This is a topic that is alternatively the bane and boon of fantasy players everywhere. The topic is far too complicated for a comment so I thought I would post my thoughts here.
This is a fascinating topic and one that is always a hot topic. For what it is worth I have no problem reaching in either scenario that Mike presents.
First a hypothetical. Is it fair to assume that as a matter of theory all players can be ranked from one to “n” where “n” is the total player pool? I think that it is, and it is hard to imagine anyone arguing with this. From this it follows that we can translate this directly into rounds; again for argument’s sake, and about this there can be little argument.
Assuming a 500 player pool we, in theory, can generate a ranking list from 1-500, set out by rounds. Assuming that everyone agrees that the values are close to correct and that the list is a fair and reasonable one, we now can predict exactly where players will go. Again this is a theoretical point; we have an agreed upon ranking list, based on projected values and apply it to a draft.
It is only from this framework that “reaching” exists? Can you see why? The idea only makes sense insofar as a concrete, reasonable draft ranking list exists.
That such an animal does not exist though is irrefutable fact. Not only is the entire concept of projections subjective, even within a given player there is so much margin for error that even assigning a round to a player once you get past round two is a futile endeavor.
Using the BaseballHQ YTD+Proj spreadsheet and combining pitchers and hitters for 12 team dollar values as a rough guide, we see that players 12-24 ranged from $30-24. Players 25-36 ranged from $24-$22. Players 37-48 ranged from $22-21. Players 49-60 ranged from $20-18. To skip a bit, Players 133-144 ranged from $10-9. What we see is that the top and bottom of each round is roughly $2 after the second round.
Conclusions from the hypothetical:
Let’s assume for argument’s sake again that our reasonable, concrete draft rank list reflected roughly the same spread of values; again a perfectly reasonable proposition. What can we glean?
1. Once you get past the twelfth round you are dealing with commodities that are primarily uncertain and roughly equal. That is because when you are dealing with speculative, uncertain commodities there is no rational basis for picking one over the other unless there is some standout reason. This is where all of our hard work comes in, but it is a fact that no matter how hard you try and no matter how much you know you are still an underdog in the context of a particular player to be correct.
2. The round differences in value are far smaller than we make them out to be. And that assumes that we have a fixed value for each player. One could take the 48th player at a projected value of $21 and assuming no margin of error at all, can draft them one round earlier and not lose anything more than $1 or $2 at most.
3. In a given player’s forecast there is a +/-$5 margin for error, as established by BaseballHQ research. To put that in the context of this discussion, that is a five round difference from high to low, based on nothing more than normal fluctuation of a player’s dollar value.
In the context of what actually happens on the field we see know a compelling argument that there is no such thing as “reaching” in any real sense. When a given player’s forecast can vacillate by five rounds on nothing more than chance, to what end can we advise players not to reach? In fact the opposite is true, if you have good reason to believe that a certain players will outperform his forecast significantly, the gain to be had must not be measured by comparison to a round but by the opportunity cost of drafting another player.
What are the outcomes?
Let us presume that I love Andre Ethier for example, and though he has a forecast of $15 I think he will be worth $23, again for argument’s sake. There are only a few scenarios:
1. He generates $23 and I drafted him. In this case it doesn’t matter where I drafted him as long as the round in which I drafted him does not have an expected value of around $18 or so compared to his $15 forecast. So even here we have a 1.5 round margin.
2. He generates $23 and I missed out on him. Good luck to you finding another $15 forecasted player that will produce $23. If you know more than two per year and are right more than half the time then you should be playing for much higher stakes than you are. So assuming you can replace a $15 forecasted player with another, you are no better or worse off, but have to pay the $8 opportunity cost associated with missing out on Ethier.
3. He generates somewhere between $20-10 (+/5 margin for error). We will ignore the range from 20-23 for ease and length of this damn article. Here as long as I drafted him in any round where the expected value ranges from $10-$20 I have not lost or gained no matter what he actually produces, since I cannot reasonably be expected to hit a player anywhere more accurately than his innate margin for error. Again if you can do this, or think you can, then perhaps you are the exception to the Texas Sharpshooter Fallacy and are actually a sharpshooter.
The long and short of it is that you should not hesitate to reach if you have specific reasons for doing so. You cannot routinely reach beyond the +/- $5 range and win, but in a particular instance you can do so without hesitation. If you are wrong you probably won’t lose much, but if you are right you stand to gain a lot more.