I read a new blog a little bit today and was disheartened by something I read.
"Of course, common prospect wisdom tells us (if not to just ignore minor league, especially pitchers, numbers altogether) that numbers put up by pitchers too old for a league mean nothing really."
If "common prospect wisdom" says to ignore minor league stats, I need to change common wisdom, because that is false, in my opinion.
I want to take a vote to see how many people believe this. If you say no, they are not predictive, I would like to have an opportunity to try to change your mind.
I don't want names of players, years they played or any information as to who this person is. I want you to give me some data and I will tell you what kind of player they are likely to be. Essentially a Baseball Reference season is enough data to run this but three years, or even three levels worth is the best.
Give me three seasons of data from any level(essentially a BR stat line)::
age Level W L G GS CG SHO GF SV IP H HR R ER BB IBB SO
and I will give you these MLB stats for their career line:
SP/RP Impact Arm(yes/no) ERA k/9 bb/9 h/9 hr/9 debut Age MLB %
Give me this data(essentially a BR stat line):
level age AB R H 2B 3B HR RBI SB CS BB SO
I will give you this: (based on 500 AB)
AB R H Dbl Tpl HR RBI SB CS BB SO AVG OBP SLG
I can also give you peaks and minimums that I expect for hitters too, but I will only do that if the range is great.
Seriously guys, test me. I want you to trust the numbers. This isn't infallible but it does work.
If it's off, let me have it. I want people to realize that predicting MLB success from even low level minor league stats is possible.