The Latest Updates to the FanGraphs Auction Calculator

Fans of the ‘Graphs are probably familiar with our Auction Calculator. It’s an intuitive tool, especially for standard leagues, but there are some quirks. If this is your first time hearing about the calculator, I recommend skimming this tutorial. I’ve also penned an advanced version of the tutorial which includes tips and tricks for converting the auction values to snake draft tiers.

Recently, we added two big updates to the calculator. I also sat down with site owner/dark overlord David Appelman to talk about some of the assumptions made by the calculator and how to get around them. For the remainder of the article, I assume you have some experience with using the auction calculator. If you’re new to the tool, I recommend you familiarize yourself with it before reading on.

Updates and Hacks

Undoubtedly, you’ve noticed the first update – we’ve uploaded 2016 end of season values as a stat source. We’re using this in our End of Season Rankings series. This is also a great way to test your strategies against what actually happened in your personal leagues. After all, that’s why the calculator is so highly customizable.

As noted in that linked article about catchers, the calculator does have one notable weakness when grading past production. Partial seasons can cause some confusion. Gary Sanchez makes for an easy example. In a 12-team, two-catcher league, he produced $8.60 of value according to the calculator. However, that’s assuming you only used Sanchez. In reality, you probably got a few dollars of production from a Cameron Rupp or Derek Norris.

The other update is a much need minimum innings parameter. Past tutorials have focused on “tricking” the calculator into producing the right number of positive value innings. Let’s say your league may have two starters, two relievers, and five generic pitcher slots. Inputting those values produced misleading results. Often, the calculator believed it was most efficient for every team to roster 1,000 or fewer innings. If your league has a 1,500 innings cap, then rostering 1,000 innings is the same as punting wins and strikeouts.

The minimum innings option helps to address this issue, but you’ll still need to make some manual adjustments before the calc spits out relevant prices. Let’s take a peek behind the curtain. The calculator is fairly conservative. A starting pitcher is assumed to be 150 innings. A reliever is assumed to be 50 innings. If you enter six starters and three relievers, it will assume everyone is rostering 1,050 innings (6*150+3*50). For the minimum IP setting to work, only the generic pitcher slots can be optimized by the calculator. This is important: if you specify the type of pitchers to be used, the minimum IP setting is ignored.

In the past, the solution was to list more pitchers than your league allows you to start. For example, if I know I’m going to roster seven starters and four relievers for nine starting spots, I would list the 11 pitchers and take two spots away from my bench tally. If you know the exact distribution of your pitchers, use starter and reliever slots. Don’t worry about the minimum IP. If you want some flexibility by using the minimum IP setting, I recommend a distribution similar to five starters, two relievers, and five unspecified pitchers.

Remember, starters equal 150 innings and relievers equal 50 innings in the calculator. If you input 1450 IP and nine generic pitchers, the calculator will tell you to roster only starters and a handful of elite relievers (at low prices). If your league has a high inning cap, you may have to list more pitchers than you’ll actually roster to get a realistic starter/reliever price split.

As an extension of hacking the pitcher values, I learned from Appelman exactly how the bench slots function. Bench players are all assumed to be $1 or less players. Basically, they’re replacement level. The bench does not specifically distinguish between pitchers and hitters. For example, let’s say you have a one-team league with 20 starters and five bench spots. The 21st to 25th best players will be listed at ~$1 regardless of position.

Therefore, if you set a minimum IP threshold that can’t be reached by the number of pitchers you allocated, the calculator won’t automagically convert bench players into pitchers. It’ll just leave you short in the innings category. Of course, you can always draft $1 pitchers at that point – it’s just that the calculator may not tell you to do so.

To produce prices that match your preferences, the number of bench slots you list should equal the number of $1 players you plan to roster. If you’re a stars and scrubs guy, you may want to list some of your starting positions as bench. Conversely, if you like quality depth on your bench, you should create a couple extra placeholder starting spots. This is basically what we did to optimize the pitcher values.

Guidance, Not Rule

The auction calculator is a one-size-fits-all tool. It’s designed to offer guidance rather than a precise price sheet. I use the calculator to give me some idea of the cost of each asset class. I then adjust to the actual conditions in the draft. If Mike Trout should be about $55 but actually goes for $70, I know I’ll need to overpay for a couple early picks or else I’ll be left with too much cash at the end of the day.





You can follow me on twitter @BaseballATeam

3 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
tomjefmember
7 years ago

Too bad you didnt add a HD/SV category!

tramps like us
7 years ago
Reply to  tomjef

Ditto. Sure changes values of closers

tomjefmember
7 years ago
Reply to  tramps like us

For reasons I dont understand, I found that if I added both HD and SV (making it a 5×6 with my league parameters), it came pretty close to reasonable values.