stealthdevil wrote:Okay. maybe I should've titled this post a little differently. what I'm trying to do is get some feed back on what areas of a pitching prospects game should get the most emphasis when trying to predict how much of a fantasy impact player they will be. I posted the formula I was working with earlier in this thread. so far the only advise/idea that I've gotten is that minor league statistics should play a role in the formula. I have kind of addressed statistics w/ the performance measure, but it isn't weighted very much and doesn't use exact stats. So do you guys have any thoughts on this?
Ok some ways to improve you ratings.
1. Use MLE stats (major league equivalents) you can find these numbers from different sources but what it does is it takes minor league stats and adjusts them to major league equivalent based on minor league level and what different league (i.e. ny penn).
2. Take those stats to paint a picture of the range of stats in key categories (k/9,bb/9,k/bb,hr/9,gb/fb,whip for pitchers and ops and ct% for hitters) and in more traditional categories (BA, HR, SB, ERA) that the player is likely to produce.
3. Take those ranges and assign a major league value/grade to each line.
4. Create a percentage of chance that the player produces each stat line.
5. Weigh each stat base on the chance that the player produces the stat line and create a mean stat line, assign a value/grade to that stat line.
PS: You also have to include stuff like IP/MLB seasons and injuries to produce better grades. Kerry Wood would be an A+ if you take his healthy years but there are only a few, same with Mark Prior.
This shows his upside potential, downside, and most likely production.
Example: Justin "Joba" Chamberlain, RHP, NYY
Gets injured/Career over 0, 3% chance, Grade F
Misses tons of time, 3MLB seasons, 8K/9, 3.80ERA, 17% chance, Grade D
OK MLB pitcher, 14MLB, 7.5K/9, 3.80ERA, 30%chance, Grade C+
Good MLB pitcher, 15MLB, 8K/9, 3.50ERA, 30%chance, Grade B
Very Good MLB pitcher, 16MLB, 9K/9, 3.20ERA, 15%chance, Grade A
Legend, 19MLB, 10K/9, 3ERA, 5%chance, Grade A+
MEAN: about, 15MLB, 8K/9, 3.75ERA, Grade B (For most prospects very high but Joba has shown his stuff at MLB level and has high celling so mean numbers are a little high)
1. Your numbers could be much different
2. This is off the top of my head, percantages with no research on this wont be very good. Not my actual prospect grade for him. (dont have one yet
)
3. Didn't use actual MLE for my projections or use calculator for numbers so not perfect
I like this system but it has obvious flaws. It is not good for projecting pitchers with less than 200-300 MiLB IP or batters with less than like 600-700 MiLB ABS otherwise you are going on a small sample size of a player just jumping to a new level. There may be some MLE for college but I have not seen any yet, that would though help tremendously when talking about players with limited MiLB experience. If you cant find good MLE for a player look for a few scouting reports and similar players along with injury history to produce stat lines. Also this is a VERY new system of player evaluation of MiLB players (compiled different parts of this from many sources most not really having anything to do with MiLB, started to use it this year) and requires a ton of work but it could pay of if a few tweaks are made. Hope this helps, haven't tested this a lot but I think it works ok.
Stuff I need to add:
1. Injuries, because just doing research on individual players (except in a few cases) is not worth it.
2. Using scouting to produce stat lines (in cases of small sample size).
Cafe Home
Fantasy Football
Fantasy Basketball
Fantasy Hockey
by jmlevy » Tue Mar 04, 2008 2:25 am 
