happy 08 everyone -- first time logging back in since the fall here.
I'm in my 6th year of fantasy (10 team roto 5x5; both leagues; 9 pos. hitters plus DH/5sp/2rp; 4 keep), and each year I get a little "more sophisticated" in my tools and data analysis as I enter the draft and prepare for the season. The Cafe has certainly been a great help, and I'll thank those who post in advance for help.
A tool I looked a little closer at for the first time last year was VORP - I utilized it to target Utley in a trade as there was a significant drop from him to the 2nd tier of 2Bs - (of course, that gap narrowed quite a bit as the season went on with the production of - off the top of my head - Cano/Phillips/Johnson/Uggla etc.)
In theory anyway, VORP seems very valuable - (imo, keeping your emotions in check and not reaching when you don't have to is a key in mid-late rounds) - and I'd like to hear how some Cafe Members use it in their drafts: Where do you get your data? How much do you rely on it vs other statistics, gut/instinct etc? What's been your success over time utilizing the data set? Do you use it for batters as well as pitchers? In short, is it useful, and if so, how do you use it?
In my league, I'd bet I'll be one of only a few delving that deeply into the numbers - should I have an advantage from using VORP to build my team?
Are you using VORP from Baseball Prospectus (PECOTA) or the concept of VORP, applied to fantasy?
As an example, there are owners who come up with projections for all the players, then determine the mean and standard deviation for each category and total up the number above or below the standard deviation for each player in each category to get total points for each player. Then you can use that number as a "VORP" by figuring how much better, say, Chase Utley is over the #13 second baseman (in a 12-team league) as compared to, say, Alex Rodriguez over the #13 third baseman. In a 12-team league, of course, the #13 player at that position is the "Replacement Player," the RP in VORP.
Roger Angell: I was talking with Bob Gibson and I said: 'Are you always this competitive?' He said: 'Oh, I think so. I got a three-year old daughter, and I've played about 500 games of tic-tac-toe with her and she hasn't beat me yet.'
I'm not sure, but I think the fantasy one. perhaps I'm not clear, but I'm looking for what you describe in the 2nd part of your reply --- something to measure over replacement player. but, of course, not interested in constructing projections/means/deviations all by myself.
BobbyRoberto wrote:Are you using VORP from Baseball Prospectus (PECOTA) or the concept of VORP, applied to fantasy?
As an example, there are owners who come up with projections for all the players, then determine the mean and standard deviation for each category and total up the number above or below the standard deviation for each player in each category to get total points for each player. Then you can use that number as a "VORP" by figuring how much better, say, Chase Utley is over the #13 second baseman (in a 12-team league) as compared to, say, Alex Rodriguez over the #13 third baseman. In a 12-team league, of course, the #13 player at that position is the "Replacement Player," the RP in VORP.
When a player has multiple eligibility, you have to place them in one position, right? BJ Upton, for example. Would it mess with the RP if you had Upton in both 2b and OF?
Yeah, I'd put that player at one position (but noting that he's eligible at both). Whichever position he'd provide the most value is the one to put him at. Most likely, that would be at 2B.
Roger Angell: I was talking with Bob Gibson and I said: 'Are you always this competitive?' He said: 'Oh, I think so. I got a three-year old daughter, and I've played about 500 games of tic-tac-toe with her and she hasn't beat me yet.'
jswede wrote:How much do you rely on it vs other statistics, gut/instinct etc?
Well, it isn't really a statistic per se: it's a measurement using all the other statistics out there, or at least the statistical projections.
If you don't want to deal with the #s yourself then don't worry about VORP: any draft guide you look at should have already incorporated the idea of VORP into making their rankings. If you try to adjust for it after looking at your rankings, you'll just be over-emphasizing the whole idea.
0-3 to 4-3. Worst choke in the history of baseball. Enough said.
as far as figuring people with multiple positions of eligibility, I find out what position they are most valuable at, such as BJ Upton is more valuable at 2B than OF. I leave him where he falls in the OF, but I don't count him toward the position totals. Then when I am calculating my VoRP totals I calculate it for Upton at OF, or any other player at their worse positions
BobbyRoberto wrote:Are you using VORP from Baseball Prospectus (PECOTA) or the concept of VORP, applied to fantasy?
As an example, there are owners who come up with projections for all the players, then determine the mean and standard deviation for each category and total up the number above or below the standard deviation for each player in each category to get total points for each player. Then you can use that number as a "VORP" by figuring how much better, say, Chase Utley is over the #13 second baseman (in a 12-team league) as compared to, say, Alex Rodriguez over the #13 third baseman. In a 12-team league, of course, the #13 player at that position is the "Replacement Player," the RP in VORP.
Can you walk me through how to do this?
So I- 1. Calculate the total points per player based on my league scoring system. 2. Calculate the mean scoring of all players for that position. 3. Calculate the standard deviation by subtracting the total player points from the mean.
Is the resulting # basically my fantasy VORP? Do I want to do it by position, or by total player pool? I would think by position, no?
Position, a quick and dirty way to find out who is the top 12 or 20 or whatever the number of teams in your league is I average all the stats for the position. Get the average for whatever stats you use, then subtract them from the player's projections. You can a value you can use to rank originally. Then you can get the standard deviation from the 13th ranked player, or 21st, depends on the number of teams.
Broncmet724 wrote:Position, a quick and dirty way to find out who is the top 12 or 20 or whatever the number of teams in your league is I average all the stats for the position. Get the average for whatever stats you use, then subtract them from the player's projections. You can a value you can use to rank originally. Then you can get the standard deviation from the 13th ranked player, or 21st, depends on the number of teams.
What do you mean by "get the standard deviation from the 13th or 21st ranked player"? Standard deviation is a measure of variability (or spread) between scores/values. What you are describing appears to be something completely different.
Cafe Home
Fantasy Football
Fantasy Basketball
Fantasy Hockey
by jswede » Sun Feb 17, 2008 12:25 pm 
