Positional Production in 2021

For years, it was taken as a given that fantasy replacement level varied by position, such that a C, a SS, and a 1B with the same exact line would have different values because of how they compare to their peers. That has been challenged more often lately, especially in leagues that have multiple utility spots and no corner or middle infield spots, like standard Yahoo leagues. In Ottoneu, most people who create values are still assigning players a primary position, setting a replacement level for each position, and adjusting values for each player based on that. I went back to look at production by position, based on Ottoneu FanGraphs Points scoring, in 2021 to see how the positions compare. Is C really that much weaker than everyone else? Should you pay a premium for MI production? Is 1B much better than any other spot?

Methodology
Before diving into the results, an explanation of how I did this. Ottoneu provides a Team Production table for each team (you can find this for your team by clicking the “Tools” icon in the header and clicking “Team Production” underneath that). This breaks down how each team did position by position. For roto leagues, it shows games used and all stats for that position; for points leagues it shows games used, each stat that impacts scoring, points per game and total points.

I gathered individual team data from every team in 10 FanGraphs Points leagues. These leagues were not selected completely at random, which I recognize could bias my data. I tried to identify leagues that a) had no strange rules that might impact how positions accrued points and b) were relatively competitive so stats weren’t weighed down by dead teams. This isn’t a perfect representation of the world, but it is useful.

I then added up total points for all 120 teams across each position and divided that by the total number of games played, to get a mean P/G for each position. I also found the median, minimum and maximum P/G among all the teams.

The Data
Here is a summary table outlining what I saw:

The next table is based on the 960 team/position pairs I reviewed. A team/position pair is, for example, SS production for the team Beck Panthers in league 32 or OF production for the team Avon Old Farm Winged Beavers in league 649). I get 960 pairs because I have 10 leagues in the data * 12 teams per league * 8 positions per team. I ranked every team/position pair by points per game and broke them into quartiles. The table shows count by position within each quartile.

I’ll go position by position and review what I found.

Is 1B the clear-cut strongest position?
This has traditionally been true in FanGraphs Points and other formats that value batting skills over running skills. After all, we all know that 1B is where you drop your slow, oversized, poor defensive player whose bat is too good to bench. It’s the last resort when your RF can’t run or your 3B loses his reflexes.

But by both mean and median points per game, SS was better than 1B. That’s not the whole story, though. First base does better by mean total points because teams filled out games at 1B more easily than SS. First base also outscored SS by P/G in six of the 10 leagues I looked at. In fact, 1B was the highest scoring position in four of the 10 leagues (SS was highest in three, and 3B was highest in the other three).

So while I don’t think we can say 1B is the clear-cut strongest position, it might be fair to say it is the easiest position to fill with strong production. It was first or second in points per game in eight of the 10 leagues (seven of 10 for SS, three of 10 for 3B, and two of the 10 for Util). Teams filled out more games at 1B than any other position, including C where you have a second spot to make it easier to fill out games played.

In practice, if I have a player with 1B eligibility and another position, I am likely going to prioritize using them at the other position. Historically, I would have never used them at 1B (for example, avoiding playing Kyle Schwarber at 1B when he could be in my OF), but I think that has changed as 1B has come back to the pack. In 2022, I wouldn’t hesitate to use a 1B/OF or 1B/3B or even a 1B/2B at 1B when needed, but the data suggests that finding other options at 1B should be relatively easier than at other spots.

Is this SS data legit?
Shortstop stands out as an elite offensive position, especially given the historical reputation of the position as home to glove-first, up-the-middle defenders. Not that long ago, an elite bat at SS was a rarity, but now? In 2021, among players with more than 400 PA, eight of the top 40 players by P/G are SS-eligible. Four of the top 12 are SS-eligible. And those players are all going to be SS-eligible again in 2022. And that doesn’t include Willy Adames, whose production since the move to Milwaukee would put him in that group. Nor does it include Wander Franco, who seems to be heading that direction. Not does it include Francisco Lindor who has been in that tier in the recent past.

This creates some interesting decisions for fantasy managers.

• If you have an OF/SS eligible player like Fernando Tatis Jr., are you actually better off using him in the OF than at SS or MI? I think the answer is no, you should not move Tatis to the OF as a rule. While SS had higher production than OF this year, MI had lower production than OF, by a fairly large amount. Moving Tatis to the OF means playing your second-best option at SS, and upgrading your top bench MI to your MI spot. Teams start a total of 36 MI and 60 OF. Looking at players with more than 400 PA in 2021, the average P/G of the 37th to 48th best MI (the guy you would be promoting to your lineup to replace Tatis when he goes to the OF) is 4.36. The average P/G of the 49th to 60th best OF (the guy you are benching when you move Tatis to the OF) is 4.55. So in general, you are going to be better off using your 5th best OF instead of your 4th best MI. However, the gap isn’t huge. The range of those OF is 4.46 to 4.64; for the MI it is 4.08-4.63. And your 4th best MI might be a top 30 MI while your fifth best OF might be well outside the top 60. So maybe the better answer is: it depends, but, if you have Tatis, there are almost certainly true that, at times, matchups will dictate that you are better off with him in the OF. Don’t be scared to put your SS-eligible players in the OF (or really any other spot).
• If you have a 2B/SS like Marcus Semien, should he be locked into a 2B spot instead of SS? In short, no. But that is mostly because no one should be locked into any spot if you have the flex to move them. For most teams, Semien is probably going to fit in at 2B more often than SS, because there is more depth at SS. However, if you happen to have Ozzie Albies or Max Muncy or Jose Altuve, Semien might make more sense as your SS than your 2B. The great thing about a guy like Semien is the flexibility to build a MI around him. You can lock in great production at one MI spot and then if you stumble on two great SS? Great he is your 2B. If you grab two great 2B? Fine, put him at SS.
• Should you punt SS knowing you can get good production at a lower cost? Punt is too strong a word. But I do think SS might be a position where you can wait or go cheap. We’ll have to see how things look later in the off-season, but I think you can afford to hold back from going to top prices on Tatis, Trea Turner, Semien or Corey Seager (the top four SS by P/G) if you can spend a lot less on Xander Bogaerts, Trevor Story, or even a guy like Adames and still get good production out of the position. Obviously you want to have the best production you can at every spot, but if you are forced to make trade offs due to budget (and you always are), SS isn’t a bad place to do it.

How bad is 2B?
Oh, it is not good. In half the leagues I looked at, teams got more production out of MI than they did 2B. Shortstop is so much deeper and stronger than 2B that for many teams their second best SS was better than their starting 2B. However, I think there may be some anchoring from previous seasons that is influencing this. I don’t have a great way to look this up, but I suspect managers used a lot of their 2B/SS eligible players at SS, rather than 2B, which may have been sub-optimal.

Second base lacks elite production. Only two MI scored over 7.0 P/G in 2021; one was a SS (Tatis) and the other was a SS until August, when he became a 2B/SS (Turner). Looking at the 360 team/position pairs for MI at my data (120 teams at 2B, SS, and MI), the top 30 by P/G includes 19 SS, 8 MI, and only 3 2B. The bottom 30 by P/G includes three 3 SS, 11 2B, and 16 MI. The quartile chart above shows the same – 2B is over-represented in the bottom two quartiles. Second is a spot where you risk getting really weak production if you don’t get someone solid. Having Altuve, Brandon Lowe, Muncy or another strong producer at 2B makes a huge difference when the alternatives could leave you with an ugly number at 2B.

I suspect that in 2022, more 2B/SS eligible players will be used at 2B as managers adjust to the relative depth of the positions, which will bring up the 2B scores and bring down the SS scores. I also suspect that 2B will still fall short of OF and 3B, while SS will still end up above those two.

But C is worse, right?
Oh, yes. Yes, it is worse. Very much so. By a lot. It is a half-point or more per game weaker than any other position. And despite the significant advantage teams have in filling out C games by virtue of having two spots, teams still averaged just over 156 games at C.

Of the 960 team/position pairs I looked at catcher made up 12 of the bottom 20 and 24 of the bottom 50. You can see in the quartile data above that C is a mess. Other than OF (which is pushed towards the mean because it averages 5 spots, unlike the others), C has the smallest representation in the top quartile, the smallest in the second and by far the most in the fourth quartile. That’s not great. Oh, and the number two C by P/G in 2021 just retired, so the position just got thinner.

C is alone on an island, making up an outsized portion of the bottom quartile and having very little representation in the top half. What that means to me is that if you don’t prioritize getting good C production, very likely get really bad C production. Even at 2B, if you miss out on the top guys, you can get decent production from the next tier down. But at C, if you you miss, you miss badly. You can compete if your “bad” positions are getting solid, middle-of-the-pack production. If you get basically nothing from a spot, even C, that hurts. There are a few top catchers and I expect to make them a priority in my auctions and drafts this year. Among the guys I will be targeting: Yasmani Grandal, Will Smith, Mitch Garver, Willson Contreras, and J.T. Realmuto.

How do you compare OF to the other spots?
I noted above that the quartile data does OF a disservice, but honestly all of this data does. Because the production data is by position and not lineup spot, when you see that OF is fifth in mean P/G at 5.22, that is overall team production across all five OF. On a player-by-player level, 16 of the top 30 players by P/G are OF-eligible. That includes Tatis, who we should probably leave out, since he got OF late and was primarily a SS, so it’s more like 15 of the top 30. But that is still half the elite bats and OF do not make up half an Ottoneu lineup.

What this is really telling us is that Ottoneu requires a deep OF, and the bottom end of that OF depth is pulling down the average. And there are two ideal ways (in my opinion) to handle this need for depth:

1. Spend in the OF. Especially if you save at a spot like SS, you can go out and pay up to have 4-5 OF among the top 40. It isn’t cheap to pay 2nd OF prices for your 5th OF, but it allows you to set-it-and-forget-it most days in most spots.
2. Platoon and track daily lineups closely. Maybe only count on two OF to play daily and have 7 or 8 more you can platoon based on matchups to try to maximize what you get out of them. This is where the flexibility of a guy like Tatis (or Mookie Betts, or Brandon Lowe) comes in handy. Having guys at MI or 1B or 3B that are OF-eligible, even if you primarily use them at their other position, gives basically deepens the OF pool for you. All of the sudden your second 3B or your fourth MI can also be your 6th or 7th OF.

Can 3B close the gap to the top two?
Third base settled in comfortably behind the SS/1B pairing at the top and ahead of OF (and Util) below it. And, to be honest, I don’t have a ton to say about 3B. It’s a solid position, with enough depth to go around, but not a ton more. But, as noted, 3B was the most productive position in three of the 10 leagues I reviewed.

And the 3B landscape will shift in 2022, as Alex Bregman and Manny Machado lose SS eligibility, adding two more bats to the top of the 3B list. However, the position is also losing its 2021 MVP, Vlad Guerrero, Jr. And without Vlad, there really isn’t much of an elite tier at 3B. José Ramirez is the only 3B over 6.5 P/G. Rafael Devers, Austin Riley, and Manny Machado are the only others over 5.75 P/G.

Plus, even with Machado and Bregman, there is only barely enough “good” production left to go around. Only 12 3B put up over 5.0 P/G and over 400 PA. Two of those 12 (Ty France and Eduardo Escobar) are going to be used at 2B/MI in most leagues. A full season of Alex Bregman probably helps. Wander Franco might be better used at 3B than SS for some teams.

I don’t think 3B is going to make the leap to the types of scores SS and 1B put up this year. Maybe as SS and 2B even out, 3B gains on SS? But I think it will remain third best, on average. Losing Vlad is just too much to make up.

As for strategy, don’t skimp at 3B. Every team should be able to field a solid option at the spot (either by getting a top 10 guy or using matchups and platoons and working the wire) but as you can see from the lack of players over 5 P/G, it’s a position that gets shallow quicker than you think.

What about Util?
I used to believe you had to have a strong util to compete. It was an open opportunity to add an extra 1B who mashes. But as 1B has fallen back to the pack, util appears to be more of a “fit in a guy you want in your lineup today” spot rather than a “you have to have a stud” spot. In the ten leagues I looked at, util was the second most productive position twice, third most productive once, fourth most productive three times, and fifth most productive four times. Overall, it was very close to OF, roughly the 4th or 5th most productive spot. 63 teams had their util produce in the bottom two quartiles; 57 in the top two.

On top of that, 12 of the top 14 P/G scores at Util had the same player – Shoehei Ohtani. Basically, if you didn’t have Ohtani, you probably had fine util production but nothing special. And given that Ohtani managers didn’t win every league, that means fine-but-not-special util production is sufficient to compete and win. That’s a small surprise for me and will probably change how I think about depth next year. I won’t worry about having a clear stud at util – I’ll build the best team I can and my best hitter without a position each evening will go into util, and that will be fine.

Unless I can get Ohtani.

Won’t the eligibility changes from 2021 to 2022 make all this analysis obsolete and pointless?
Boy, I hope not. This was a lot of work! Ottoneu accounted for the unusual 2020 season by allowing all players to keep position eligibility from 2019 in 2021. Basically, in a typical season, a player is eligible at any position where they had 10 games played or five games started that season or the previous one. So for 2021, that would mean in 2021 or 2020. But since 2020 was only 60 games, players kept eligibility from 2019. In theory, this made 2021 the deepest season ever for multi-position players. As a result, 2022 might look a lot different.

Regardless of how 2022 looks, I think the analysis above is useful. I tried to account for changes where I could and I don’t think the changes will massively alter production by position.

That said, stay tuned to this space – I’ll be back with a deep dive on positional eligibility changes, hopefully next week.