Introduction:

Today, I will introduce a game theory approach for comparing baseball projection systems. The day’s venture will not be a typical statistical analysis. I won’t be using any Chi-squared tests, nor will I calculate Type I or Type II errors. I won’t be evaluating MSEs or the like.

Instead, I will look to determine the profitability potential of each projection system by simulating what would have happened in a fantasy auction draft. Instead, I’ll play a game.

What do I mean by this?

Read the rest of this entry »

The MLB playoffs are now upon us. We’ve had back-to-back one-game division title games! We’ve got a statcast broadcast! Hey … even hugs are a-plenty this postseason!

Just a reminder … I am recapping my bold predictions for 2018. You won’t see anything like “Giancarlo Stanton will hit at least 25 homeruns” – that would have been too easy a prediction. Sure, I could have filled up my list in March with much more likely calls to boast that “Ariel Cohen got 50% of his predictions right!” But that isn’t the point here. I don’t expect to get most of these correct.

The aim of the exercise is to choose a few unlikely outcomes, yet achievable upsides (or downsides) – so that the analyst can highlight certain players. My general rule for bold predictions is to target somewhere between a player’s 70^th & 90^th range of percentile possible outcomes, or in other words, predictions which are about 10-30% likely.

Read the rest of this entry »

The date is March 19, 2018. You are about to compete in a live NFBC fantasy baseball auction. You are prepping vigorously for your draft auction. You are reading over your player lists, mulling over which OFs you will go for at the auction table.

Fast forward to September 27, 2018. Now that the season is almost all in the books, you can now look back at your fantasy auction and see all the good, the great, the bad and the terrible decisions you had made just 6 months prior.

Let’s start with a simple OF decision. Which player should you have bought back in March?

Mike Trout (OF, LAA)

OR

Eddie Rosario (OF, MIN)

Sounds like a fairly easy decision, no?  But to help you out, before you answer the question … I’ll provide you with some 2018 statistics (as of 9/26):

Don’t answer the question just yet … let me also provide you with a full season dollar valuation for the two players in question.

For those new to subtraction, Trout’s valuation is larger than Rosario’s valuation by $15.30, or another way to look at it … Trout was worth almost 70% more than Rosario was in 2018.

So which player of the two should you have purchased for your fantasy team on March 19?

The answer is ……  Eddie Rosario!!!

Wait just a minute … Trout had the same number of RBIs, but more Runs, more SBs, a higher BA, and more SBs than Rosario. You may even ask, “Ariel, didn’t you just tell me that Trout was worth 70% more than Rosario?” How can that be???

Well, to answer that – you need one more piece of information, namely, what the price was to acquire each player.

First, let’s define three quantities:

• $Value – The full season 5×5 roto value of each player. For this, I am using FanGraph’s auction calculator on YTD 2018 stats, with NFBC standard settings (15 teams, mixed AL/NL, $260 budget and positions – 9 P, 2 C, 1B, 2B, 3B, SS, CI, MI, 5 OF, U). This represents what a player was actually worth in 2018.
• $AAV – The average auction value for each player in 5×5 roto / NFBC style format. For this, I am using the average of a set of actual NFBC online auctions run by Andy Saxton that Todd Zola of Mastersball had provided. This will represent the cost that it would have taken to acquire a player back in March. For those players who weren’t drafted, or who were only drafted as a reserve, we will set a nominal price for them of $0.10.
• $Bargain – The difference between the $Value and $AAV. This represents the profit that each player had provided over the 2018 season, from his initial pre-season draft price.

Let’s go back to that OF comparison now …

Winning fantasy baseball is All. About. Value.

Although Mike Trout will finish the season as a top-10 performer, if you had purchased him in a fantasy auction this year, you would have paid some $47 – which meant that you LOST MONEY on him. Had you purchased Eddie Rosario, which would only have cost you about $12, you would have profited … by over 75% of what you paid.

Below are the top 20 most profitable players for 2018:

Assorted Player Notes & Facts (in no particular order):

• Blake Snell was clearly the best pitcher to buy pre-season 2018 at a roto auction. For a mere $7 average auction value, he will finish as the 3^rd most valuable pitcher this season worth $36, behind deGrom ($40 value) and Scherzer ($39 value). His 21 W currently leads all of MLB, his 1.90 ERA leads the AL, and add in a healthy 211 Ks too (only 14 pitchers have more than 200 strikeouts).
• The next most profitable starting pitchers were Mike Foltynewicz ($15 bargain), Walker Buehler ($14 bargain) and Miles Mikolas ($13 bargain). [Does Mikolas remind you at all of Colby Lewis?]
• Blake Treinen was clearly the most profitable RP in roto this season, with 37 saves and spectacular roto ratio stats (0.79 ERA / 0.83 WHIP). He even added nearly 100 Ks to boot. He was the 16^th most expensive closer to purchase pre-season at an AAV of under $10.
• Yasmani Grandal ($19 value) and Yan Gomes ($13 value) were the two most profitable catchers this year. Grandal could be had for only $4 AAV, and Gomes’s AAV was less than $1.
• The three most valuable undrafted players were Jesus Aguilar ($23 value), Miguel Andujar ($22 value) and Nick Markakis ($19 value). Of those three, Aguilar and Andujar were high skilled players not assured of playing time – which is why their draft price was so low. Javier Baez’s price was also depressed, because most projection systems didn’t give him enough plate appearances, with the Cubs’ logjammed IF roster on opening day.
• Nick Markakis on the other hand, was simply a mistake by pre-season drafters. Markakis was handed the cleanup spot on day #1 in Atlanta. Surrounding him the lineup was Inciarte, Albies, Freeman, and Acuna (knowing he would come up to the MLB shortly). They all cost a minimum of $12 in AAV, which suggested a good ATL lineup this season. Given who the players surrounding him were in his lineup, Markakis, who has a history of a high batting average – should have been projected for a much larger R and RBI total. At the very least, he should have been drafted in NFBC formats.
• Of the top 20 bargain players, Christian Yelich was the most valuable returning a $43 full season value. Yelich had an MVP-caliber season amassing a 33/104/.321 campaign with 21 swipes. He was also the most expensive player on this list to buy, with an AAV of $24. Javier Baez is the next closest to Yelich in pre-season cost at an AAV of $11.50. The huge bargains this year typically came from players which cost just $2-8 pre-season.
• Of the top 20 highest priced players pre-season only 1 player returned a profit – Mookie Betts ($47 value / $39 AAV). The other 19 highest priced players averaged a loss of $16 (A $Bargain of -$16).
• Jacob deGrom ($40 value / $29 AAV) was the highest priced pitcher pre-season who turned a profit. Only deGrom, Justin Verlander and Aaron Nola cost over $20, yet turned a profit. To note, pitching this season accounted for about 37% of NFBC auction budgets, which is rather high historically, and much higher than the traditional 70/30 split rule would indicate.
• The average cost of the top 20 bargain players (including undrafted players at $0.10) was $5. The average returned value of the players was close to $25.

Opening day is now just one short day away. The days this week seem to creep by ever so slowly for a baseball lover. We have already had our fantasy drafts and auctions, and now we want to watch/see how our teams will perform, both fantasy and in real life. Baseball is back … and this year, we get to watch real baseball games that start to count in March!

Below are my 2018 bold predictions. Some of them stem directly from a small stretch of the ATC Projections, which can be found here on FanGraphs. Others come from my own personal analysis on the player, or team situation. The rest arise from some sheer wild optimism, but I have convinced myself that it could happen.

Read the rest of this entry »