Introduction to the FIC Rankings
We've already shown you how last season shook out. Now it's time to show you our rankings and projections for the 2012 season. We could simply take our rankings, show how we rank the players, and expect you to take them at face value. However, the rankings will make a lot more sense to you if you have at least a general idea of how they're created. For that reason, we're releasing this introduction to go along with FIC's preseason rankings for 2012. I'm going to try to stay under 2,500 words (for the sake of both my and your sanity), but I'm not going to make any promises!
While I believe strongly in our rankings system, I always like to recommend that you don't use FIC alone in determining your cheatsheets. There are a ton of terrific resources around the net. You should always use all of the tools at your disposal.
In conjunction with the release of our 2012 projected rankings, we recently submitted our rankings to Fantasy Pros, which has a program that both shows and compares a variety of Experts' Rankings. If you're looking for multiple sources regarding who to target in your draft (and when), I highly recommend checking them out. That's an invaluable tool that can not only help you sort out your own rankings, but also get an idea of what your competitors may be thinking.
Without further adieu, let's talk a little bit about FIC's Rankings, how they're created, and what we think separates us from the pack.
The 5X5 Score
I would say that this is the most crucial aspect of how we create our rankings, but that would be understating things a bit. When we're releasing in-season rankings, this will be the only thing that goes into producing them.
We show you points, but our rankings are not based on Points Leagues
Firstly, it's important to point out that these values are not determined in a way similar to how you would find typical fantasy points leagues scored. Many of you have probably played in points leagues (be it for baseball or football) that award you a specific amount of points based on the magnitude of each event. For example, here are the values Yahoo lists for a standard fantasy points league.
- Runs - 2 Points
- Hits - 0.5 Points
- Home Runs - 4 Points
- RBI - 2 Points
- SB - 2 Points
- IP - 1 Point
- Wins - 5 Points
- Saves - 5 Points
- Earned Runs - -0.5 Points
- Strikeout - 2 Points
There are some obvious shortcomings that any rotisserie player would see here if they were planning on using the same rankings for rotisserie play and points play. Let's examine a few of them....
- In a rotisserie league, you're going to have to create a far more balanced roster. Many of us have seen owners tank one category (if they're lucky enough and they're playing in a league that goes beyond the 5X5 categories, maybe even two), but it will come back to bite you more often than not if you ignore one or two categories and load up on another. In points leagues, you can get away with that a bit more. It's irrelevant how you're accumulating points, so long as you're accumulating them.
- The points system has little to do with balance. You accumulate the same number of points for each Run, RBI, SB, and Strikeout. The frequency with which these events occur isn't taken into account at all.
Our point system accounts for values far differently. We do assign values (or points) to categories based on factors such as the frequency with which they occur and their effect on other categories. Still, I'm getting ahead of myself. Let's start with the basics.
Establishing Baseline Values
Now that we have part of the points vs. roto battle out of the way, let's talk a little bit about how we establish our baseline values. In order to create these values, we must look not at the whole league performance (counting categories listed above), but focus more on the player pool that a given fantasy owner is going to use to create their teams. Here are the positions used by Yahoo in a standard 5X5 league:
- 1 C
- 1 1b
- 1 2b
- 1 3b
- 1 SS
- 3 OF
- 2 UT
- 2 SP
- 2 RP
- 4 P
- 5 Bench
That's 23 players per team, or 276 total spots. Using the above information, we can say with certainty that there will be at least:
- 12 C
- 12 1b
- 12 2b
- 12 3b
- 12 SS
- 36 OF
- 24 UT
- 24 SP
- 24 RP
- 48 P
- 60 Bench
To establish the player pool that determines our values, that is the baseline that we can start with. Of course, most of the UT and P spots are going to be filled with players who qualify at other positions, which requires us to adjust beyond simply carrying 12 players for each (non-utility) position. The majority of those utility roles are going to be filled with 1b and OF, as those are positions that lean the most towards proficiency with the bat (rather than the glove), so the player pools used to create our average fantasy player use more players at 1b and in the OF than they would at C or SS. With regard to pitchers - who are liable to occupy at least one or two of those bench spots - we run a little deeper into the starting pitching pool than the reliever pool, as teams generally carry five or six starting pitchers.
We then create our baseline based on those player pools. Since we're looking for our baseline player (really a non-existent set of numbers) to be above average - you want to win your league, don't you? - we don't rank the top 20 or 30 players beforehand and establish our values from there. We sort and re-sort to find the top 20 or 30 players at each position in terms of Runs, (re-sort for) RBI, etc.* We then place the numbers from all positions into a larger pool, figure the sum of each category and divide it by the total number of players in the pool. By doing so, we create both the pool of statistics that fantasy owners will be working from and can create that baseline player.
* In terms of handling ratios for this process - Hits and At Bats are, of course, used rather than average. Hits, Walks, Innings Pitched, and Earned Runs are used rather than ERA and WHIP.
Using these guidelines, here is the baseline player - or an average player for a team that would be expected to win all categories:
|Player||IP||ER||H + BB||ERA||WHIP||Wins||Saves||Strikeouts|
After we've established our baseline, we really only use two factors to establish the values that we assign to each event - be it a Home Run, Stolen Base, Run Scored, etc. We determine the event's frequency - or rarity, if you will - as compared to other events. We also take into account how much of an impact that event has on other events that affect your rotisserie team.
A Simple Explanation of Why We Focus On Frequency... Comparing Runs to Stolen Bases
|Default Points League Score||41616||39608||18208||6558||68976||12145||6215|
The roto players in the audience are probably asking themselves right now why an event such as a run - which occurred more than six times more frequently than a SB - is given equal credit to a stolen base in Points Leagues. While a run is certainly an event of greater magnitude for a real baseball team than a stolen base, players in standard 5X5 roto leagues - where all categories carry equal importance - are going to find greater value in the stolen base.
The same can be said for each of the "counting" statistics across the board. We'll notice that there were more Runs than RBI (players score on errors... wild pitches... double plays), yet they get equal credit in Points Leagues. We'll notice that there were more than 4.5 runs scored per Home Run, but each Home Run is worth only twice as much as a Run Scored (more on that in a bit). This applies to pitchers as well, as not all baseball games result in a save (but each results in a win).
Unlike a standard Points League, roto owners need to use a more balanced approach. To simplify things, an owner could win the HR category by 30 Home Runs in roto and still finish with the same amount of points in the category (12) as they would if they had won the category by 1 Home Run (also 12). In a Points League, they would net an extra 116 point edge over their opponents in the category based on that 29 homer boost. By placing a higher value on events that occur less frequently, you're more likely to find balance with your fantasy squad.
Accounting For Events That Affect Other Events
That header may be confusing for some at first glance, but I will try to make sense of it for you. Each time a player hits a home run, four different things occur that affect his fantasy value:
- His HR increase by one (let's start with the obvious)
- He gets a hit (increases his batting average)
- He scores a run (increases his Runs Scored)
- He drives in at least one run (increases his RBI)
While the player who hit that home run is credited with the points (based on their assigned values) in every category, the magnitude of the event - as well as the fact that it affects three other categories - is given more credit because of the effect that it has on three other categories.
We apply a similar logic to a stolen base. It's a little harder to quantify, and it's on a smaller scale, but when a player successfully steals a base:
- His SB total increases by one
- He increases his run expectancy (there can be some cases [i.e, stealing second to make first base vacant with Albert Pujols at the plate] where some sabermetricians would argue that this is not true)
If ratios always seem like the most difficult thing to quantify when you're putting together your cheat sheet, it's because they are. The trick is not looking at the ratios, but looking at the total numbers used to accumulate them. Let's look at two hypothetical pitchers:
Without even looking at any of the other categories, we can tell you that Pitcher A is going to have a greater effect on his team's ERA and WHIP. Why? They had roughly the same ratios, didn't they? Let's look at a team that has accrued 1,000 innings with a 3.40 ERA and 1.20 WHIP to see how these two players will change things on the whole.
|Team Starts With||1000||378||1000||200||3.40||1.20|
|Team With Player A||1215||450||1199||238||3.33||1.18|
|Team With Player B||1170||435||1152||235||3.35||1.19|
While, at first glance, a savings of .02 in the ERA category and .01 in the WHIP category isn't dramatic, it's certainly a difference. The fact that we're using two pitchers who are identical with their ratios should tell you that quantity means something. As such, we use our baseline player to modify ERA and WHIP based on how many innings a pitcher has thrown. We do the same with batters, as a player who hits .330 in 600 at bats is going to impact your team more than a batter who hits .330 in 400 at bats.
We then compare each player's performance in the ratio categories (Batting Average, ERA, WHIP) to the figures that our "baseline player" has established. Before plugging the number into the value we've assessed each point of Batting Average, ERA, or WHIP, we use the weight of that player's impact on the categories dependent upon their total At Bats or Innings Pitched. At this point, we take the player's weighted* ratio and multiply it by our assigned value for ERA, WHIP, or Batting Average.
* Using a random number (not our value), if our baseline is 400 AB and a player has 600, his batting average will have 150% of the impact that a player directly at the 400 AB threshold would.
Adding Up The Values
If you think this is where things start to get easier, you're right. After using the baseline to establish our values for both the counting and ratio categories, we simply add each player's performance (modified by those values) up to create each player's FIC 5X5 Score.
Positional Adjustments For Our Overall Rankings
In order to create our positional adjustments, we use the median options from each position for a fantasy league. Using first basemen as an example, if the player pool used was 24, we would take the average FIC 5X5 Score of the players ranked ninth through sixteenth at the position to create the overall position's value. Much like we handle the ratios above, we would then divide each player's overall FIC 5X5 score by that positional value figure*. We then assign a player's grade for the overall rankings based on his performance against the average at his position.
* Unlike the baseline player, we do not subtract any values here. If a player performed at 84% of his positional value, his FIC 5X5 Score is multiplied by 0.84 to establish his positionally adjusted value.
It's also important to note that players with multiple position eligibility are run at all positions where they're eligible. Their grade for the overall rankings is used only at the position where they score the highest - in other words, their ideal position.
So Why Do We Show You The Score?
You can go to any number of sites and find out that they like, for example, Miguel Cabrera better than they like Paul Konerko (let's just be brutally obvious so nobody gaffs at this). There aren't very many sites that will mathematically show you based on their scoring system exactly how much more they like Miguel Cabrera than Paul Konerko.
More importantly, the FIC 5X5 score is a vital tool that can help you establish tiers of players. For example, you'll note that the gap between Prince Fielder and Adrian Gonzalez (our fourth and fifth ranked 1b, respectively) is relatively small. However, you'll see that there's a pretty steep dropoff between Gonzalez and the next best option (Eric Hosmer). This is something you will obviously notice most often when we drop from the top end players to more of a middle tier. There are points ten players are ranked 200 or more points apart. As you get to the middle and lower tier options, the separation doesn't tend to be nearly as great.
Other sites will tell you that one player is ranked ahead of another, but the gap between players is not always equal. We show you just how big that gap is!