Why managing risk is the key consideration in end game picks, and whether an owner should seek high or low variability.

When I write about the concept of value and its lack of vitality, I recognize that I am partly a lone voice railing against the traditional measures of “value” when dealing with the top of the player pyramid. As a result one can easily criticize my theories based on the notion that I am giving owners carte blanche to choose whatever player they want.

In a way that is true. My next column at RotoTimes’ premium site will attack the premise that in the later rounds who you pick is important. So not only am I now attacking the hobby’s most sacrosanct commandment at the top of the pyramid, I am also attacking it at the bottom.

In fact, it makes little to no difference to pursue “value” in later rounds as well. Each General needs to carve out a plan, and decide what criteria are to be used in the later round picks, value be damned. Often this may need to be done on the fly in a short amount of time, especially in online drafts. The concept of risk management is key; given what has come before the General needs to decide whether to maximize or minimize risk. Only then can one consider whether to go for “value” or “reliability” or “stats” or whatever other criteria is used.

I think that the best “default” strategy is to is just select whomever one wants and believes will vastly out produce the round. You try to hit a homerun with every pick, and cut loose those that are not homeruns quickly to replenish from the free agent pool. However, this is an individual choice mapped out ahead of time and planned for by the expert player, so that decision on whether to deviate or follow course can be made swiftly and correctly.

There is a sliding scale that must be considered. On the one hand we have the probabilistic theory of projections, as I have pointed out previously. On the other we have the risk involved with the projection. Consider these to be akin to the supply and demand curves in basic economics. The intersection of the supply and demand curves is the optimum solution.

A similar principle applies here, the optimum choice for the individual is the balancing of risk versus projection, and for each owner the answer may be different. The difference is that in our case, the General can choose the degree of risk he wants to accept. One can make the selections in later rounds by determining exactly how much risk is desired.

Tim Dierkes of RotoAuthority.com recently had a posting about the value of Greg Maddux and how he was undervalued compared to more risky pitchers such as Ubaldo Jiminez, as Jiminez and others like him were going earlier in the ADP reports.

For most fantasy players Tim is correct; in a non-keeper league where you are trying to win this year it is often best to take the surer thing. But what exactly is a “sure thing?”

When a player like Maddux has a $9 projection, we are really saying that he will produce between $4 and $14 65% of the time. For a late round pick that $14 could be very juicy and even if he produces a $4 season it isn’t the end of the world. If this fits in with your overall plan then so be it and select him.

Maddux is a relatively secure commodity, and we think we can be relatively certain what we are going to get. However, the best strategy and analysis by the General considers not just that Maddux might be secure but also the more important question of what degree of risk is warranted.

For example, in a vanilla 12-team league pitchers like Maddux are easily replaced. So the General may very well decide that he wants to gamble boldly, taking for example a pitcher like Jiminez who presents a greater reward. If the risk succeeds he wins a lot, if it fails he loses little since he can cut bait very quickly and minimize the damage.

The proof of this would be that last year one could substitute James Shields for Ubaldo Jiminez and have the precise dilemma posed here. Yet taking the surer thing would have been an error.

Ron Shandler’s research establishes that pitchers who dig early holes in April have a far worse outcome than those that do not; it is extremely difficult to dig out of an early hole. So, guided by this premise the General can easily choose to maximize risk looking to hit a homerun with the knowledge that he will cut bait quickly enough once the season starts.

It is hoped (but not certain) based on Shandler’s research that if you have selected a lemon it may be clear from the get go (consider for example Adam Loewen last year versus Shields). If it isn’t, then that is the downside of taking the risk and you will pay the price. But merely advising fantasy players to go for the sure thing cannot be right either.

One can argue in response that even in one month the damage is done and it should be avoided. Fair enough; but the benefit might also be gained in April and the sad fact is that in a given month what we are really talking about is only the marginal difference between the two choices, Maddux and Jiminez in this example. This may be far smaller than one thinks.

So in a way, I guess the charge that I am giving carte blanche is right. However, for players with strong judgment, galvanized by experience, I think this is precisely correct; such players should have carte blanche regardless of any notion other than selecting the player one prefers, with no other criteria applicable.