Using the Auction Calculator in 2022: A Beginner’s Guide

I’ll admit that the first time I looked at the auction calculator I was overwhelmed and confused. What do all those numbers mean? How come some of them have parentheses around them? What’s an aPOS? I use a snake draft in my league, not an auction. Why do I need dollar values? Well my friend, if you have asked yourself those questions I hope to answer some of them in this post and prove to you that our auction calculator is a very valuable tool whether you play in a 10-team ESPN home league or you’re an NFBC regular.

—

An index of terms

ADP – Average Draft Position pulled every day from NFBC draft data, reflecting the past 30 days.

PA – Plate Appearance projections (depending on the projection system you have chosen). Not used in ranking calculations, but very necessary for all formats.

mAVG, mRBI, mR, mSB, and mHR – Z-Scores (also know as Standard Deviation Formula) by category. This is the calculation that compares player projections across the player pool to see how they compare. For example, Salvador Perez’s RBI projection is compared to every other player’s RBI projection in a Z-score calculation that looks like this:

Salvador Perez’s mRBI = (Salvador Perez’s projected RBI production – Average of the player pool projected RBI production) / Standard deviation of the player pool projected RBI Production.

or, more simply put:

mRBI = (player RBI – player pool average RBI) / player pool RBI standard deviation.

The Player Pool – The players who are considered when calculating averages and standard deviations, necessary for the z-score calculation. How do you get a good player pool? More on that later.

PTS – Sum of the mAVG, mRBI, mR, mSB, and mHR scores.

aPOS – Positional adjustment, taken as the last rosterable position player’s PTS. This is easy to see in a 12-team, 2 catcher league (the default settings of the calculator) Go here, and notice that Jacob Stallings is the 24th best catcher and therefore the aPOS for catchers (with these settings) is the same as Jacob Stallings PTS, just multiplied by -1. This effectively makes Stallings worth $0, but you can’t get him for free in a draft. Therefore, Stallings is valued at $1 because he’s worth rostering with these league settings, but not at any more than the minimum $1 bid in auction leagues. Non-auction leaguers can see this as simply, don’t draft anyone listed below Stallings.

Dollars – Sum of PTS, aPOS, and a $1 adjustment for the minimum bid allowed on a player. Since we added $1 to Jacob Stallings’ value, we need to add $1 to everyone else’s.

—

What are z-scores and who is included in the player pool?

The player pool is possibly the most important piece to consider and the reason that you may come to a different mRBI, for example, if you try to calculate a Z-score by yourself. You may have noticed above that Perez is being compared to “every other” player. But how do you determine how many players he should be compared against? Well, how many catchers are rostered in your league? How many roster slots need to be filled by managers in your league? That’s part of what helps determine the player pool. But, the other part is how do we determine who the top catchers are to be included in our player pool when we have not yet ranked them? Our auction calculator optimizes this process by randomly sampling and then iteratively re-sampling players to be selected in the player pool until the best players have been selected, all within one simple click of the “Generate Projections” button. That’s pretty cool and something I did not know about before doing the research to write this piece. In my first piece on Z-scores, written a few weeks ago, I was simply trying to rank shortstops against other shortstops and I hard selected the player pool using projected WAR. Some took offense to such a wild and crazy idea, but I was simply looking to compare a single position and use that exercise to demonstrate the Z-score calculation. But, that would be difficult to do for the entire player pool and the calculator optimizes that process for you.

Let’s take a stab at calculating Perez’s mRBI as I laid out above and see how close we get. Remember, we won’t be able to be precise because our averages/standard devs are going to be slightly different than the calculator, since we can’t optimize in the same way. Perez is projected by steamer to accumulate 102 RBI in 2022. If we use a player pool that matches the calculator’s default settings (14 batters plus 6 out of the 10 bench spots designated for hitters with 12-teams in the league):

we can approximate the player pool at 240 (12 teams * 20 roster spots). I took a random sample of all the players in steamers projections to get those 240 players. This is not a good way to do it because steamer projects a lot of players who you would likely not include in your player pool if you were making it by hand. But, for the purpose of this demonstration, it will work. I then took the average of RBI projections and got 9 with a standard deviation of 23.37. Now, we can calculate Perez’s mRBI or RBI Z-Score as:

mRBI = (102 – 9) / 23.37 = 3.97

When we load up the calculator and check on Perez, he’s listed with an mRBI of 4.4. I got close by randomly sampling the player pool, but the calculator’s optimization method does a better job and that is the reason for the difference, close though it may be.

Now what you have is a way to look at category scores by position. Clicking over to the catcher’s tab, we see that Salvador Perez is by far the best-projected RBI-producing catcher and the best overall catcher from a points perspective (PTS) at $7.0. But, the only other catcher with a positive mRBI (the parentheses reflect negative values) is J.T. Realmuto at $0.4. Does that mean all the other catchers are harming your RBI totals? One of the quirks of Z-scores is that players will have negative values, representing projections below the average of the group. In this case, you can look at Perez and Realmuto as the only catchers who have above-average RBI projections within the player pool.

Summing up all of his scores (mR, mRBI, mHR, mSB, mAVG), adding his catcher player adjustment (aPOS), and a $1 increase, Salvador Perez gets bumped up to $33.7. But now it’s time to point out something tricky. Due to that big aPOS adjustment, when you click back over to look at all batters together, you see that Perez finds himself in the top 12 picks when we sort by Dollars. In fact, he’s the fourth most valuable hitter overall. Does that make sense? Well, if I have the 4th pick in a 12-team draft, I’m not taking Salvador Perez with my first-round pick even if he is projected for 617 PAs. Our friend Ariel Cohen wrote about this exact issue back in 2019 and his article gives some excellent suggestions on how to handle this conundrum. Next, I’ll discuss how you can adjust these ranks to your preference.

—

How does this apply to my league?

This part will be short. Don’t skip over the league adjustments at the top. Are you in an OBP league instead of AVG? Then, make those adjustments so that you are appropriately valuing players. Are your roster settings different? Maybe you don’t have 10 bench spots, so make sure your inputs are exactly like or as close to your league settings as possible. Remember, that’s what determines the player pool. Lastly, if you’re in a points league, keeper league, or AL/NL league, the settings of the calculator can be customized. How about an Ottoneu league? The calculator has that all worked out for you. Just click on the “Preset” that matches your Ottoneu style.

—

Finally, what’s the best way to use this?

Let me now, with an image, show you the most important aspect of the FanGraphs Auction Calculator:

This export will give you a jumping-off point for your personalized rankings. If you’re in an ESPN league and your league mates are using ESPN ADP as their ranking system, trust me, you’ll be better off. One of my favorite things to do is to sit with my rankings list, sip a cup o’ joe, listen to Paul and Justin debate rankings on The Sleeper and the Bust, and alter mine as I listen. Some will argue that this defies the purpose of comparing players because when you add weights or change value arbitrarily, your rankings are no longer based on standard deviations. But if all I’m really after is a ranking system, then I don’t see the issue with this. If I’m in an auction league and I’m pinpointing value, I may want to be a little more cautious. Each manager has to find what works for them.

Finally, for those who understand that drafting is more about roster construction than anything else, you can sort your output table by stolen base value, by home run value, by projected PAs, etc., When you’re in the 8th round and realize you are light on stolen bases, you can sort by mSB to see who is left with the most SB potential value.

—

My goal was to write a beginner’s guide to the auction calculator and now I fear I may have taken it a little too far. There’s always more to learn and the RotoGraphs crew plans on putting more out there for advanced users. Know that the calculator’s value comes from the optimization of the player pool according to your league settings and the exporting of sortable ranks so that when round 6, pick 3 comes up with your name on the clock, you’ll be able to value your players accordingly and pick the best player available.

—

FAQs:

• What does altering the position priority setting do?

Notice that the default settings prioritize C over SS over 2B. In this case, multi-position eligible Trea Turner (2B/SS) is treated like a SS and is given a SS aPOS. If you change that so the priorities are listed as C over 2B over SS, then Trea Turner is counted as a 2B and is given a 2B aPOS. One important thing to note is that if you prioritize OF over C, most of the C eligible players will still be counted as C. One specific example of this can be seen with Dalton Varsho. He is both C and OF eligible, but he ends up being a replacement-level OF and is needed to fill out the C pool. So, Varsho is always treated as a C.

• What is experimental mode?

If you really like a player but are worried about their PA projections, “Experimental Mode” will fill in stats for any player with less than 600 projected PAs with replacement level stats. This way you can get a sense of what that player will produce if they are coupled with a waiver wire pickup to fill out the missing PAs.

• What are some other ways to value players?

Two other common valuation formulas are Percentage Value Method (PVM) and Standing Gains Points (SGP). Like Z-scores, each has pros and cons. I recommend reading Jeff Zimmerman and Tanner Bell’s The Process for an in-depth look at SGP.

• More Questions?

Add them into the comment section and I’ll do my best to add them into the FAQs.