Picture this: you see two players left on the board, and you decide that one of them will be your next acquisition. You think long and hard, and your brain says that both players will likely be worth $15, and you can get either of them for around that price. When it comes time to bid on a player, you decide to target the player with more upside, because there’s a better chance he beats your projection than the other.
But why? If you valued both players at $15, then they should be worth the same amount come draft day. The players’ upside should be factored into the value you place upon them. Simply put, a $15 player should be equal to another $15 player. Simple as that.
So how do we go about factoring upside into a players’ value? Well, there are two ways, and one is much simpler than the other.
The first method is not the simplest method, but it’s more precise. When running projections for each player, you can simply input the player’s mean projection, thus allowing upside (and downside) to move the prediction. This involves doing full-on projections for each player, and for most owners, this is simply not an option due to the time commitment involved.
The second method is much simpler and easier, but it does involve a little bit of legwork. By simply doing a breakdown of the likelihood a player will be worth certain amounts, we can create a weighted average value. In the end, this should give us nearly identical results to the first method mentioned, but it doesn’t involve an entire projection and valuation system.
In its’ simplest form, here is what an analysis of some random player would bring us:
The table above just shows us that the player is assured to be a $5 player in the evaluator’s eyes. This is how most valuations look after the user has decided on a projection and arrived at the appropriate dollar value. This is certainly not realistic, but it is the simplest way to look at the idea. To be more relevant and realistic, here is a slightly more detailed approach of how the user may have come up with the $5 figure.
If you do a little math, this arrives at the same conclusion as before: the player is worth $5. This is what most owners do when they are settling on a value, but they usually do it in their heads. In a lot of ways, owners tend to double-count upside: once in their original projection, and again when favoring one player over another.
If you wanted to expand on the idea and apply more (and uneven) factors, here is another example.
As you can see, the overall value is not necessarily equal to the median or mean value. It is a weighted average of outcomes that can skew one way or another depending on the probabilities assigned.
Now that we have the general idea in place, what different categories can players fall under, and what do those categories mean for their value?
Players under the high variance label have high risk, but also carry a hefty reward if things go right. Most hitters in this category have high strikeout rates and fluctuating home run power. A great example of a high variance player is Cincinnati Reds’ outfielder Drew Stubbs . If all goes well, Stubbs could hit .270 with 30 homers and 40 steals. If things don’t go so well, we’re looking at .230 with 15 homers and 20 steals. That’s at least a $15 difference in his value. Pitchers in this category will have higher walk rates, but also high strikeout rates.
Injury, Playing Time Risks
For these players, it all comes down to how often they see the field. They may miss time due to injuries, positional competition, or reasons associated with prospects (service time, development, etc.). From the injury side of things, you have guys like Carlos Beltran  and Jose Reyes . Both of them can be fantasy stars if they’re healthy for an entire season, but they tend to slip in drafts due to concerns over how many games they will actually play. If you want an example from the prospect realm, Bryce Harper  fits pretty well this year. He could break camp with the team, or he could be stuck in the minors for another few months (or the entire year even). These players also fit in the “high variance” group, but they need their own subgroup to help identify them.
Batting Average Reliant
By “Batting Average Reliant,” I’m talking, of course, about players that depend upon the cruel mistress known as batting average on balls in play. These players aren’t going to hit homers or steal bases, and their ability to score runs lies on the back of BABIP due to a lower-than-average walk rate. Omar Infante  fits this category perfectly thanks to a career walk rate under 6%, a .310 career BABIP and a .275 career batting average. This also is relevant for pitchers with low strikeout rates that are going to rely on their defense to make plays.
Steady, Stars Edition
These players are giving you cost certainty thanks to their durability and skill set. Most of the time, hitters in this category tend to have lower strikeout rates and decent power, and those players tend to be stars. With your first round picks, you’re likely paying for players you know will produce at a high level.
Steady, Non-Star Edition
Some steady players aren’t stars, though. You know what you’re getting from guys like Nick Markakis  and Alexei Ramirez  at this point in their careers, but they aren’t going to be taken in the first couple of rounds. You pay for the lack of real downside, but you get a bit of a discount due to the lack of upside.
Low Upside, High Downside
These players just suck the fantasy juices out of your entire roster, and you have no reason to own any of them.
So, how do these players’ value ranges look graphically? Like this, of course:
You will notice that the “Low Upside” group is not included here, and that’s simply because their upside puts them at $0, so nobody cares. Not even their parents.
Also, this graph does not accurately show the weighted spreads of these players, but that is not what it was intended for. If we were to show the weighted splits, you’d see heavy upside/downside ratings for the “High Variance” players and not a whole lot in the middle. The Steady players would have a good deal of their value clustered in the middle of their range, with more outlier types on the outer edges.
The moral of the story is simple. Upside should seriously be considered, but it shouldn’t be done at the last minute. You valuations should already have upside and downside factored in, and you should refer to them as if they do. Double-counting upside is a common mistake, and one that leads owners to fill their rosters with too many riskier players. Don’t let yourself become one of those losers.