Sunday, October 20, 2013

Sunday was the final day of my first fantasy baseball season. I finished in second place by 0.5 measly points in a standard roto mixed league. I loved every minute of it, despite originally only doing it to support my son in his first fantasy endeavor. His sport is baseball, and I'd never done fantasy baseball so I figured joining my own money league would motivate me to learn the ropes enough to help him. He also ended up finishing second in his league, so it was a nice intro for both of us. Over the course of the really long six month season, I learned that fantasy baseball is a lot different than fantasy football. First off, there are a lot more guys you have to know, and their roles on field don't necessarily translate into the stats they can supply to your fantasy team. I whiffed on stolen bases pretty hard in my draft, for example, being more concerned with filling field positions than accumulating stats in every category. Second was the baseball language. I'm pretty open minded and adventurous but I never expected to be coveting the guys with the nastiest, most filthy "stuff". Hitters are reduced to speed or power, and rather than home field advantage, you have hitter's parks and pitcher's parks. Finally, after years of playing fantasy football and basketball, it came as a shock that a professional baseball player could miss ten days and two starts over a blister. A blister!

I ultimately realized something else about fantasy baseball. Very simply, the best players play for the best teams. Teams that score a lot of runs and win a lot of games tend to roster players that score a lot of fantasy points. Pitchers on those teams get a lot of run support and rack up the wins, right Mr. Scherzer? Although this seemed obvious, the only way to really know is to look at the numbers. I assigned each team a numerical value corresponding to their win-loss record. A "1" was the best, a "10" was the worst winning percentage in MLB. I plotted that number against the fantasy value of the top 100 hitters. Honestly, the correlation isn't as strong as I suspected, but one of the reasons is that only 4 of the top 100 fantasy baseball players according to ESPN's player rater came from teams with a terrible winning percentage rank of 8-10. 65 of the top 100 fantasy baseball players play for teams with a rank of 1-3. So in fantasy baseball, useful fantasy players disproportionately play for the top teams.



I really did this little analysis in order to see whether the same was true of our favorite fantasy sport. Do the best fantasy football players, in terms of fantasy points per game, play for the best NFL teams? My analysis used data from the complete 2012 season, with teams ranked according to their win/loss records. I used a cutoff of 7 fpt/game in standard scoring, which yielded data for 128 offensive players (excluding K).
The answer is no, overall it doesn't matter what the team record is, fantasy value is fairly evenly distributed amongst teams with very different records (13-3 through 4-12, which corresponds to rank 1-10). Only the very worst teams (KC and JAX were both 2-14 for a rank of 12) show any decrement in average fantasy points scored.

One trend that emerges upon closer examination is at QB. Matt Ryan and Peyton Manning averaged 19 fpt/game (their teams ranked 1st) while Matt Cassel and Blaine Gabbert averaged about 10 fpt/game in 2012. There is a nice correlation between fantasy value and team rank when you look at QBs alone (at left). Thus you can make a case for the best fantasy QBs coming from the best NFL teams based on last year's data.

If you remove QB from the plot (at left), the WR/RB/TE stats fall in the same range (7-15 fpt/game) regardless of team record. This means that for fantasy football, you are just as likely to find fantasy gold to fill your flex spot on a 4-0 team as a 1-3 team. Good to know, since we are kind of predisposed to believe the baseball truth--good begets good, and bad is, well, bad. We can certainly find examples from our current season--compare the Denver and Jacksonville WR stats--that support our bias, but the overall data refute the correlation in football.

The top five WR/RB in PPR scoring so far this year (with their team records) are: JC (4-0), Victor Cruz (0-4), Demaryious Thomas (4-0), Julio Jones (1-3), and Adrian Peterson (1-3). Among widely owned players that have played the first four games, the five worst in the same league's scoring are: David Wilson (0-4), Lamar Miller (3-1), MJD (0-4), BJGE (2-2), and Vincent Brown (2-2). Thus far, it appears that fantasy goodness and badness can be found on any team in 2013 too.

So don't go overlooking a Robert Woods or Marlon Brown on waivers this week because their teams are average, and don't assume Donnie Avery or Dexter McCluster is the answer for your fantasy team because their team is great so far. I say beware of grabbing Matt Cassel or Mike Glennon until (unless) their teams improve. I mean, as always, get whomever you want, I want you to like your team and all, just be careful using team record as justification for your choices.

No comments:

Post a Comment