The Real Fallacy of Value Based Drafting
What do Phoebe Cates and cheap beer have to do with proper drafting? Todd explains.
Value has become a buzzword in fantasy sports – one which I feel is misused let alone overused, but that’s a story for another day. Let’s just leave it at this. Whenever you want to say value in reference to fantasy sports, instead say potential. You’ll be shocked at how many times it makes the thought so much more practical and realistic. But I digress.
For those unaware, value based drafting (VBD) is a process that adjusts rankings based on the player’s position. The idea is 15 points (as an example) from your QB, RB, WR, TE, D/ST or K are not equally helpful.
This oversimplification illustrates the reasoning. Two-person league, typical scoring, me and you. The player pool is Aaron Rodgers, Drew Brees, Eddie Lacy and Pierre Thomas. What can I say, we’re Green Bay and New Orleans fans. You have first pick who do you want? If you ask me if the league is PPR I’m kicking you out.
Hopefully, the obvious choice is Lacy since the difference between Lacy and Thomas is much bigger than the difference between the signal-callers. Perhaps this is intuitively obvious, but soup it up and apply it across all positions and you have the basis for VBD.
I may be wrong, but it’s my understanding that icon Joe Bryant was the first, at least publicly, to introduce VBD to fantasy football. Allegedly, he borrowed the notion from fantasy baseball, where VBD is better known as positional scarcity.
The idea behind VBD is to properly rank player relative to positions. A QB projected to score 200 fantasy points should be ranked differently than a RB, WR or TE projected for 200 points.
Each position is adjusted relative to a baseline. Some set the baseline to be replacement level, or the last active player drafted at each position. Others compare each player to a mean or average at their respective positions.
Opponents of VBD will cite the above as its primary flaw. They’ll question using replacement level, noting the adjustment assumes a fixed player pool. Using QB in a 12-team league as the example, the adjustment would be subtracting the projected points for the 12th ranked QB from all QB. But an integral element of the hobby is exploiting matchups and it’s possible a team has the 13th and 16th ranked QBs and deploys the one with the more favorable matchup. This skews replacement. To account for this, I’ve seen some lower replacement to the 10th best QB (or whatever). Others use historical data, looking at the previous season and averaging the points of the 12th ranked QB each week.
The argument against comparing against a mean is what does that really accomplish? It’s elegant and sounds all smart and stuff, but does it accurately rank players across positions or is it just a fancy-ass mathematical exercise?
Not to mention, even if one of these were theoretically correct, bye weeks and especially flex spots (if your league uses them) completely trashes replacement. Where do you set replacement or compute the mean when some players will be at flex? Do you fill in all active spots then put the next highest projected players in a flex and count up the number of each position worthy of an active roster spot incorporating flex? Seems like a plan but now you have that whole matchup thing screwing things up.
I’ll let you in on a dirty little secret. None of this matters. It’s all mental masturbation. The real flaw has nothing to do with proper setting of the baseline. It’s much simpler than that.
In order for VBD to be effective, everyone has to rank players in the exact same manner. Not only that, they need to share approach and philosophies. In a vacuum, you and I can expect the same production from a player. But there’s more to projection than that. There’s upside and downside. There’s reliability and durability. There’s consistency. The player may share a bye-week or have a perceived difficult matchup during your playoffs. You may already have a player on the same team on your roster and you prefer to diversify. The point is, not only will everyone have different expectations for each player, but intrinsic potential to each fantasy squad differ.
We may be bordering on the intuitively obvious, and that’s fine. But the implication with VBD is to continually take the top-ranked player on your board as they offer the most potential, at each selection, to your squad. That’s just wrong-ditty-wrong-wrong-wrong.
Let me be clear. Some form of value based ranking should be done. Just don’t be married to the list and robotically take one off the top.
The goal is to put as many players that you rank highest on your roster. Sometimes to accomplish that you need to deviate from the order. Proper ranking is a science; effective drafting is an art.
Using another oversimplification to demonstrate the point. It’s several rounds into the draft and you’re waiting on your QB and feel it’s time to get one. Your draft spot is one before the wheel and the person on the turn has Peyton Manning. Everyone knows to wait and take the QB on the comeback pick. This isn’t rocket science.
Now take that scenario and pump it up with steroids and spread it over every team in the draft, looking at each squad’s construction and owner’s philosophy. Your objective is to assimilate all that in a manner to equip your roster with as many players you rank as high as possible. This may mean jumping down the ranks to get a player you know will be taken before your turn. This may entail skipping over your top-ranked player for a round or two, sensing he’ll be there at a later pick.
Again, this may be intuitive; something you’re doing already. And if you are, kudos.
But, I am hearing and seeing far too much emphasis on the D in VBD. Be it in mocks or just advice doled out, D can’t be mechanical, it has to be fluid.
If I had it my way, I’d eliminate VBD from the vernacular and call it VBR – value based ranking. Actually that’s not true either. I’d call it potential based ranking. Or....PBR.