FanPost

The first annual VEB Free Agent Prediction contest


Hi! It’s been many years since I last made a fanpost here, but this was too big to put into a comment. I think it’s easier to have this info in one place anyway. As the title suggests, I had the idea to do some type of a friendly free agent prediction competition for the VEB community this offseason. After asking a couple times in the comments, there seemed to be enough interest to make it worth the time and effort to put it together. What follows is how to participate in this competition and an explanation of how the scoring works:

How to participate

Here is a spreadsheet listing the 55 free agents that we will be predicting. This is everyone listed on either Fangraphs’ top 50 free agents or MLB Trade Rumors’ top 50 free agents. I went through Fangraphs’ list first and then I found 5 players on MLB Trade Rumors’ list that were not in Fangraphs’ list: Corey Knebel, Eddie Rosario, Hector Neris, Ryan Tempera, and Danny Duffy. If I missed someone, it’s too late now.

For each player, put your guesses for how much the player signs for, how many years they sign for, and three picks for guesses of who you think they’ll sign with. Scoring will be covered below, so make sure to take that into consideration for strategy purposes.

To fill it out, you’ll first have to either copy this template as a google sheet or download it as an excel sheet. To copy it, go to File -> Make a copy. To download it as an excel sheet, go to File -> Download -> Microsoft Excel. After filling it out, you have a couple different options:


If you’re using google sheets, you can share your predictions spreadsheet with me by doing the following:

Hit the "Share" button in the top-right corner. Put my email (bamarkham1@gmail.com) in the "Share with people and groups" section. Afterwards, post a top-level comment in this fanpost letting me know that you sent me your predictions. I can then reply to that comment either confirming that I received them or telling you I haven’t received them. You don’t have to put your email in the comment as long as you put your username is in the appropriate place in the spreadsheet.


Note that even if you filled it out as an excel sheet, you can upload that spreadsheet to google drive and then share it to me using those above steps. Otherwise, you can share your predictions spreadsheet by doing the following:

Attach it in an email and send that email to me (bamarkham1@gmail.com). Then make sure to post a top-level comment in this fanpost telling me that you emailed me predictions. I can then reply to that comment confirming I got them or letting you know I didn’t.

Anyway, I will be writing a python script to score these (otherwise it would not be feasible to score them all), so please try to fill it out as intended. I will try to look at each submitted spreadsheet and make sure there’s no issues, but I would also appreciate it if everyone can do their part to make sure their spreadsheet is filled out correctly. Make sure to put your VEB username on the first line so I know who you are.

In the "Total guarantee" column, please specify the total in terms of millions of dollars. You can put "$40.5m" as your guess, or "40.5", "$40.5", or "40.5m", but please don’t put "40,500,000" or anything else with that many numerical digits. From context a human will know what you mean, but software is generally not good at recognizing context, and writing something that could would take way too much effort on my part.

For the team predictions, please use the team’s name without any abbreviations (so "Diamondbacks" rather than "Dbacks", "Athletics" instead of "A’s", "Yankees" rather than "NYY", etc.

Scoring

Your score for each player’s prediction will be scored out of 40 points. More is better, so a perfect prediction will get you 40 points.

You will be scored on three categories:

  • Total guarantee of the contract
  • Average Annual Value of the contract
  • Which team the players signs with

Here is an explanation for how the scoring is calculated for each category:

Total guarantee: 20 points possible

Half of those 40 possible points come from guessing the total guarantee. If you guess the total guarantee exactly, you’ll get 20 points for this section. It’ll be calculated like so:

20 x (1 - ( | actual total guarantee - predicted total guarantee | / actual total guarantee ) )

In other words, I’ll take the absolute value of the difference between what the player signed for and what they ended up signing for. Then I’ll divide that by what they signed for to get the percentage difference between your guess and what they signed for. Since a smaller difference is better and that difference will be expressed as a fraction, I’ll take 1 minus that number. Then I’ll multiply that by 20 to get your final score.

If you were off by more than 100% (for instance, if you guessed someone would sign for $100m and they ended up signing for $40m), then your percent difference will be capped at 100%, meaning you’ll get 0 points for the total guarantee, but your score will not go negative.

The amount used for the actual total guarantee will not include team options, but will include player options and years that are guaranteed following an opt-out. They will not include mutual options, team options, or vesting options. Any guaranteed buy-outs will be included however.

Average Annual Value (AAV): 10 points possible

You’ll notice that the sheet you get doesn’t ask you to give the AAV of the deal, it’ll be calculated by taking the total guarantee of the deal divided by the years. I think it makes more sense to score based on the AAV instead of the years because then a person doesn’t get knocked twice when they predict a player to sign a long-term deal but that player instead settles for a one year deal at a high AAV so he can re-enter free agency again the following year.

The formula is the same as above but uses AAV in place of the total guarantee and 10 as the multiplier rather than 20:

10 x (1 - ( | actual AAV - predicted AAV| / actualAAV) )

Similar to the total guarantee, the number of the years used to calculate the AAV will only be years that are guaranteed, including player options and opt-outs. Years involving team options, mutual options, or vesting options will not be included.

Also similar to the total guarantee, your percent difference will be capped at 100%, so if you guess say, a $20m AAV and the player signs for an $8m AAV, you’ll end up with a score of 0 for that prediction.

Team picks: 10 points

You can guess up to 3 teams that the player signs with. On this one, calculating your score requires no math: If your first pick ends up being correct, you get 10 points. If your second pick is correct, you get 5 points. If your third choice is correct, you get 3 points.

The deadline to turn this in and other details

Predicting the contract details of 55 players takes a lot of time, so I’m going to give everyone a little over a week to make them all and get their predictions submitted to me. That deadline will be Monday, November 22nd at 10pm Central Time. Any players that are included on this spreadsheet but sign before that deadline will not be scored, only free agents that sign after that date will be used.

We will use MLB trade rumors as the arbiter of this, because they have a very good track record of being correct on a signing once they announce it. If MLBTR releases an article with language similar to "Player X to sign with Team W for contract Z" prior to the deadline that will be counted as a signing prior to the deadline. After the deadline however, we will wait for a more official announcement of the signing before grading predictions for that player.

This competition does not cost any money to participate in and does not pay out any money for winning or being one of the best predictors. This is just for bragging rights among the VEB community. Feel free to do that among yourselves though in any manner you want. Perhaps if this is a hit, we can continue it next year but with some type of awards for the top performers. VEB, SB Nation, and Vox are not actually involved in this at all, I am organizing this on my own. Let me know in the comments how you would like to improve this competition next year, assuming we do it again. Also let me know if anything in this post is unclear and you are unsure how to do anything regarding filling out or turning in your predictions. Otherwise, start making some predictions! I will try to post everyone's predictions in a Google spreadsheet on Tuesday, November 23rd, the day after the deadline for turning them in.