For some time, it looked like there was a pretty good chance I would win at least one of the leagues I'm in using the RotoWire Commissioner Service (I will, alas, be buying some other lucky colleague a steak dinner next month). My team from our June 1 redraft ran in first place for some time, but weakness in the pitching staff (looking at you, Jon Lester) has pushed me back to second place, a full 12 points behind Derek VanRiper and just two points ahead of Chris Liss in third.
Figuring out the way player valuations change over the course of the season is key to making the trades and pickups necessary. Seeing Lance Lynn go in the eighth round clued me into the fact that there might be some value to be had there despite the constant drilling of “Buy Low, Sell High” into everybody's heads by analysts these days. Seeing R.A. Dickey last until the 17th was a pretty good sign keeping him would be better value than selling, even with the inevitable regression from his absurd early-season run.
There was one question in particular I wanted to answer: was there a dominant strategy to the midseason draft game? Picking in these mock drafts is a matter of balancing out preseason value and what we've seen in the season to date. Who excelled? Those who picked the guys everybody thought would be good, or those who picked the guys were were already playing well?
Let's look at the data.
Personally, my hypothesis would be that teams filled with players with great preseason projections would outdo those filled with flavors of the month given the notorious fickle nature of baseball statistics. There were hints of that – VanRiper's team had over $300 worth of preseason value but just $40 of actual value at the time of the draft -- but nothing statistically significant showed up:
Statisticians typically require a p-value of .05 (95 percent significance) before a result like this is considered valid. Not even close here.
Nor do we see anything particularly notable when we look at the clusters that show up when we plot total preseason value and total actual value against each other:
Each cluster has its own successes and failures. As such, I think it might be more educational to instead look at a couple of the top and bottom teams and see which picks are driving their successes and failures respectively.
Up and down VanRiper's squad we see established names, from Mike Stanton and Jose Reyes down to Rickie Weeks and Max Scherzer. For the most part, Van Riper gobbled up players with obvious talent regardless of their play – Matt Moore (R12) didn't faze him despite the early struggles, and Moore has rewarded him with a 3.43 ERA, 111 K and nine wins. Scherzer has brought in a 2.88 ERA and 1.138 WHIP from the 18th round. Weeks has 15 HR and 12 SB to go with a shockingly acceptable .267 average out of the 14th.
Getting consistently good value from most of his picks – the only big misses were Peter Bourjos, Eric Hosmer, and Dan Haren – has put VanRiper well atop this league.
The early picks of struggling stars Robinson Cano (1st), Adrian Gonzalez, (2nd) and Ryan Zimmerman (4th) created the basis for my team. Cano and Zimmerman each have at least 20 home runs, 50 runs and 50 RBI; Gonzalez pulled it together to pull an average above .300, a whopping 75 RBI and a solid 12 HR despite his struggles with the Dodgers.
But the real home runs came by taking less-established players with high picks. Mike Trout in the third was one I thought may have been a reach, but the Millville Meteor has continued to show he may be the best player in the entire league, much less the best rookie. He has a sharp 22 HR, 61 RBI, 97 (!) runs and 38 (!) SB to go with his .335 average since June, the perfect player to build an offense around. Edwin Encarnacion was my other “Buy High” special out of the sixth round, as he provided another 20-homer player to help put my team top in the category.
The pitching staff killed it, though – Brandon Morrow's injury and undue faith in Jon Lester and Roy Oswalt left it middling in most categories.
Schoenke's team had major injury struggles, to be sure – Yoenis Cespedes, Brett Gardner, Johan Santana and Shaun Marcum all missed significant unexpected time. But there were also a few players popped simply too early. Jason Kipnis in the third proved disastrous, as he has managed just a .237 average and six home runs since the draft. Nelson Cruz never quite recovered from an early-season slump, hitting .248 with average power numbers out of the fourth round.
The Ryan Dempster pick was crushed by a trade to Texas, something nobody could have truly predicted, and Anibal Sanchez saw similar struggles with his move to the American League. Not even Adrian Beltre's power surge (23 HR, 59 RBI, 62 R) could save it.
Again, injuries hurt – Brandon Beachy, Matt Kemp and Alex Rodriguez, in this case. But that's not enough to explain Ray's struggles. Carlos Gonzalez was a bit eager as the No. 1 pick. Lance Lynn turned out to be a huge reach in the eighth, as did Mike Aviles (12th), Bryan LaHair (15th) and Tony Campana (21st). The common thread? All started out hot with little career success to back it up – even Gonzalez never had No. 1 pedigree. Acceptable picks later in the draft (Campana aside), to be sure, but too risky for how early they were picked.
Ray also bet on iffy closers (Heath Bell, Frank Francisco) and didn't get nearly enough innings out of his starters thanks to Lynn's transferal to the bullpen and Stephen Strasburg's shutdown. C.J. Wilson's collapse (unforeseeable) and Chris Capuano's nosedive (perhaps less so) didn't help things.
Such is fantasy baseball. There's no one template as to how to go about things, and no cheat sheet wins a league by itself, and injuries can decide an uncomfortable amount of winners and losers. For the most part, though, it would appear the best processes involve the most established talents – I would throw Trout in that group purely based on his pedigree – a simple maxim to live by when the June trading season comes around next year.