that later.

All I was responding to was this: Kirilloff is not even 2nd after Rooker for me. Garlick has been the best hitter in ST. Granted it's spring training. He also had 100 pt higher OPS at the MLB level in 2019 compared the Kirilloff numbers at AA. Garlick's AAA OPS was over 1,000. His history suggests he is better prepared to play at the MLB level. Garlick, despite an option and 2+ minimum salary years remaining, was passed over by the 24 worst MLB teams *twice* this offseason. That tells me more about his MLB readiness than his passable first 53 PA in MLB (ignoring his awful next 23), or his OPS while repeating the PCL at age 27 with the lively ball used at AAA in 2019: https://mlb.nbcsports.com/2019/09/03/triple-a-homers-increased-by-nearly-60-percent-this-season-thanks-to-juiced-baseball/ As Jayson Stark of The Athletic notes this morning, that number includes a 59 percent increase in homers in the Pacific Coast League over last year and a 57 percent increase in the International League. Stark talks with a baseball executive who tells him that, from a development perspective, the PCL has now become essentially useless, and they are sending prospects to Double-A instead because the juiced ball is preventing clubs from accurately assessing players. If we're talking a 2 month opening for regular MLB starts, I'd at least be open to the idea of Kirilloff skipping AAA. I wouldn't automatically hand it to Kirilloff, but I wouldn't give it to Garlick by default either. At the very least, I would have played Kirilloff through to the end of spring training before making my decision; more likely, I would have sought other alternatives in the offseason. Off the top of my head, Bellinger (142 wRC+) and Conforto (160 wRC+ in 197 PA but trending down, 118 wRC+ over his last month) are not exact matches for Kirilloff's 127 wRC+ performance in AA coming off his wrist injury, but nothing bright-line disqualifying. Their teams didn't have to consider a cancelled 2020 and delayed 2021 minor league season either -- Bellinger was promoted from AAA to make his MLB debut on April 25, a scenario not possible for the 2021 Twins. Nor did their teams prefer Garlick as a 2-month MLB starter!

One start in spring training wouldn’t mean much. You’d need multiple starts to stretch out beyond a reliever’s workload anyway. And both guys are coming off health issues. I’d actually be surprised if they transitioned either one to starting during the 2021 season — seems more like a “next offseason” thing — but the White Sox have been a little surprising lately.

May actually didn’t walk that many guys as a starter in MLB, or even ultimately fail as a starter. He walked 7 in the disaster that was his very first game, but the remaining 9 games as a rookie he only had a 7.6 BB% (he gave up a ton of runs, though). Second season, 16 starts, 1.9 fWAR, 20.3 K%, 5.0 BB%. (League was 17.0 K% and 6.7 BB%) ERA was 4.43 but the league SP ERA was 4.14. I don’t know if that would have held up over time, but he never really got a chance to start again after that.

There's no perfect metric to compare between levels and leagues (and nothing is going to rescue selective small samples like spring training and Garlick's 53 MLB PA in 2019), but wRC+ is going to have fewer flaws than the raw OPS numbers that you quoted. if you include the postseason -- just 5 more games -- Kirilloff's season OPS would have climbed to .793, or a ~127 wRC+. Considering he started the year playing through a wrist injury too, that number isn't noticeably different than, say, Bellinger's AA wRC+ of 142. Bellinger had another ~90 PA at AAA but it's not clear that those 90 PA were critical to his development. I don't want Kirilloff to skip AAA, but with the AAA season starting a month late this year, if the alternative to sending Kirilloff down was 2 months of starting Kyle Garlick in a contending season, maybe I'd rather give Kirilloff a shot first. Of course, in a non-Rooker world, I'd probably just use Arraez a lot more in LF (maybe I would anyway!). Maybe Garlick could hang around as a 4th or 5th outfielder -- roles for which he's not in direct competition with Kirilloff.

I think you make a fair point about Thorpe, but those K differences aren't *quite* as dramatic as they look due to changing contexts. The 2019 IL K% was 22.8%. Even just a few years prior, in 2014 for Trevor May, it was only 19.4%. Thorpe still outpaces his league by a wider margin, but the league difference wipes out over half his K% advantage over May.

I believe the rule is that MLB FA signings can't be traded before May 15th. Obviously that doesn't come up too often!

I wouldn't go that far. Garlick's 2019 and AAA OPS figures have been in hitter's leagues. His 2019 MLB wRC+ was only 114, with a 35.8% K rate. He's probably not as bad as his 2020 Phillies numbers suggest, but I'm not terribly eager to find what midpoint he could settle at. (ZiPS projects him for a 72 wRC+.) By comparison, Kirilloff was in a pitcher's league in AA and posted a 121 wRC+, despite a wrist injury in the early part of the season -- then had an insanely good AA postseason that's not reflected in those numbers either.

It's not the spot, though -- it's the playing time. Astudillo might play once a week -- and there's nothing wrong with that! That means everybody else is healthy and hopefully productive. They can always reassess later and use the spot differently. You missed Arraez in this analysis -- he's got a 3 WAR projection and might be our best leadoff hitter. I hope he gets more playing time than just backing up 2B/SS/3B. Garver and Jeffers might benefit from a DH or 1B game too. Also, there's no DH for the first 3 games of the season, so that takes 3 more starts away from our best projected hitter in Cruz. (And while I sympathize with the Cruz rest theory, I trust that the Twins are aware of his age and are already taking that into consideration with their current roster plan -- they don't need Kirilloff for that purpose.) And all that just to get Kirilloff irregular at-bats over a month of MLB, without a set defensive position, starting his service clock before he's even a MLB regular, after a cold spring training, 1 active game in 2020, and no career AAA experience. Seems sub-optimal to me. And none of this is set in stone. If Rooker looks really overmatched the first few weeks, we can always swap him for Kirilloff. If someone gets hurt, we can make a move too. If those things happen, or Astudillo is somehow starting 50% of games, and the Twins won't promote Kirilloff, I'll be right there criticizing them. But right now, I just don't see it. The current roster construction and expected playing time distribution look perfectly rational.

The White Sox have said they still consider Kopech and Crochet as starting prospects, although both will open the season in the bullpen (in part to help manage their workloads). They're probably the White Sox strongest SP depth right now.

I get that you don't like Astudillo or a 9th reliever, but neither of those spots have anything to do with Kirilloff. Kepler, Cruz, and Arraez are projected 3 WAR players. Sano's projection is only 2 WAR but he was a 3 WAR guy in 2019 and the team is committed to giving him a chance to do it again. These are not the guys we want to bench so we can achieve playing time equity across the whole 26-man roster. I laid out the numbers in my post above, but trying to give even half-time starts to Kirilloff would have a noticeable impact on other roster goals. Kirilloff is a very good prospect but his performance and experience to date do not suggest any particular urgency to try to squeeze him in right now at the expense of these other goals. Mind you, I don't necessarily think your plan is bad -- I am excited to see Kirilloff too -- but I see nothing wrong with the plan the Twins are apparently choosing either.

The White Sox bench looked really bad, even before the Jimenez injury. I don't think they really had "plenty of options on any given day" at DH -- if Vaughn didn't make the team and produce right away, they were almost certain to have a replacement level or worse player in their lineup every single day, even at full health. The Twins, on the other hand -- I don't know that all of our LF options will be great, but between Cave (2.8 fWAR in 660 PA), Rooker, and even Arraez, I'm pretty confident we can put together an above-replacement group, even before Kirilloff enters the picture.

This is a good summary of who is left:

Fangraphs has those, but the 2020 data isn't loading for some reason. You can still view previous years: https://www.fangraphs.com/players/max-kepler/12144/spray-charts?position=OF&type=battedball&pid2=12144&ss1=2018&se1=2018&ss2=2019&se2=2019&cht1=battedball&cht2=battedball&vs1=ALL&vs2=ALL