Is This Just Fantasy?

Fantasy baseball season is here and so ends the next 5 months of my life.

Each year following my two drafts – Mayberry (a yearly re-draft league) and Sons of Pitches (a keeper league in its second year) – I use ESPN projections to calculate how each league will theoretically play out in what has become my favourite task of the season. Its highly imperfect because it becomes out of date before the season even starts when players get picked up and dropped. Also, injuries and slumps take their toll and the projections are never super accurate. However, it does give an instant perspective on what categories each team is strong/weak on and an overall pecking order of who is expected to be the better performing teams over the course of the season.

I thought I’d pull up the draft results of 2014 and compare it with how things shook out in the league throughout the rest of that season. The long in the short of it, is that the drafting projection place finishes proved fairly accurate in Sons of Pitches and less accurate in Mayberry. For example, using the draft projections from SoP, three of the top four were forecast to finish: me (1), Burnham Beeches (3), Tequila Mockingbirds (4) – and ended the year with me (1), Burnham Beeches (2), Tequila Mockingbirds (3). Mayberry was a lot more scattered, with the likes of NC Sting Ray projected to finish 8th then coming tied for 1st.

That being said, looking at the charts below, the point projections for Mayberry seem to tally up slightly better than SoP, with less of a difference between the bars:

Mayberry 2014 – Fairly close between Projected & Actual

Inline image 1

SoP 2014 – Wider Discrepancies
Inline image 2

This actually rings true when you calculate the variance for both leagues projections vs actual:
Mayberry = -3
SoP = -8
And also the standard deviation between each individual difference compared to the average difference:
Mayberry SD = 15.3
SoP SD = 21.4
Anyway, what do these projections tell us, if anything?
1.  Moves don’t matter? The argument should be that, the more transactions made during the course of the season, the further apart the Draft Projection vs Actual score should be. Interestingly though, the number of transactions over the course of the season bore little relationship to the overall projected points. In Mayberry there were 1,317 add/drops/trades, whereas in SoP, there were 849. So we’re talking about an average of 132 moves per person in Mayberry vs 85 in SoP. Strange therefore that Mayberry should be closer to projections than SoP. Why would this be the case? Maybe waiver transactions are generally lower impact than drafts -> most of the guys you pick up or drop tend to be drafted in the later rounds and wont have the same impact as a Mike Trout or Clayton Kershaw…? Best explanation I can come up with.
2. Inactive managers skew the results a lot. Especially looking at SoP, where Team Joe Wahl and London Towners grossly underperform their projected score as their owners gave up/stopped playing throughout the year. The zero sum nature of the scoring system means that the discrepancies are exacerbated once people give up, because there are more points to go around to other league managers.
3. Fantasy talent/knowledge rises to the top. Managers in SoP were able to take strong drafts and maintain top finishes throughout the year. In Mayberry, I think the talent was more evenly distributed and you can’t keep a guy like NC Sting Ray down. He outperformed his projections by 22.5 points,
So what have we learned….probably not a lot. Conclusion is: Be really good at fantasy baseball and try not to have inactive managers in your league.
Perhaps a better study would be individual categories for the draft and actual results.
Sons of Pitches Projected Place Actual Place Mayberry Projected Place Actual Place
Clown Question Bros 1 1 Chudley Cannons 1 3
Team Joe Wahl 2 8 Clown Question Bros 2 5=
Burnham Beeches 3 2 Timonium Centipedes 3 7
Tequilla Mockingbirds 4 3 Blind Your Ponies 4 8
Boston Navigators 5= 9 Jacksonville NC Hot Shots 5 10
London Towners 5= 10 Windsor Castles 6 1=
I Hate Zack Greinke 7 6 Kentucky Beerbellies 7 5=
Joe Buck Yourself 8 7 NC Sting Ray 8 1=
Out of Options 9 4 OBX Barnacles 9 4
Last Man Stanton 10 5 Pikesville Rhinos 9 9
For this year, the projected Mayberry league is expected to be:

SCORE

PARK 61
MCMILLAN 58
GRANT 67
SAUN 61
LMMP 62
BARNEY 17
KENTUCKY 62
ELEPHANT 54
MILLER 51
CHIEF 55

Initial thoughts – Barney’s Tigers Blue Tiger had one of the worst drafts I’ve ever seen. This became clear as the draft was playing out (Cabrera at #2 overall, Kimbrel in the 2nd round, etc) but only becomes more obvious when you tally up his projected score to…17! Christ.

Another successful draft for me – I swear I don’t play with the numbers or anything. I think this was largely down to my focus on Batting – as I have 3 batters on my bench, more than anyone else, which gave me good scores for Runs, HRs and RBIs.

I’m going to be scrounging the waiver wire for Saves. Don’t know where all the closers went during the draft but have to hope for some serious churn in depth charts to snag saves. I’m also in dire need of Ks and Ws since most of my pitching staff is injured to start the year, which is far from ideal. Not enough research by me, rookie mistake.

Sons of Pitches is more difficult to project as the bench in our keeper league has been expanded to 6 spots and this kind of throws off the projections with prospects maybe not even getting to play this year. So I have decided not to redo the task for this league. Also, I’m about to repeat as Champion so who needs projections anyway?

Standard

Leave a comment