Our projection system is pretty complex. We don't give 'gut' projections on players. All of our predictions use
mathematical algorithms that look at key player performance indicators based on historical performace. We generally look at
these indicators over a 3-year period with the most recent history weighed the most.
However, it's more complicated than just taking the weighted average of the past 3 years. For example, when we look at a player
we look at the following things to predict performance:
Peformance Metrics -
When determining AVG, OBP, and OPS we look at the player's LD% (line drive %), FB% (fly ball %), GB% (ground ball %),
IFFB% (infield fly ball %), and IFH (infield hit %). This helps use eliminate how 'lucky' a player was from year-to-year and
gives us much more accurate projections going forward.
We also look at K-Rate, BB-Rate and the HR/FB rates in a weighted manner where the most recent data counts the most.
This helps us predict categories like HR, Ks, and BBs.
We do much of the same things when looking at pitchers. We discount the traditional metrics like ERA when predicting current year performance and use the more
reliable metrics like FIP and xFIP to generate pitching projections.
Lineup Projections -
As you may have noticed, we already have our projected lineups on the website. This is because our projection algorithms
need to know where a player is hitting in the lineup. Think of Hanley Ramirez
. If he's hitting leadoff then his runs will
increase and his RBIs will decrease. If he's hitting 3rd the opposite will be true. So, as projected lineups shift as
we head towards opening day, our projections will adjust automatically.
Team/Lineup Strength -
It's also important to look at team/lineup strength and our system adjusts projections accordingly. A player with
similar skill sets will produce better counting stats playing for the Yankees than the Astros. Our projections take this
factor into account.
How do you predict the performance of somebody with less than 3 years of experience (like a rookie)?
Admittedly, these players are the hardest to predict. For rookies we will look at their minor league numbers to get a general
guide of potential peformance, but then discount their projections significantly because the pitching in the majors is so much more difficult.
For other young players who have 400+ ABs in the majors we begin to ignore minor league performance and focus on their MLB metrics.
Our system is generally a little less optimistic on younger players because we don't try to predict if this is the year
they will 'blossom'. We go strictly by their past performance. We just try to show a realistic base for these players and let you adjust a player's value if you
think this could be the year they drastically improve.
We try to take an unbiased approach to our projections to give fantasy owners the best possible baseline in which to start your research. Our draft software
allows you to customize our projections if you think a particular player will do better or worse than what our projections say.
We advise people to start with our projections and then modify them based on any additional insight you may have (or even a hunch). Our draft software
adjusts the player values as you adjust their projections. If you think Jay Bruce will hit .285 this year then enter into our draft software
and see him ascend up your own customized rankings list.
OK, so I understand how you generate your projections but how do you assign player values?
This gets pretty complex but here's a general idea:
1) Let's say you are in a 12-team, 25 player league. Our draft software
finds the 'optimal' 300 players that should be drafted.
2) It then takes the optimal 300 and determines the average value of each scoring category.
3) After determining the average value it figures out how far each player deviates from the average and does this for every category for every player.
4) It then can give the proper weight to a HR/RBI/SB/Win/Save/etc... and assign scores to each player in the top 300.
5) We then bottom adjust for position scarcity.
6) After the position scarcity adjustment we find the player score in relation to the sum total of all top 300 player scores.
7) If Mike Trout has 2% of the total score then he is worth 2% of the total money available.
Our complex model is the most accurate player valuation method available. We believe we are taking a more advanced approach to fantasy baseball projections/valuations than has ever been done before.
Do I have to understand all of this math stuff to use your draft software ?
Absolutely not. All of this stuff happens behind the scenes while you work with our easy to use draft software
. We just explain it for those that like to know what's running under the hood.
Do you think your projections are the gospel?
Absolutely not. In fact, we include multiple projection systems in our software, as well as a Composite set that we think these are the best overall projections and provide the best starting base for any fantasy manager.
Our draft software
allows you to
customize the projections and it will dynamically generate the correct auction values based on your changes and league setup.
It's a powerful tool that will help you dominate your draft.