Closing the book on 2016 predictions: Each preseason I rank-order
players at all six core offensive positions.
At season's end I examine which of my top 10 QBs/TEs/Ks/DST finished
among the top 10 scorers at their respective position, and which of my top 20
RBs/WRs finished within their respective top 20. Then I compare these results with those of
ESPN, CBS Sports, and Yahoo! Sports.
It’s not overly scientific. For
example, my preseason 22nd, 23rd, 24th, and 25th
ranked RBs turned out to be top 20 scorers.
For this annual study, close doesn’t count, so none of those predictions
(including ranking 40-something ADP RBs like Tevin Coleman and Isaiah Crowell
just outside the top 20) factor in to my final assessment. Too bad for me, but over multiple years across
various positions, these things will mostly even out.
Here’s the accuracy percentage for my site, compared to the final
preseason rankings on ESPN, CBS Sports, and Yahoo! Sports:
FF4W: 54%
ESPN: 55%
CBS Sports: 50%
Yahoo! Sports: 55%
And for context, here are the results for the other two seasons I’ve
tracked this:
2015:
FF4W: 44%
ESPN: 48%
CBS Sports: 51%
Yahoo! Sports: 54%
2014:
FF4W: 58%
ESPN: 48%
CBS Sports: 51%
Yahoo! Sports: 49%
So over this three-year period, Yahoo! Sports has led the way with 54%
accuracy. FF4W is second at 52%. And CBS Sports and ESPN are at 51% and 50%,
respectively.
Group Think—one of the main problems with fantasy sports
prognostications—has something to do with how close these results have
been. Almost everyone is getting their news from the same sources, or the sources of those sources. Where FF4W has gained a slight edge over two of these three has been in taking more chances. And that can make a small but tangible difference to fantasy managers.
Maybe 52% of the time, these chances pay off. And that can make a small but tangible
difference to fantasy managers.
For example, this past preseason FF4W picked more top 10 fantasy
scoring QBs (six) than any of the other sites.
My “bold” moves included tagging Matthew Stafford (QB-17 ADP) in the top
10 and Eli Manning (QB-9 ADP) at #17.
Most experts picked the opposite.
My preseason top 50 bold predictions went completely against the grain. Each one was an outlier based on more than 100 industry experts. I got 19 right, 26 wrong, and five were borderline picks. “Borderline” accuracy is defined by whether my bold pick—while wrong—was still more accurate than conventional wisdom. For example, the
Dolphins were my 12th ranked DST (compared to a DST-23 ADP). They finished the season ranked 14th,
earning a “borderline” tag, as they were every bit the bargain I expected, but
didn’t quite cross the threshold.
So if you took a chance on one of my bold predictions this preseason, you had about a 50% chance of getting a bargain, and nearly a 40% chance that that bargain would be a steal.
That’s it for the close-out. Tomorrow
I’ll offer some final thoughts on the year before signing off.