Prospect Ranking: Statistics
A while ago, I asked you for idea on how to best rank and evaluate prospects. There were many great ideas, but I think actual on-field results top them all and deserve most recognition/value.
Here is an attempt to plug actual statistical results into a "Master Prospect Ranking" formula that I will use in a few months to rank the top prospects in the Twins' system.
(Remember - this is just the statistical portion. I will add other components as time goes on.) After the jump I'll explain my thoughts...
The first thing I've done is find the league average of five major statistic categories in each of the five Twins' minor leagues (IL, Eastern, Florida St., Midwest, GCL) and assign penalties/rewards for being below/above average in each of the categories. There is no penalty for playing at the same level as the average hitter in the league. (Though there are flat values for each league, which I'll explain later.)
Let's take for example the International League, home of our Rochester Red Wings.
Batters
Batting Avg: .262 if >= 10 points above = +10
if >= 10 points below = -10
OBP: .328 if >= 10 points above = +10
if >= 10 points below = -10
SLG: .396 if >= 10 points above = +10
if >= 10 points below = -10
BB%: 8.78% if >= 5 points above = +10
if >= 5 points below = -10
K%: 19.67% if >= 5 points above = -10
if >= 5 points below = +10
Pitchers
ERA: 3.93 if >= 1 point above = -10
if >= 1 point below = +10
WHIP: 1.354 if >= .33 point above = -10
if >= .33 point below = +10
BB/9: 3.2 if >= 1 point above = -10
if >= 1 point below = +10
K/9: 6.9 if >= 2 points above = +10
if >= 2 points below = -10
K/BB: 2.17 if >= 2 points above = +10
if >= 2 points below = -10
(NOTE - These averages will obvoiusly change and I'll base the values off the new value whenever I rank the prospects.)
From this formula (you obviously use one or the other) you will get one solid number. This presents another problem: it is obviously easier to be an average hitter/pitcher in a hitter/pitcher-friendly league, as the case may be. So here's the ranking of the five leagues in order of their strengths, and bonuses that will be awarded in certain cases:
Hitter-Friendly Leagues (based on OPS)
If average/below average in 3/5 categories:
If above average in 3/5 categories:
Pitcher-Friendly Leagues (based on FIP)
If average/below average in 3/5 categories:
2. (tie) Midwest (A)
4. Eastern (AA)
5. International (AAA)
If above average in 3/5 categories:
What do you guys think? How accurately would this rank statistical performance, and what improvements could be made?
0 recs |
8 comments
Comments
Interesting...
But I just don’t know if there is some ‘perfect’ statistical analysis of prospects. Prospects are such an inexact science.
They’re all interesting, so I would be curious what this would produce.
by SethSpeaks on Aug 23, 2009 10:25 PM EDT reply actions 0 recs
Just noticed that I forgot all about E-Town.
Whoops.
Viking Vigil - The 2nd greatest Vikings community on the internet!
http://www.vikingvigil.com
Twins Fix
http://www.twinsfix.com
by Andersklasen on Aug 23, 2009 11:08 PM EDT reply actions 0 recs
I'll be interested to see what you come up with.
I agree with Seth—rating minor leaguers is an inexact science. But it’s fun. Be sure to post your results here.
by Jesse on Aug 24, 2009 1:15 PM EDT reply actions 0 recs
Looks like you're re-inventing MLB metrics for the minor leagues...
… but you have to do it the hard way because league averages and park factors are not readily available for the minor leagues (even at bb-reference).
I’m not entirely sure why bb-ref does not have this information yet. I mean, Rochester, New Britain, Fort Myers and Beloit play long seasons with balanced schedules no? Once PF’s and LgAvg’s are obtained its straightforward to get OPS+, ERA+ and many other good stuff which would help a lot.
Of course these metrics won’t provide a perfect ranking system because some “value” peripherals are better for projecting than others and there’s the age-relative-to-league issue but it would really help because its too much to keep track of the offense level of each league.
by DavidRF on Aug 25, 2009 2:01 PM EDT reply actions 0 recs
Above I said that I would need to add age/tools components to any rankings.
As to league averages and park factors, these are both readily available on the web. I know that bb-ref has minor-league averages of at least some major statistics, and that baseballthinkfactory.com has all the minor-league park factors (at least, through a few years ago).
Adding park factors would be the next step in these statistical rankings, along with creating a fluid point-reward system (allowing a player with an OBP of .500 to get more points than one with an OBP of .380, for example.)
Viking Vigil - The 2nd greatest Vikings community on the internet!
http://www.vikingvigil.com
Twins Fix
http://www.twinsfix.com
by Andersklasen on Aug 25, 2009 2:41 PM EDT up reply actions 0 recs
Cool...
If you just had OPS+ and ERA+ for minor leaguers, that would be a great start. I mean Parmelee’s .265/.359/.458 line is a bit hard to wrap my head around. Its not bad and I know that FSL is a pitchers league (.252/.322/.364) but how much exactly do I adjust his numbers? If someone tells me his OPS+ is 120 or 140 or 160, then I have a much better idea of how good he’s doing.
by DavidRF on Aug 25, 2009 8:29 PM EDT up reply actions 0 recs
For batters, start with...
….strike zone judgment.
Cases in point: Ruben Rivera as a minor leaguer compared to Johnny Damon.
Bloggin' the bloggers since 1938.
by Johnny Safron on Aug 25, 2009 9:20 PM EDT reply actions 0 recs
This
I totally agree with this. Stats that revolve around batting eye/pitch recognition are big. Then after that you’ve got contact and power.
"You can't sit on a lead and run a few plays into the line and just kill the clock. You've got to throw the ball over the damn plate and give the other man his chance. That's why baseball is the greatest game of them all."
~ Earl Weaver
"In God we trust. All others must provide evidence."
~ Billy Beane
by AdamOnFirst on Aug 26, 2009 12:52 AM EDT up reply actions 0 recs

by 















