The Value of Followers in Fantasy Leagues: Part I

I don’t know about you, but as a kid I was inundated with messages about not being a follower. “Don’t be a sheep,” they said. “Be a leader,” they said. “If your friends all came out of the bullpen to throw bulk innings after, would you do that too?” they asked.

And now I want to call all those people and say, “Have you been watching the MLB postseason? Everyone wants to be a follower!”

The opener/follower structure is certainly having it’s moment and it’s easy to understand why this has value to MLB teams. You can split the opening innings between a left-handed and a right-handed arm, making lineup construction a little more difficult. You can let your follower get to a third time through the order without that third time starting with the big bats at the top. And you can guarantee you get one of your pen arms in the game for the first inning, when you can control the matchups.

The value to fantasy teams, however, is less obvious and worth evaluating.

I started digging into this topic because of a discussion among Ottoneu players, but this isn’t really unique to Ottoneu. Followers helped me win an ESPN league this year and I spent a lot of time debating their value in a FanTrax league, in addition to using them in some Ottoneu formats, but not others.

But so much of my decision- making on followers was based on surface level numbers and gut feel. I wanted to dive deeper. To do that, we are going to answer a series of questions that should help us figure out how to think about followers in fantasy leagues. The questions we will ask are:

• Part I:
  • Are there actually certain pitchers who are “followers” and who can be reliably counted on to follow?
  • Can you roster pitchers to use as followers specifically?
  • How effective are followers relative to other pitchers?
  • How much better are individual pitchers when they follow vs. when they start?
• Part II:
  • How does scoring format impact follower value?
  • How do rules around innings caps and pitcher usage impact follower value?
• Part III:
  • Why does this come up so much for Ottoneu?
  • How should Ottoneu managers think of followers?

Today we will just start with the first four questions, understanding followers writ large. Then we will dive into league rules and how they impact follower value.

Are there “followers?”

This seems like an odd question – of course there are followers! – but I think it’s an important one, and I want to make a distinction between “pitchers who follow” and “followers.” Master of the Ottoneu Universe, Niv Shah, pulled a list of all follows that were identified on Ottoneu lineups pages last year. Over the course of the 2026 season, Ottoneu tagged 139 pitchers as expected to follow, for a total of 261 expected follow appearances. Based on a manual count of the RotoWire probables grid, they seem to have identified as many as 146 pitchers who were tagged as an expected follower at least once, totalling 293 follows. The discrepency, given RotoWire is a data source for Ottoneu, is likely impacted by the manual nature of my RotoWire count, plus some differences how the two sites make backward-looking updates. Suffice it to say, for our purposes, there were something like 260-300 expected follows this year, and about 140-145 pitchers were responsible for those follows.

I am focused on “expected” follows because the question of the value of a follower in fantasy assumes that you know a pitcher is slated to follow and can slot them into your lineup. A reliever who sometimes throws multiple innings isn’t the same as a follower, for a variety of reasons. They can get saves and holds; followers can’t. They are used as-needed, rather than planned, meaning you can’t plan around them. They’re a different beast and need to be valued differently, alongside other relievers.

So there are 140-145 “followers,” right? I don’t think so. To truly be a follower for fantasy purposes, you need to follow multiple times. I would argue at least five times. Any less than that, and you aren’t really a follower, you are just a pitcher who sometimes follows. Of the 139 followers identified by Ottoneu, only nine of them followed 5+ times. Three of the nine followed exactly five times. Only two had 10+ follows – Erick Fedde with 12 and Miles Mikolas with 10.

If you want to be generous and say, “hey even a guy who follows three times can be picked up and used to follow,” that list increases to 26. Still not a ton. Those 26 pitchers made 120 follows, 64 of them by the nine with 5+.

Can you roster “followers?”

So if we grant that there are some true followers, somewhere between two and 26 of them, they only really have fantasy value if you can identify and roster them. That can be a challenge depending on your format.

I played in my first ESPN head-to-head points league this year, my lone redraft league, done as an excuse to draft with my son and my nephew. I crushed them both, and then felt a little bad, and still haven’t pointed out to them that I won the league. How did I crush them both (and everyone else)? Well, I’d like to think I drafted well, managed my lineups well, and used the wire well, but part of it was identifying that we had daily moves (first come, first served for FA, no FAAB) and a scoring system that favored followers (more on scoring systems coming later). If I saw a guy tagged as a follower, I picked him up and voila. I could drop him the next day and be done.

In my CBS H2H (2x weekly FAAB) or my Ottoneu leagues (48-hour auctions for FA), or even my FanTrax dynasty league (daily FAAB), I needed anywhere from 24- to 96-hours notice to be able to pick up a follower. That creates a complication. Anyone who tracks this stuff during the season knows that teams often announce an opener and follower very late, sometimes on the day of the game. Joey Cantillo is set to start, until it is announced within 24 hours of the game that he’ll be the bulk guy after an opener.

If you were chasing followers, you had a short window to learn about this news and pick Cantillo up. In most of my leagues, and I suspect many of yours, that wasn’t possible. If you already rostered Cantillo, that was a different story – and later we will discuss if you should use a guy to follow when he is already on your roster.

That means that in order to effectively chase followers, you need to find the guys who follow regularly, not just often. We noted that there were between two and 26 “followers,” depending how many follows you think is necessary to earn the title, but many of those followers were hard to predict.

Erick Fedde, the league leader with 12 follows, was a traditional starter in his first appearance this year. Then he followed in his next. Then he started two, followed one, started four, followed two, and so one. He would eventually make four and then later six straight relief appearances, but not all of those were true follows – they were not announced ahead of time and in some cases he went an inning or less. So if you wanted to pick up a follower, you had Fedde all year, for 33 total appearances, only 12 of which were predictable follows. If he was worth using as a reliever or a starter, that might be fine – like Cantillo, if you already rostered him and wanted to use him when he followed, you could. But if you were looking for a follower specifically, he was using up a roster spot for 180 days so he could be used about a dozen times.

And that’s the best case scenario. Zack Littell made 12 relief appearances, but only were identified as follows ahead of time and he never followed more than twice in a row until August.

Ty Madden shows another risk. He was an announced follower four times. His first two appearances were 5 and 6 IP bulk outings. He never went beyond 3.1 after that and that 3.1 was the only time he went more than 3 IP. He was more “reliever who can go more than an inning” than he was a follower, and that impacted his volume.

So rostering followers requires a lot of churn and a lot of making guesses as to who will follow when. When a guy follows once, you have to decide you think he is worth picking up after that first follow and then hope he follows again next time. Which he probably won’t, given the majority of followers only follow a couple of times all season, if that, and rarely more than once in a row. And if you pick the guy up and if you are right that he will follow again, you have to ride that train until they stop following, which is probably after 1-2 appearances, and then make another guess – now that the guy had a non-follow, will he go back to following? Again, the data tells us “probably not,” which means you probably cut and move on.

As a side note, this answers a question I didn’t ask: what about daily vs. weekly leagues? Given how unpredictable and sporadic followers are, they are basically impossible to chase in weekly lineup leagues. If you have a pitcher on your roster you otherwise like, and they follow, that is probably fine. If you like Bubba Chandler or Emmett Sheehan or Joey Cantillo or Ian Seymour, don’t cut them just because they followed one or more times. But don’t go after followers in weekly leagues.

How Effective are Followers?

So we have now established that there are very few pitchers who are, in fact, followers, and far more who are starers or relievers that sometimes follow. And we have established that those followers often move in and out of the following role and it can be very hard to predict if they will stay in that role or not. So finding and rostering followers is, bluntly, a pain.

But no pain, no gain, right? And if followers are good enough, then who cares if there is some pain – go get that value! But what if followers aren’t, you know, good?

Well, that seems to be an issue, too. Of the 261 expected follows flagged in Ottoneu, only 236 actually happened (this will come up again later). For whatever reason, they were flagged as a planned follower and then didn’t follow. The other 236 outings were…not ideal, but not terrible.

In 899 innings (3.81 IP per appearance), they had a 3.80 ERA, 1.27 WHIP, 7.51 K/9. None of that is bad, but none of it is helping you much, either. Looking at our player rater, there were 27 pitchers in 2026 with an ERA between 3.60 and 4.00, a WHIP between 1.23 and 1.31, and at least 20 IP. In 12-team 5×5 leagues, they ranged in value from $3 to -$11. Only four returned positive value, and only one of those four had a K/9 close to our followers. Followers, as a whole, were low K, low win, no save relievers who put up mediocre numbers that were worth probably negative value in 12-team 5×5 leagues.

Looking at those numbers, you can understand why MLB teams use followers. That’s production an MLB team can and will take from a bulk guy! MLB starters as a whole this year (including openers) had a 4.24 ERA and a 1.29 WHIP. Followers did better. But not better enough for us.

How Much Better are Followers as Followers?

This is important because even though I just told you the average follower isn’t worth rostering, that doesn’t mean every follower is unrosterable. The average MLB SP isn’t worth rostering either, but that doesn’t mean we cut Tarik Skubal.

There are two ways to consider how much better followers are when they follow: individually or as a group. Looking individually comes with a lot of data concerns, most notably that the sample for any individual pitcher is small. As a group, however, following does seem to help. Taking only followers who also started at some point, those pitchers had the following numbers:

Well, that is quite a bit better, at least on the rates. The K’s aren’t as good and the volume is lower, but those rates are improved! That does suggest that there are pitchers out there who are not good enough to be used as starters but are worth putting in your lineup with the follower boost.

The problem, however, comes when we look at the individual pitchers. Here are the nine pitchers who had five or more follows with those same splits:

So those guys did pretty well following, which is probably why they got so many chances to do it. But there are some questions worth asking. I think we can agree Griffin Canning didn’t do enough as a follower to have value. Same for Spencer Arrighetti. Mikolas barely improbed enough to be worth rostering. Trey Yesavage had a 3.69 ERA, 1.14 WHIP and 97 K in 97.1 IP. He was worth $1 in 12-team 5×5’s this year. Mikolas as a follower had half the innings and barely more than half the K/9. So he wasn’t worth rostering, even if you believe he can maintain such huge gains when he follows.

Seymour, Bello, Littell, and Miles did enough, but you have to believe they can maintain outsized gains in their “as follower” splits – if they revert closer to the gains of 1.16 in ERA and 0.18 in WHIP, things get iffier. Seymour is the only one who looks interesting then.

Conclusions So Far

The major takeaway here is that not only are there very few regular followers, but there are even fewer who pitch well enough as followers to matter and, given the small samples involved, it’s hard to know if that would continue.

Even if, in theory, you like the idea of rostering a follower, it is really hard to find someone who can sit on your roster as a follower. There are some guys worth using when they follow if you happen to roster them, but there aren’t many of them, and even they aren’t really sustainable or predictable enough to roster them specifically as followers, with rare exception.

Those exceptions have less to do with the pitcher’s involved and more to do with your league structure, which is what we’ll cover in part II.