Ottoneu Tiered Rankings Intro

I have done tiered Ottoneu rankings for FanGraphs Points leagues for a few years now, and every year I write a version of the same intro over and over again. This year, I am just going to write this up, link back to it, and then for the intro to each ranking, I can talk a bit about the position overall. For today, no rankings just yet. Just a little discussion of methodology and other such things. Let’s do a little FAQ.

Why are your rankings for FanGraphs Points leagues?

That’s the most popular format, even if it’s not (in my opinion) the best (if you aren’t playing 4×4, you should be). It’s also pretty applicable to most other formats. If you want to convert from FGPTs to SABR Points, basically all you have to do is downgrade pitching a bit vs. hitting. Hitting scoring is exactly the same in both points formats. For pitchers, you do want to downgrade high walk and/or low K pitchers. Innings are worth less and avoiding hits isn’t a differentiator because hits don’t cost pitchers points.

If you want to convert from FGPTs to 4×4, you need to assume a shallower value curve (the top players are not worth as much and the dollar values are spread more evenly. You’ll also want to boost your values for HR-suppressing pitchers and for high-OBP hitters. 5×5 is a different beast but there are a million 5×5 rankings out there that you can use.

Do I need to adjust for H2H leagues?

Yes, a bit. In H2H leagues, without an IP cap, pitchers who go deeper into games are more valuable. So boost up anyone you think will be going into the 7th often. And I downgrade catchers, as well. While season-long Ottoneu leagues allow you to start two catchers per-day (while still holding you to 162 total games at the position), H2H leagues only have one C spot. I choose not to roster two C most of the time, because when my starter gets Sunday afternoon off, there is a decent chance the backup does, too. So I just get less value from that spot, which causes me to downgrade C values overall.

Do you use projections for your rankings?

Yes and no. My rankings are a bit art and a bit science. The science bit comes from pulling values from a variety of projections and sources into a single spreadsheet to inform my thinking. This includes Steamer, ATC, ZiPS, THE BAT, and mock drafts from the Ottoneu Slack Community. The idea is to have a wide-range of values because different sources value players differently. This gives me a sense of the range a player falls in.

So what is the “art” portion of your rankings?

With those data sources in place, I do an initial ranking of players based on those inputs. I don’t simply sort by an average or something like that. I move players up or down based on the range, what I expect from the player at first-glance, etc. Then I go back, position-by-position, and look at each individual player. This is when I write the quick one-line notes on each player and when I adjust rankings based on things I learn, such as a player playing through injury, adjusting their swing, learning a new pitch, etc.

Why tiers and what do the tiers mean?

While I include ordinal rankings for each position, the tiers are what I really focus on. If you like a guy at the bottom of a tier more than a guy at the top of that tier, I probably don’t disagree strongly. It’s both difficult and, to be honest, a fool’s errand to pretend I can draw hard and fast lines between each player. So I don’t try.

This year, I am doing something new with the tiers, which is setting a single set of value buckets and using it for all positions. Every position will have players in tiers of $0, $0-$1, $1-$2, $3-$5, $6-$9, $10-$14, $15-$20, $21-$27, $28-$35, $36-$44, $45-$54, $55-$65. In the past, I have ranked the players and then tried to draw lines between tiers. This year, I ranked the players and forced myself to take a stand on a player who landed between tiers. If I think a guy is probably $44 or $45, the exercise of determining “am I happier to win him at $50 or lose him at $45” helped me make smarter decisions (I think).

By the way, those first three tiers overlap because I realized that in the past I was drawing too clear a distinction between $1 and $0. There are a bunch of $1-$2 players that I like and think should be rostered cheap. There are a bunch of $0 players I think have no value and don’t belong on a roster. There are also a bunch of players in the middle. So here is how I interpret those tiers:

• $0 – I won’t roster this player
• $0-$1 – I probably won’t roster this player, but if the right circumstances align, I can see picking them up to fill a positional need, to gamble on upside, etc.
• $1-$2 – These players probably should be rostered.

Should I use these for my first-year auction or for future auctions or both?

These are designed for first-year leagues, because it isn’t possible to account for inflation for every league at the same time. If you expect 30% inflation in your 5th year league, these tiers are a good starting point, but the dollar values are going to be too low for you. Make sure to adjust for inflation.

How do you handle multi-position players?

I rank every player at one spot – the spot where I think they are most valuable. I also rank 2B and SS together as a single ranking because a) so many players qualify at both and b) with the MI spot, I find that neither position is meaningfully more valuable than the other. So any player who qualifies at C is ranked at C. Any player not ranked at C but with MI eligibility will be ranked at MI. Then I’ll rank OF, then 3B, and finally 1B/Util.

Do you boost player values if they have multi-position eligibility?

Yes, a small amount. I don’t generally think being eligible at more than one position is worth that much but it can help with roster construction and I can see spending an extra dollar on a player because of it. But that depends on the position. Yainer Diaz doesn’t get any credit for 1B eligibility – you are never going to use him there and if you do, something has gone wrong. Same with 2B/1B-eligible Brandon Drury. 1B eligibility doesn’t matter to me in that context.

The overlaps that do matter are those between MI, OF and 3B. I will boost Mookie Betts for being 2B/SS/OF. I will boost Christopher Morel for being 2B/OF and Nick Senzel for being 3B/OF. I’ll even boost Vaughn Grissom and Thairo Estrada for being 2B/SS.

Who do you include in your rankings?

I started with the entire Ottoneu player universe and then removed anyone who met ALL of the following criteria:

• Rostered in less than 3% of leagues
• Had a $0 projection from Steamer based on both Justin Vibber’s Surplus Calculator AND Adam Kaufman’s Value Calculator.
• Were not selected in any of the three mock drafts the Ottoneu slack community had run at the time

I have since removed a few players who retired or signed overseas, but I currently have 853 players in my sheet.

When will we get to see the rankings?

Starting tomorrow, I will be rolling them out position by position. I’ll have a full ranking list up early next week.