Should You Target Openers for Ottoneu

Ever since the opener became a thing in MLB, I have been wondering if I should be aggressively targeting openers among my RP corps in Ottoneu. I typically like to carry 6-8 RP on a roster, and having the option to extend my bullpen by locking in an opener on occasion seems like a nice way to get some extra pen innings. But opener innings aren’t exactly relief innings. Yes, they are pitched by relievers, and they are short-burst outings like relief appearances, but they don’t offer any hope of a save or hold, and they are often not as short-burst as they would be from the pen.

I finally got around to trying to answer that question for myself and, to be honest, I am not sure I have a clear answer.

I used a combination of Baseball Reference’s Stathead, which allows me to find specific games (all starts of 3 IP or less) and FanGraphs’ Splits Leaderboards, which allows me to isolate “as SP” vs. “as RP” splits.

I then tried to focus on guys we would consider openers. I started with all pitchers with 20+ innings this year and eliminated anyone with zero relief appearances or zero starts. I then took out anyone who was a reliever for <50% of their appearances and anyone who averaged more than 3 IP per game started. This allowed me to remove players who are either swing men – sometimes SP, sometimes RP, but are used as traditional starters, not openers – or who were starters at one point in the year before transitioning to the pen (or vice versa), such as Garrett Whitlock or Matt Brash.

In the end, I was left with 39 pitchers who made a total of 74 starts of three innings or less.

The first thing to notice is that, on average, this group opened less than twice on the season so far. This immediately made me doubt that targeting openers for my rosters was a good strategy – openers barely exist. In aggregate, 74 appearances isn’t a ton. And on a per pitcher basis, having a guy get a total of three opening appearances over the year just can’t have THAT much of an impact.

The next thing I looked at was how they performed and, in aggregate, the answer is “poorly.” They threw a total of 136.2 innings as starters and put up 470.8 points, for 3.45 P/IP across their starts. If we stopped here, the answer would be clear – not only should you not target openers, you should not use them.

However, there are five pitchers who made one start each bringing this average way down and those five pitchers have something in common besides bad numbers as openers: bad numbers as relievers. None of them are pitchers you would roster in Ottoneu regardless of role, and if we take them out, the numbers look better.

Taking out the five appearances as SP by those pitchers, we are left with 69 opening appearances and for 3.97 P/IP. Still not good, and still not worth targeting, but better.

But we can narrow things down further. Let’s focus on the nine pitchers who have made more than two appearances as an opener. Maybe those nine are uniquely interesting targets? Anyone making just one or two appearances as an opener isn’t taking that role often enough to matter, anyway, and maybe teams are using guys in that role more often if they are good enough to justify it.

At first glance, that is no better – nine pitchers throwing a total of 71.1 IP at 3.42 P/IP. But again, there is an outlier. Reiver Sanmartin of the Reds has been used to start four times for 11.1 IP and has put up -42.7 points. Like the five relievers who made one start each, Sanmartin isn’t a guy you should roster regardless of role, so let’s leave him out, too.

That leaves us with eight relievers who have made three or more starts and averaged three or fewer IP per start:

Looking at this list, things seem more promising. That is a pretty solid 4.78 P/IP – you would gladly take that from a starter and so adding an extra couple innings at that rate a few times a year is useful.

But, we are talking about a pretty small benefit. There are 46 pitchers who have made 10+ starts this year with greater than 4.78 P/IP. That makes these openers decent, but not great, as SP. Those extra innings they are offering are helpful, but not game-changing.

The difference between a good RP who never opens and a similarly talented RP who opens a few times a year is pretty small. In fact, if that opening appearance comes at the expense of a relief appearance, it might be a negative.

Using Beeks as an example, he is putting up a solid 4.52 P/IP as a starter and 5.81 as a reliever (which is not that solid). If you swapped out his five starts for five additional relief appearances of one inning each, you would lose 14.7 points but use up 4.2 fewer innings. If you can find a starter to give you just 3 P/IP over those 4.2 innings, it’s basically a wash. And you should do far better than that from a random SP.

Even if you account for the lost predictability (I can guarantee I will get 9.2 innings of Beeks as an opener, but I might only get 60% of his relief appearances in my lineup), there still isn’t a benefit. Now I “lose” 26.3 points and use up 6.2 fewer innings, so any 4 P/IP SP can make that up for me, no problem.

My takeaway here is that openers don’t really matter for Ottoneu – they might even hurt. You should be aware if a player on your roster is opening and try to get them in your lineup, as a good reliever locked in for 1-2 innings while opening up a relief spot in your lineup for someone else is likely a benefit. If a pitcher is good enough to be on your roster as a reliever and gets a shot to open, they are probably good enough that you want those innings.

But you shouldn’t change your strategy on finding or acquiring relievers to account for openers. Go get the best relievers you can, and only worry about whether they open after the fact.