Birds of a feather, whatever

Some fantasy team owners like to have multiple players on the same team on their team, others don’t, and a third group is largely agnostic. Is there are any reason to have a preference? I don’t think so.

Agnosticism is my default position on this issue, but there are some instances in which I can see a reason to stack players on the same team. Let’s address the fundamental principles behind the question, which should give some broad insight as to the possible exceptions to my stance.

The idea of stacking players on the same team is one of consolidating assets, and therefore consolidating risk. By doing this, you are increasing the likelihood of stark short-term variance.

So, if maximizing variance over short scoring periods is to your advantage—say, in a daily contest where you compete against multiple teams for the highest score—– then I think there is merit to stacking as a gambler’s philosophy. In a full-season league, however, even a weekly scoring format, I assume the peaks and troughs will cancel each other out over time, leaving you with nothing more than the sum of individual value of each players’ production.

In a H2H league with weekly scoring periods, stacking players simply increases your team’s performance variance, including its vulnerability to the effects of inclimate weather. Building a team this way may cater to a player’s strategic preference and risk tolerance, but I’m skeptical it actually constitutes an advantage.

In a roto league, I don’t believe player stacking to have any significant impact whatsoever. That is to say that given two identical players, one of whom hits fourth for a team whose third hitter you also own, and the other player being the fourth hitter on a team from whom you own no players, I see no strategic advantage to pick one player over the other. Or, expressed slightly differently, I don’t see any reason to pick a lesser player ahead of a better player because you own other players on his team.

Now, that I’ve given my conclusion, I’ll briefly explain my reasoning.

Quite simply, in the game of baseball, one player does not directly take opportunities from another in a zero-sum sense. On a basketball team with a lot shooters, every shot Stephen Curry or Monta Ellis take is one fewer shot David Lee can take, so there is a form zero-sum dynamic at play.

However, in baseball, each player is going to get four or five plate appearances, and those are his opportunities to produce. Chase Utley only influences Ryan Howard’s number of opportunities to produce in the extremely general sense that a successful offense generates more PAs for the team as a whole, and to no greater degree than Placido Polanco does.

Some may think having two hitters on the same team helps because you can double up on stats by having an owned player drive in another owned player, but that is a myopic way to view things.

I want to own players who are going to have the best chance at being driven in the most often and who are driving in as many runs as possible. Who drives in my players and who my players drive in is inconsequential; stacking players does not increase or decrease the volume of opportunity.

One may argue that if I own Kevin Youkilis and David Ortiz and Youk homers in front of Papi, this takes away the opportunity for Papi to produce. I disagree for two reasons. One, the opportunity is rooted in the PA, and the context of that PA is beyond the control of the player. Two, in this respect this is a zero-sum game. A player can only be driven in once, so if Youk drives all those runners in, then you’ve achieved all the production that can be squeezed from those players. How that production gets split between Youk and Ortiz is nothing more than cosmetic.

It should be emphasized that none of this implies that aren’t reasons to own players on certain teams, just that one shouldn’t feel dissuaded to own multiple players on the same team. The Boston Red Sox are a potent offensive force that puts runners on all over the place, scores tons of runs, and plays in a hitter-friendly ballpark. These are all reasons to value Red Sox hitters. But, these are all factors that should be taken into account when pricing players in the first place.

The team for which a player plays has profound impact on his value, but that player should be no more or less desirable to an owner because of the composition of his team’s individual player’s team affiliations.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Blind resume
Next: Are there two Jonah Lehrers? »


  1. Jim G. said...

    I have two teams doing very well – one with Ryan Braun and Prince Fielder, the other with Miguel Cabrera and VMart. No complaints.

  2. dht said...

    One way stacking would increase the overall production, at one extreme at least, is taking park effects into account. 

    Stack Colorado hitters or Safeco or San Diego pitchers. 

    But without doing an approximate calculation, I would only stack by opportunity.  Say I had to choose between adding another Rockies hitter versus a comparable non-Rockies hitter.

  3. Brad Johnson said...

    I do tend to stack Padres relievers…but I think that’s has different incentives built in than stacking say the Reds lineup.

  4. Rorgg said...

    I’m pretty agnostic, so just by the way randomness works, I tend to stack a little on one or two teams.  This year, it’s Cleveland (Santana, Sizemore, plus I carried LaPorta and Asdrubal for short times—made a trade offer for Choo but couldn’t land him) and St. Louis (Pujols, Carpenter, Berkman to start, and I’ve picked up Lohse and Craig along the way).

    Couple ChiSox and a couple Angels, but that’s about it.

  5. Brad Johnson said...

    I ignore team (but not park) and go for whichever player best helps my roster. So anything I stack is random. This year, the Giants (Sandoval, Torres, Posey, and Belt) and Rays (Zobrist, Joyce, and Damon) have shown up on more than one team.

  6. TomtheBod said...

    I came across this problem by default.  It just so happened that I ended up with 5 out of the 9 Kansas City Royals’ starting lineup:  Alex Gordon, Billy Butler, Jeff Francoeur, Eric Hosmer, and Mike Moustakas.  I was not actively trying to pursue members of the same team.  I’m a white sox fan and I hate the Royals.  Nevertheless,  I was just hitting up the waiver wire for prospects and people who were playing good and I ended up with these guys. 

    I finally was able to trade one of them; Billy Butler along with Jose Tabata for Adam Lind.  I’m still rolling with 4 but trading is like pulling teeth in our league.  I’m happy with Gordon as a starter and I spot start Francoeur and Hosmer.  I’m still waiting on Moustakas to heat up.  If they keep marching Mouse out there, I think he’s gonna catch fire after about a month, which would be just in time for the playoffs.

  7. Chad said...

    No reason to stack up in roto as the overall stats are all that matters.  If two guys happen to be on a high powered offense then they are both desireable but no moreso than two guys on separate but equally powerful lineups.  I see the same thing in football where people are drawn to having a QB/WR tandem thinking they “double up” on the stats.  It just doesn’t make any sense.

  8. Derek Ambrosino said...

    In fact, in football, it is almost a defensive strategy to own WR1s on the teams that super elite QBs throw for. Since WRs get more pts for yards and receptions than the QBs get for yards and completiions, a 40 yd TB pass from Manning to Wayne is going to get Wayne more pts than Manning. This is a good way to neutralize the light-it-up QBs because every time they throw to your WR, you gain more than the QB-owner does. Owning the combo does prevent against that.

    The sentiment is more understandable in football because if Reggie Wayne scores than Peyton will too, so there’s a desire to own both. But, really you just want a QB who is going to throw a lot of TDs and a WR who is going to catch a bunch of them – if that’s Manning and Wayne, those are two great players to have at their respective positions, but their whole is not greater than the sum of their parts, and they have advantage over a Rodgers/Roddy White pair.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>