Value of starting versus closing

DMZ · August 31, 2008 at 8:00 am · Filed Under Mariners 

WoTYC, 2, Paul B:

“I’m still curious about how Morrow’s value as a starter compares to his value as a reliever. Potentially, and making appropriate assumptions of course.”

Interestingly, almost two years ago baseball analysts had almost exactly the same argument about Jonathan Papelbon. Papelbon stayed a closer. If I can boil the whole thing down, when you account for the value of the innings they pitch, a top-tier closer is worth as much to their team as almost all starters are. That varies, of course, by how the manager uses them. “Only the 9th, only 1-3 run lead” closers end up being a lot less valuable than ones that pitch in ties and can pitch in the eighth.

So take a look at, say, WPA on fangraphs, which is a fair measure of how well a player did when it could most affect the team’s success.

Of the top 30 right now, the breakdown is essentially:
18 starters
7 relievers who’ve closed since the start of the season
5 late-inning relievers.

There are a couple issues to look at then:
– Is it easier to find a replacement for him as a starter or a reliever?
– Is his stuff particularly well-suited to one role or the other?
– If Morrow is better in short stints throwing all out, is he that much better that it outweighs the contribution he’d make in innings as a starter?

In rough order: it’s easier to find relievers who can close. Every year, a bunch of closers and late inning relievers show up and do well without a previous resume. They’re converted starters, failed starters, castoffs, injury recovery dudes, college closers, whatever. Pitchers who don’t have to pace themselves can throw harder, and don’t have to have the multi-pitch arsenal of a starter who’ll have to face the same hitters over and over. Pitchers with huge platoon splits or other vulnerabilities can have a good manager put them in situations where they’re well-suited to succeed, while a starting pitcher can only be so vulnerable in any area before they’re reduced to total ineffectiveness.

Look at the list of relievers in the top 30 WPA pitchers right now. There’s easily a half-dozen names on that list that at least at the start of the year would have drawn blank looks if you’d thrown them out at an average fantasy draft and certainly weren’t household names. This, incidentally, is something the M’s have done really well over recent years — picking guys out of the farm system for an eye towards filling out a bullpen with cheap, young, effective arms.

Then the question of what Morrow is suited for. Durability enters into this: when we compare the relative valuation of relievers against starters, there’s an assumption that the injury risk is the same. Your opinion on this may vary — relievers throw at maximum effort more frequently, but for much shorter times, while starters toss a hundred pitches in a row at a significantly lower exertion. Here, it’s hard to offer outside analysis. This is entirely a judgment call by teams.

Does Morrow, in particular, have characteristics that make him more suitable for relief? He does. One of the things Silver found in looking at successes and failures of relief conversions was that both strikeout rate and walk rate were associated with relief success, as was having a low ISO (SLG-AVG). Groundball rate was not, which surprised me, since that’s so closely tied with not giving up home runs. Morrow’s had some issues with walks, though that doesn’t seem to be an inseparable part of his game in the way that it is for (say) Jenks.

Morrow’s ISOs, by year: 105, 127 — that’s significantly under the major league average (which stands at 150 right now)

Silver, theorizing on the why:

A relief role emphasizes high-impact pitching and deemphasizes consistency and durability. A low ISO is a good proxy for high-impact pitching, a pitcher who can take control of the at-bat with one or two great pitches. Meanwhile, a low BB rate is a good proxy for mechanical consistency.

How good does Morrow have to be as a starter to make it an even swap, entirely in terms of value contributed? Here’s Silver again regarding Papelbon:

…, and we find that a 2.00 ERA closer is roughly as valuable as a 3.69 ERA, 200-inning starting pitcher.

What we wouldn’t give for a 3.69 ERA, 200-inning pitcher. He’d be the second-best pitcher in the rotation, by a long ways. And again, it’s a lot easier to find a replacement pitcher for the bullpen than it is to find a replacement starter who can put up that kind of line.

Morrow’s stuff may be better suited for a bullpen role, and we know that he’s done well there so far. The determining question seems to be his durability: if Morrow can go 200 innings in a season in the near future without increasing the risk he’ll be injured, then he should be starting.

But here’s the other issue, which hasn’t really been discussed enough yet: over the last two years, we’ve seen Morrow’s development take an odd path, as he’s struggled not only with command but with falling in love with the fastball when he’s experiencing success with it, not mixing pitches, and so on. One of the happy storylines this season was seeing him start to overcome those difficulties. Starting forces him to confront those weaknesses in a way that late-inning relief doesn’t, and in Morrow’s case this makes the decision to try starting him the right one. If Morrow has (say) potentially three plus pitches and relieving means that he relies on two of them, throwing 85% fastballs for the rest of his career, that’s a huge amount of potential wasted. That role can be filled by anyone who can throw fast and has a decent change.

If Morrow’s conversion to starting means that he’s challenged to work on and effectively throw more pitches, work batters better, mix up his stuff and become smarter about pitching, there are a couple of possible outcomes:
– He succeeds, and becomes a more valuable starter than he’d have been a reliever over the long term
– He succeeds, but the durability issues force him back to the bullpen having improved the very things that often have held him back from domination as a reliever
— He fails, and goes back to the bullpen and we get the Morrow we’ve seen of 07-08

Since Morrow’s attempt at starting’s not going to make a difference for this year’s team and the benefits will be reaped by a future club with a shot at contention, starting’s the way to go.

This post was written as one of the requests by a USSM supporter as part of Thank You Content Week. Join them!


9 Responses to “Value of starting versus closing”

  1. Tree on August 31st, 2008 9:35 am

    I’m suspicious of Silver’s use of leverage index, since the replacement reliever isn’t going to inherit the closer’s high leverage. The second best reliever will get the highest leverage, the third will get the second, etc. This should reduce the value lost by converting a closer, and therefore raise the necessary ERA that the closer needs to have as a starter.

  2. joser on August 31st, 2008 10:22 am

    The unique thing about Morrow is the opportunity cost: the M’s have now spent not one but two first round draft picks on the hope of making Morrow a starter. If he ends up joining Fields in the bullpen on a permanent basis, that goes a long way to undoing all the good work the M’s have done in acquiring cheap bullpen arms. It would be mitigated somewhat if they could trade Putz high, but that doesn’t look lightly at this point. (Maybe next year, if Morrow washes out as a starter and goes back to the pen, and Putz reliably regains his old form).

  3. Rain Delay on August 31st, 2008 10:59 am

    #2: The M’s would actually have to sign Fields first and as of right now it doesn’t look like it’s going to get done. The M’s and Fields are about half a million off – and Boras would be more than happy to send Fields to the Indy’s and have him re-enter the draft in 2009.

  4. joser on August 31st, 2008 11:31 am

    In which case they still will have spent two draft picks to get Morrow into the rotation, and will have even less to show for it.

  5. Teej on August 31st, 2008 3:28 pm

    I’m rooting for the M’s to not sign Fields, to be honest. It sucks to not get your first-round pick in any year, but if the M’s fail to sign him, then offer Raul arbitration and let him walk, they should end up with four picks in the first round/sandwich round next year. One of those picks could be used on something more valuable to this team than a reliever.

  6. SequimRealEstate on August 31st, 2008 3:39 pm

    Rain Delay, What is you source of information re: Fields? I asked the authors of this blog that question a couple of weeks ago and was assured the deal was done and they were just waiting for the timing of the announcement.


  7. Teej on August 31st, 2008 3:51 pm

    What is you source of information re: Fields?

    Hickey had it yesterday.

  8. SequimRealEstate on August 31st, 2008 4:44 pm

    Thanks Teej. While I am at it thank you DMZ for another excellent post. The exlaination was different than I expected. Appreciate the education.

  9. vj on September 1st, 2008 6:42 am

    When the M’s drafted fields, they may have still be believing that they had a shot at contention this season and that they would need him for that. Now, not so much any more. Indeed, taking a compensation pick in his stead in the next draft might be the better option. That compensation pick significantly lowers the leverage for Boras’ strong-arm tactics. A first-round pick cannot pitch an infinite number of seasons in independant baseball.

Leave a Reply

You must be logged in to post a comment.